Human Preimplantation Embryo Selection

HPE_Prelims.qxp 7/18/2007 2:59 PM Page i Human Preimplantation Embryo Selection HPE_Prelims.qxp 7/18/2007 2:59 ...

Author: Kay Elder | Jacques Cohen

236 downloads 1081 Views 8MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

HPE_Prelims.qxp

7/18/2007

2:59 PM

Page i

Human Preimplantation Embryo Selection

HPE_Prelims.qxp

7/18/2007

2:59 PM

Page ii

REPRODUCTIVE MEDICINE & ASSISTED REPRODUCTIVE TECHNIQUES SERIES Series Editors David K Gardner DPhil Colorado Center for Reproductive Medicine, Englewood, CO, USA Jan Gerris MD PhD Professor of Gynecology, University Hospital Ghent, Ghent, Belgium Zeev Shoham MD Director, Infertility Unit, Kaplan Hospital, Rehovot, Israel Published Titles Gerris, Delvigne and Olivennes, Ovarian Hyperstimulation Syndrome ISBN 978 1842143285 Sutcliffe, Health and Welfare of ART Children ISBN 9780415379304 Tan, Chian and Buckett, In-vitro Maturation of Human Oocytes ISBN 978 1842143322 Keck, Tempfer and Hugues, Conservation Infertility Management ISBN 978 0415384513 Pellicer and Simón, Stem Cells in Human Reproduction ISBN 978 0415397773 Forthcoming Titles Tucker and Liebermann, Vitrification in Assisted Reproduction ISBN 978 0415408820 Aplin, Fazleabas, Glasser, Giudice, The Endometrium, second edition ISBN 978 0415385831

HPE_Prelims.qxp

7/18/2007

2:59 PM

Page iii

Human Preimplantation Embryo Selection Edited by Kay Elder MD PhD Bourn Hall Clinic Cambridge UK

Jacques Cohen PhD Reprogenetics and Tyho–Galileo Research Laboratories Livingston, NJ USA

HPE_Prelims.qxp

7/18/2007

2:59 PM

Page iv

© 2007 Informa UK Ltd First published in the United Kingdom in 2007 by Informa Healthcare, Telephone House, 69–77 Paul Street, London EC2A 4LQ. Informa Healthcare is a trading division of Informa UK Ltd. Registered Office: 37/41 Mortimer Street, London W1T 3JH. Registered in England and Wales number 1072954. Tel: +44 (0)20 7017 5000 Fax: +44 (0)20 7017 6699 Website: www.informahealthcare.com All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of the publisher or in accordance with the provisions of the Copyright, Designs and Patents Act 1988 or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1P 0LP. Although every effort has been made to ensure that all owners of copyright material have been acknowledged in this publication, we would be glad to acknowledge in subsequent reprints or editions any omissions brought to our attention. Although every effort has been made to ensure that drug doses and other information are presented accurately in this publication, the ultimate responsibility rests with the prescribing physician. Neither the publishers nor the authors can be held responsible for errors or for any consequences arising from the use of information contained herein. For detailed prescribing information or instructions on the use of any product or procedure discussed herein, please consult the prescribing information or instructional material issued by the manufacturer. A CIP record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data Data available on application ISBN-10: 0 415 39973 4 ISBN-13: 978 0 415 39973 9 Distributed in North and South America by Taylor & Francis 6000 Broken Sound Parkway, NW, (Suite 300) Boca Raton, FL 33487, USA Within Continental USA Tel: 1 (800) 272 7737; Fax: 1 (800) 374 3401 Outside Continental USA Tel: (561) 994 0555; Fax: (561) 361 6018 Email: [email protected] Distributed in the rest of the world by Thomson Publishing Services Cheriton House North Way Andover, Hampshire SP10 5BE, UK Tel: +44 (0)1264 332424 Email: [email protected] Composition by Exeter Premedia Services Private Ltd, Chennai, India Printed and bound by Replika Press Pvt Ltd

HPE_Prelims.qxp

7/18/2007

2:59 PM

Page v

Contents List of contributors Preface Acknowledgments

vii xi xiii

Section I: Morphology 1. Human oocyte and embryo assessment for ART

1

A Henry Sathananthan and Sulochana Gunasheela 2. The zona pellucida and markers of oocyte and embryo viability

15

Anette Gabrielsen and Svend Lindenberg 3. Morphology and kinetics of human pronuclei

21

Jan Tesarik, Raquel Mendoza-Tesarik, Ermanno Greco, and Carmen Mendoza 4. Human pronuclei as a mode of predicting viability

31

Aidita N James 5. Multinucleation and mosaicism in the human preimplantation embryo

41

Renee Walmsley 6. The origins and consequences of fragmentation in mammalian eggs and embryos

51

Mina Alikani 7. Analysis of blastocyst morphology

79

David K Gardner, John Stevens, Courtney B Sheehan, and William B Schoolcraft 8. Morphometric analysis of human embryos

89

Christina Hnida and Søren Ziebe 9. Development rate, cumulative scoring, and embryonic viability

101

Christine C Skiadas and Catherine Racowsky 10. Human embryo cryopreservation and its effects on embryo morphology

123

James J Stachecki and Klaus Wiemer 11. Manipulating embryo development

135

Jacques Cohen Section II: Metabolism and Immunolgoy 12. Assessment of soluble human leukocyte antigen G in human embryos Jeffrey D Fisch, Levent Keskintepe, and Geoffrey Sher

145

HPE_Prelims.qxp

7/18/2007

2:59 PM

Page vi

CONTENTS

13. Immunological aspects of embryo development

155

Carol M Warner 14. Nitric oxide regulation of the preimplantation embryo

169

Yvette M Huet-Hudson 15. Uptake and release of metabolites in human preimplantation embryos

179

Fabienne Devreker 16. Preimplantation embryo metabolism and embryo interaction with the in vitro environment

191

Yves J R Ménézo and Pierre Guérin Section III: Genetic Aberrations and Embryo Selection 17. Polar body chromosome abnormalities and their consequences for human embryo development

201

Anver Kuliev and Yury Verlinsky 18. Chromosomal status of human embryos

209

Santiago Munné and Luca Gianaroli 19. Genomic imprinting and consequences for embryonic development

235

Henry E Malter 20. Selection of viable embryos and gametes by rapid, non-invasive metabolomic profiling of oxidative stress biomarkers

245

James T Posillico and The Metabolomics Study Group for Assisted Reproductive Technologies 21. Gene expression analysis in the human oocyte and embryo

263

Nury M Steuerwald 22. Mitochondria in reproduction: future assays for embryo selection

275

Brian Dale, Loredana Di Matteo, and Martin Wilding 23. Future genetic and other technologies for assessing embryos

287

Dagan Wells Section IV: Pre-Fertilization Parameters 24. Oocyte selection in contemporary clinical IVF: do follicular markers of oocyte competence exist?

301

Jonathan Van Blerkom and Susan W Trout 25. Sperm DNA and embryo development

325

Denny Sakkas and Emre Seli 26. The sperm centriole: its effect on the developing embryo

337

Calvin R Simerly and Christopher S Navara Index

vi

355

HPE_Prelims.qxp

7/18/2007

2:59 PM

Page vii

Contributors Mina Alikani PhD

Jeffrey D Fisch MD FACOG

Tyho-Galileo Research Laboratories

Sher Institute for Reproductive Medicine

Livingston, NJ

Las Vegas, NV

USA

USA

Jacques Cohen PhD

Anette Gabrielsen

Reprogenetics and Tyho–Galileo Research Laboratories

Ciconia Research and Development Aps

Livingston, NJ

Copenhagen

USA

Denmark

Brian Dale PhD DSc

David K Gardner DPhil

Director of Research

Scientific Director

Centre for Reproductive Biology

Colorado Center for Reproductive Medicine

Clinica Villa del Sole

Englewood, CO

Naples

USA

Italy Luca Gianaroli MD Fabienne Devreker Md PhD

Reproductive Medicine Unit

Clinic of Fertility

Italian Society for the Study of

Hospital Erasme

Reproductive Medicine

Brussels

Bologna

Belgium

Italy

Loredana Di Matteo BSc

Ermanno Greco MD

Facolta di Medicina e Chirurgia

Center of Assisted Reproduction

II Università degli Studi di Napoli

European Hospital

Naples

Rome

Italy

Italy

Kay Elder MD PhD

Pierre Guérin PhD

Boum Hall Clinic

Ecole Véténaire Lyon

Cambridge

Marcy l’étoile cedex

UK

France

HPE_Prelims.qxp

7/18/2007

2:59 PM

Page viii

LIST OF CONTRIBUTORS

Sulochana Gunasheela MD FRCOG

Henry E Malter PhD HCLD

Gunasheela IVF & Research Center

Scientific and Laboratory Director

Bangalore

Tower Fertility Center

India

Hackensack, NJ

Christina Hnida, Cand Scient PhD

USA

Laboratory Director

Yves J R Ménézo PhD DSc

The Fertility Clinic

TC Laboratoire de Procréation médicalement assistée

Herlev Hospital

Clinique du val d’Ouest

Copenhagen

Ecully cedex

Denmark

France

Yvette M Huet-Hudson PhD

Carmen Mendoza PhD

Professor of Biology

Department of Biochemistry and Molecular Biology

University of North Carolina at Charlotte,

University of Granada

Charlotte, NC

Granada

USA

Spain

Aidita N James PhD Associate Laboratory Director The A.R.T. Institute of Washington, Inc. Walter Reed Army Medical Center Washington, DC USA

Raquel Mendoza-Tesarik MAR & Gen Clinic Granada Spain Santiago Munné PhD

Svend Lindenberg

Reprogenetics


Livingston, NJ

Herlev University Hospital

USA

Copenhagen Denmark Levent Keskintepe PhD HCLD Executive Laboratory Director Sher Institute for Reproductive Medicine Las Vegas, NV USA

Christopher S Navara PhD Assistant Professor, Pittsburgh Development Center of Magee-Womens Research Institute University of Pittsburgh Medical School Pittsburgh, PA USA

Anver Kuliev MD PhD

James T Posillico

Reproductive Genetics Institute

Brigham and Women’s Hospital

Chicago, IL

Boston

USA

USA

viii

HPE_Prelims.qxp

7/18/2007

2:59 PM

Page ix


Catherine Racowsky PhD

Calvin R Simerly PhD

Department of Obstetrics and Gynecology

Pittsburgh Development Center of Magee-Womens

Brigham and Women’s Hospital

Research Institute

Boston, MA

University of Pittsburgh Medical School

USA

Pittsburgh, PA USA

Denny Sakkas PhD Associate Professor Yale University School Medicine New Haven, CT USA

Christine C Skiadas MD Brigham and Women’s Hospital Harvard Medical School Boston, MA USA

A Henry Sathananthan Monash Immunology and Stem Cell Laboratories Melbourne Australia

James J Stachecki PhD Tyho-Galileo Research Laboratories Livingston, NJ

William B Schoolcraft MD

USA

Medical Director Colorado Center for Reproductive Medicine

Nury M Steuerwald PhD

Englewood, CO

A.R.T. Institute of New York and New Jersey, NJ

USA

USA and

Emre Seli MD

University of North Carolina at Charlotte

Yale University School of Medicine

Charlotte, NC

New Haven, CT

USA

USA Courtney B Sheehan Colorado Center for Reproductive Medicine Englewood, CO USA

John Stevens Colorado Center for Reproductive Medicine Englewood, CO USA

Geoffrey Sher MD FACOG

Jan Tesarik

University of Nevada

MAR&Gen Clinic

School of Medicine

Granada

Reno, NV

Spain

USA

ix

HPE_Prelims.qxp

7/18/2007

2:59 PM

Page x


Susan W Trout

Dagan Wells PhD

Colorado Reproductive Endocrinology

Yale University Medical School

Rose Medical Center

New Haven, CT

Denver

USA

Colorado, CO USA

Klaus Wiemer PhD Northwest Center for Reproductive Sciences

Jonathan Van Blerkom PhD

Kirkland, WA

University of Colorado

USA

Boulder Colorado, CO

Martin Wilding PhD

USA

Centre for Reproductive Biology

Yury Verlinsky PhD

Clinica Villa del Sole

Reproductive Genetics Institute

Naples

Chicago, IL

Italy

USA Søren Ziebe Renee Walmsley

Laboratory Director

Institute for Reproductive Medicine and Science at


Saint Barnabas Medical Center

Rigshospitalet

Livingston, NJ

Universtiy Hospital of Copenhagen

USA

Copenhagen Denmark

Carol M Warner PhD Matthews Distinguished Professor of Biology Northeastern University Boston, MA USA

x

HPE_Prelims.qxp

7/18/2007

2:59 PM

Page xi

Preface Assisted reproductive technology (ART) is a numbers game, with permutations that involve the transfer of multiple embryos . . . but the most important number in IVF is of course the number one. One embryo, one sac, one fetus and one healthy baby – the ability to choose just one embryo that will lead to the successful birth of a baby is what we all crave in our profession. Seeking just this is the name of the game, the “Holy Grail of IVF”, as suggested by some of the authors in this book. The early pioneers of human IVF very quickly observed that not all gametes and embryos had the same potential to establish an ongoing pregnancy, and only a small proportion of oocytes that fertilized in vitro was truly viable. This was quickly followed by noting that, contrary to established experience in animal models such as the mouse, there is an obvious diversity in human embryo morphology and implantation potential. Although a correlation could be seen between outcome and morphological phenomena such as fragmentation, it was generally accepted that aesthetic appreciation – ‘embryonic looks’ – could be deceiving, and even the ‘ugliest’ embryo of a cohort can sometimes develop into a beautiful healthy baby. After 30 years of clinical IVF treatment, we have learned a great deal about human embryos – but there is still so much left to explore. The absence of absolute criteria that can predict the implantation potential of an embryo brings to mind the proverbial principle illustrated by the threesome of the Japanese Wise Monkeys – ‘to see no evil, hear no evil, and to speak no evil’. The practice of blindly compensating for lack of appropriate embryonic viability testing by transferring large groups of embryos is now all but gone; the debate surrounding embryo viability has changed instead to one of aptitude – the partial failure of new tests to predict implantation has become the norm. This notion has recently been transformed into a new and exciting science, and the search for the ultimate test has begun: the race is on to achieve the happy retirement of two words: ‘success rate’. This book was planned as a means of exploring this new and exciting science, and experienced authors who specialize in embryo testing were invited to contribute their expertise. Some of the authors have their background in basic science, other are dedicated to clinical IVF; they all share the common goal of finding this ‘holy grail’ with differing approaches and strategies. Our aim was to produce a book that is comparable to a peer-reviewed work, and the authors graciously allowed us to mingle with their text as editors, patiently providing explanations and further data if it was required. Although it is difficult to cover all aspects of gamete and embryo testing in one text, we tried to make it as comprehensive and up to date as possible. It is divided into four main sections, with chapters dealing with morphology determinations, immunology and metabolism, genetic aberrations, and pre-fertilization parameters. With respect to morphology assessment, there appears to be no real consensus on how to grade human embryos based on their morphology, and it is therefore relatively easy to criticize this most basic tool. It is generally accepted that there is a correlation between cell number and implantation, yet the absolute nature of this correlation is unknown; prospectively randomized trials have never been contemplated in order to determine the real value of morphological parameters or embryo development rate. We feel that use of microscopy is not over, and the morphology debate is becoming of increasing interest, with obvious but ethically challenging work yet to be undertaken. The second section on embryo metabolism offers an exciting glimpse into the feasibility of scoring embryos by examining spent culture media, using non-invasive tests. Although large randomized trials have not been carried out in this area of research, retrospective data shows promise, and more research is needed to expand the use of this tool for embryo assessment. The third section of the book explores ways of assessing the genetic status of embryos. Some conditions such as aneuploidy and mosaicism may be associated

HPE_Prelims.qxp

7/18/2007

2:59 PM

Page xii

PREFACE

with adverse conditions during follicular growth and gamete preparation, and also correlated with clinical outcome. Cell analysis using gene expression or imprinting are exciting approaches that may one day be available as clinical tools. Mutations in mitochondria, or changes in their patterns of activity provide another potential tool for single cell or whole embryo analyses. The fourth and final section covers examples of pre-fertilization parameters: aspects of sperm function, including DNA and centriolar integrity, and investigations of follicle-specific factors that influence oocyte competence. Kay Elder Jacques Cohen

xii

HPE_Prelims.qxp

7/18/2007

2:59 PM

Page xiii

Acknowledgments We are deeply indebted to all of our friends and colleagues who generously invested their time, experience and expertise in order to contribute to this book, and very much appreciate their tolerant patience in accepting and responding to our comments, questions and editorial corrections. We could also like to acknowledge and thank Nick Dunton, who was responsible for ‘conceiving’ the book, and for getting it into the first stages of development. We are grateful to Robert Peden, Lindsay Campbell and Helen Brock at Informa Healthcare for taking over this project during its completion.

HPE_Prelims.qxp

7/18/2007

2:59 PM

Page xiv

HPE_Chapter01.qxp

7/13/2007

4:35 PM

Page 1

1. Human oocyte and embryo assessment for ART A Henry Sathananthan and Sulochana Gunasheela

INTRODUCTION AND METHODS

The human oocyte, the female germ cell, is a unique cell equipped to fuse with and incorporate the sperm cell at fertilization and to sustain early embryonic development. It needs to be assessed for maturational status and normality for in vitro fertilization (IVF) and intracytoplasmic sperm injection (ICSI) in assisted reproductive technologies (ART). It is desirable to obtain a fresh, mature oocyte for insemination, usually after ovarian stimulation with gonadotropins or after down-regulation using gonadotropin-releasing hormone (GnRH) agonists/follicle stimulating hormone (FSH). With improved methods of ovarian stimulation and better timing of human chorionic gonadotropin (hCG), the majority of oocytes approach metaphase II (MII) and could be easily harvested for ART by ultrasonography. The trend now is to harvest a single oocyte in the natural cycle with minimal stimulation. The ripe MII oocyte is ovulated in a natural ovarian cycle around day 14. As much as we assess oocytes and sperm for ART, the embryo has to be assessed for embryo transfer in ART and currently for embryonic stem (ES) cell technology, a logical progression of ART. The fertilized ovum is the embryo, which undergoes cleavage by repeated mitoses to form a blastocyst during the first week of preimplantation embryogenesis (Figure 1.1). The embryonic genome is activated between the 4- and 8-cell stages in humans and the blastocyst implants in the uterus during the second week of development. The reader is referred to atlases of ART and other selected websites and references for images of gametes and embryos.1–9 All embryologists are advised to follow any embryology textbook to appreciate the highlights of development during the embryonic period (the first 8 weeks

of development), when most of the tissue and organ rudiments are laid down in the embryo. This chapter presents images supported by pointform assessments of the relevant stages of development. These include gross morphology, assessed in the laboratory using the inverted light microscope (LM), digital images of epoxy sections (LM), as well as fine structural assessments that may not be seen routinely, visualized by transmission electron microscopy (EMTEM). For surface observations in scanning electron microscopy (SEM), the reader is referred to atlases by Sathananthan3 and Makabe et al;10 Fluorescent microscopy (FM) is dealt with elsewhere in this book (see Chapter 26). The author’s website6 has images relevant to this chapter. OOCYTE ASSESSMENT

MATURATIONAL STATUS

Preovulatory oocytes, collected from multiple follicles after ovarian stimulation have commenced the final stages of meiotic maturation, ranging from germinal vesicle breakdown (GVBD) through metaphase I (MI), to MII.11–13 Nuclear maturation goes hand-in-hand with cytoplasmic and cortical maturation. Furthermore, changes also occur in the egg vestments, particularly the zona pellucida (ZP), increasing receptivity to sperm binding and penetration. Significantly, GVBD heralds the resumption of meiosis and initiates the expansion of the cumulus during maturation. This usually occurs in the culture medium prior to insemination (IVF) or sperm injection (ICSI) and may take 2–6 hours to complete, depending on the timing of oocyte pickup after administration of hCG. The process might be completed after insemination with washed sperm during IVF. Since the oocyte is 

HPE_Chapter01.qxp

7/13/2007

4:35 PM

Page 2

HUMAN PREIMPLANTATION EMBRYO SELECTION

A

B

C

D

E

F J

G

H

I

Figure 1.1 Normal whole embryos – 1-cell stage to blastocyst (LM). (A) Activated oocyte; (B) fertilized ovum (2PN); (C) 2-cell; (D) 4-cell; (E) 6-cell; (F) 8-cell; (G) compaction; (H) morula; (I) blastocyst; (J) hatching blastocyst. (Courtesy Dr. S. Gunasheela.24)

denuded of cumulus cells before ICSI, it is possible to precisely identify the mature oocyte, which has the first polar body (PB1) at the animal pole (AP). Whatever technique is used, the oocyte should not age in culture, becoming postmature, which could lead to abnormal fertilization and development, particularly aneuploidy and polyploidy. The mature oocyte is one of the largest cells (100–120 ␮m in diameter), surrounded by a gelatinous, glycoprotein shell, the ZP, and several layers of follicular cells, composing the cumulus oophorus. The female germ cell carries the 23 maternal chromosomes (n ⫽ 23) for procreation. The sperm cell contributes the 23 paternal chromosomes (n ⫽ 23) and the dominant centrosome (cell center) that initiates embryonic development after fertilization. Both sperm and egg contribute to the embryonic genome establishing diploidy (2n ⫽ 46), the essence of fertilization. FINE STRUCTURE OF THE MATURE EGG

To appreciate the processes of oocyte maturation, fertilization, and development we need to briefly 

review the structure of organelles in the oocyte.2,11,13 Basic cellular organelles found in most somatic cells are found in oocytes (Figure 1.2). These include the mitochondria, smooth endoplasmic reticulum (SER), lysosomes, annulate lamellae, few Golgi complexes, microtubules (MT), and microfilaments (MF). The SER consists of isolated vesicles or aggregates of tubular elements. Ribosomes are rare and rough endoplasmic reticulum (RER) is absent. Cortical granules (CG), unique to oocytes, are located beneath the oolemma (plasma membrane) and play an important role in fertilization. The human oocyte has no lipid or yolky inclusions, but survives in the oviduct and uterus during the first week of development. The metaphase II spindle, located at the AP, is barrel-shaped, anastral, and aligned perpendicular to the surface (Figure 1.3). It is composed of MT but lacks a functional maternal centrosome at each pole. The spermatozoon provides the dominant, centrosome (centriole) for embryo development in humans.14–16 The layer of follicle cells just outside the ZP is termed the corona radiata (CR). The CR is composed of typical somatic cells with the usual complement of cellular organelles. The oocyte has a

HPE_Chapter01.qxp

7/13/2007

4:35 PM

Page 3

HUMAN OOCYTE AND EMBRYO ASSESSMENT FOR ART

Figure 1.2 Human oocyte fine structure. The illustration incorporates cellular organelles of immature and mature oocytes, as well, and two follicle cells that play an important role in oocyte maturation. A ⫽ aggregate of SER; C ⫽ caveolus; CCP ⫽ CR process; CG ⫽ cortical granules; Ch ⫽ chromosomes; CR ⫽ corona radiata; En ⫽ endocytosis; Ex ⫽ exocytosis; G ⫽ Golgi complex; L ⫽ primary lysosome; M ⫽ mitochondria; MB ⫽ multivesicular body; MF ⫽ microfilaments; MT ⫽ microtubules; MV ⫽ microvilli; N ⫽ nucleus; PR ⫽ polyribosome; PVS ⫽ perivitelline space; RB ⫽ residual body; RER ⫽ rough endoplasmic reticulum; S ⫽ vesicular SER; Sp ⫽ meiotic spindle; Z ⫽ zona pellucida. Modified from Sathananthan et al. (1993).2

single polar body (PB1) in the perivitelline space (PVS) beneath the ZP, which carries the chromosomes extruded during meiosis. The fully mature oocyte (Figure 1.4) shows:11,13

The maturing, metaphase I oocyte has:2,11 ● ● ●

●

● ●

●

●

An expanded cumulus and radiating CR around the ZP (LM) A polar body (PB1) in the PVS at the AP (LM) A clear, homogenous ooplasm with even distribution of organelles (LM, EM) A barrel-shaped, anastral MII spindle beneath PB1 (LM, FM, EM) One to three layers of CG beneath the oolemma (LM, EM).

(The MII oocyte is ovulated around day 14 in the natural cycle.)

●

●

No polar body (LM) No germinal vesicle (LM) An expanding cumulus and corona cells (LM) A metaphase I spindle with homologous chromosomes (FM, EM) One or two layers of CG beneath oolemma (LM, EM).

(This stage is transient, there being no interphase.) The immature oocyte (Figure 1.4) at prophase I shows:2,11 ● ●

No polar body (LM) A GV or nucleus with a dense nucleolus (LM) 

HPE_Chapter01.qxp

7/13/2007

4:35 PM

Page 4


A compact, unexpanded cumulus and corona (LM) A discontinuous layer of CG beneath oolemma (LM, EM) An agranular cortex with Golgi membranes that secrete CG (LM, EM). (Oocytes about to mature will show an eccentrically located GV at one pole.) Oocytes during GVBD (Figure 1.4) show:2,11

●

●

●

● ● ● ●

Figure 1.3 Normal and aging oocytes – metaphase II spindles (TEM). The normal MII spindle is barrel-shaped, has no centrosomes at either pole and is attached to the egg cortex. The ageing spindle is displaced centripetally and has disorganized chromosomes at its MII plate. CG ⫽ cortical granules, p ⫽ polar body; S ⫽ smooth endoplasmic reticulum; Z, zona. ⫻27,300, ⫻3500. From Sathananthan (2002),6 (2007).22

●

A disappearing GV or nucleus (LM) Breakdown of the nuclear envelope (LM, EM) Condensation of chromosomes (FM, EM) Formation of a spindle with MT (FM, EM) Uncoupling of cell junctions between CR cells and oocyte (EM).

(This stage heralds the resumption of meiosis after its arrest at the GV stage.) Aging, postmature oocytes in culture (Figures 1.5–1.7) will show:2,11,13 ●

A dense ooplasm with vacuoles (swollen vesicular SER) (LM, EM)

PB1

Zona

GV

A

D

B

C

E

F

Figure 1.4 Preovulatory oocyte maturation (phase-contrast and LM). (A) and (D) germinal vesicle (GV) stage, (B) and (C) metaphase II, (E) GV breakdown, (F) telophase I, are depicted. Note retraction of cumulus cells in (C) and (E). ⫻400, ⫻1000. (A),(B) courtesy Dr. D. Payne, Adelaide, (C)–(F) From Sathananthan et al. 2003.2



HPE_Chapter01.qxp

7/13/2007

4:35 PM

Page 5


Figure 1.7 Aging oocyte – cortical granules (CG) and smooth endoplasmic reticulum (SER) (TEM). CG crowd beneath the surface with large aggregates of SER. Hypertrophy of SER is primarily induced by gonadotropin stimulation, during maturation. ⫻35 500. From Sathananthan 2002.6

Figure 1.5 Changes in aging oocyte ultrastructure. A, aggregate of smooth endoplasmic reticulum (SER) (hypertrophic); CG, cortical granules (crowded, displaced); Ch, chromosomes (scattered); G, Golgi; L, lysosome; Lb, lipofuschin body; M, mitochondria (dense); MV, microvilli (short); S, vesicular SER (swollen); Sp, MII spindle (displaced); Z, zona pellucida (hardened). From Sathananthan 1997.11

●

●

●

● ●

Normal or abnormal MII spindles, displaced from the surface (LM, FM, EM) Loss of spindle MT causing chromosome scatter (LM, FM, EM) Crowding of CG beneath oolemma or their centripetal migration (LM, EM) Few lipofuschin bodies with aging pigment (EM) Large hypertrophic aggregates of tubular SER (EM).

ASSESSMENT OF FERTILIZATION

A

B

Figure 1.6 Meiotic and mitotic spindles – chromosome scatter (TEM). The MII spindle (A) and that at syngamy (B) are disorganized. Some chromosomes have scattered outside the spindle zone, which can cause aneuploidy in embryos. A ⫻17 000, B ⫻10 000. From Sathananthan 2002.6

Fertilization begins with sperm–egg membrane fusion and culminates at syngamy, when the genetic constitution of the embryo is established. The oocyte is activated to become an embryo, the beginnings of life. The early events of fertilization cannot be visualized in the laboratory, except for the appearance of the second polar body (PB2), usually alongside PB1. These events, however, can be seen by TEM and FM, which are both invasive procedures.2,3,16 About 12 hours after insemination or ICSI it is easy to confirm fertilization in the laboratory, when two distinct pronuclei (PN), male and female, appear in the ooplasm. This stage is currently used to predict 

HPE_Chapter01.qxp

7/13/2007

4:35 PM

Page 6


normal development and prospective implantation. Each PN has about eight dense nucleoli that align adjacent to apposing pronuclear membranes in normal pronuclei. Apart from pronuclei, the alignment of nucleolar-associated chromatin is equally important in PN assessment, since this condenses to form the maternal and paternal, homologous chromosomes later at syngamy. This chromatin, however, is more difficult to see in the laboratory but is clearly visible in sections of ova, particularly by TEM (Figure 1.8). The dominant sperm centrosome has now been released from the sperm neck and is activated to form a sperm monoaster, that will eventually duplicate to establish a bipolar spindle (bipolarization) at the onset of mitosis. The zygote centrosome in the sperm aster has now duplicated centrioles and has become functional by attracting maternal ␥-tubulin, which nucleates MT, best visualized by FM. We believe that the restoration of the functionality of the dominant sperm centrosome is the most significant event of oocyte activation that initiates embryo development.15,17 A normally fertilized ovum 3 hours after insemination shows:2,3 ● ●

Abstriction of PB2 into the PVS (LM, EM) An incorporated, decondensing sperm head in the ooplasm (FM, EM)

●

●

●

A developing female pronucleus in ooplasm beneath PB2 (FM, EM) A sperm tail, midpiece, or sperm centriole in the ooplasm (FM, EM) Evidence of CG exocytosis all around the oocyte (EM).

(A fertilization cone may be evident at site of sperm incorporation.) A normally fertilized ovum 12–14 hours after insemination (Figure 1.8) has:2,3 Two pronuclei, male and female, associated in the central ooplasm (LM) Two polar bodies – PB1 with chromosomes and PB2 with a nucleus (LM) Nucleoli aligned close to apposing PN nuclear envelopes (LM, EM) No CG or few beneath oolemma after IVF (EM) Crowding of organelles, mostly mitochondria, around PN (LM, EM). (Delayed CG exocytosis has been observed after ICSI by TEM.) An abnormally fertilized ovum (dispermy) during IVF (Figure 1.9) has:2,3,9

●

●

●

● ●

● ●

Three pronuclei (two male and one female) (LM) Two polar bodies (PB1 and PB2) in PVS (LM)

Figure 1.8 Normal bipronuclear ova (LM and TEM). These bipronuclear ova, after monospermic fertilization, seem normal. Note alignment of nucleoli adjacent to apposing pronuclear membranes. What is more significant is the alignment of chromatin, associated with nucleoli, which would condense to form the male and female chromosomes at syngamy. ⫻35 700. From Sathananthan, 2003.6



HPE_Chapter01.qxp

7/13/2007

4:35 PM

Page 7


Chromosomes AL

Female pronucleus Nucleolus

M Interbody Microtubules

Nucleolus Male pronucleus

SPINDLE

Male pronucleus Mitochondria

Figure 1.9 Dispermic tripronuclear (3PN) ova (TEM). Classical images of 3PN ova at the pronuclear stage and syngamy. Note chromatin (dark specks) and nucleoli located toward adjacent membranes of the pronuclear envelopes. The spindle is tripolar enabling the ovum to divide into three cells, instead of two. ⫻5000, ⫻8000. From Sathananthan et al. 1995.9

●

●

●

A triploid chromosome complement 69XXY or 69XXY (LM) Two male centrosomes and two sperm asters (LM, FM, EM) A bipolar or tripolar spindle at syngamy (LM, FM, EM).

(May cleave into two or three cells, show normal early cleavage; and even develop to term.) An abnormal fertilized ovum (digyny) after ICSI shows:3 ●

● ●

●

Three pronuclei (two female and one male) caused by suppression of PB2 (LM) A single polar body – PB1 (LM) A triploid chromosome complement 69XXX or 69XXY (LM) One male centrosome and one sperm aster (FM, EM).

(Will not cleave normally beyond the 6–8-cell stage.) Structurally abnormal PN ova usually have:2 ● ● ●

●

●

‘Silent fertilization’ (Figure 1.10) may occur when:2,3,15 ●

●

●

(Can cleave normally and develop to term.) An unfertilized oocyte – parthenogenesis shows:2,11 ● ● ● ●

A single pronucleus – female (LM) A single polar body – PB1 (LM) Is usually haploid (n ⫽ 23) (LM) No crowding of organelles around PN (LM, EM).

PN with fuzzy irregular outlines (LM) PN of unequal size located peripherally (LM) PN not closely associated in the central ooplasm (LM, FM) Nucleoli not aligned against apposing PN envelopes (LM, EM) PN showing incomplete incorporation of chromatin with micronuclei (EM).

●

●

Sperm nuclear decondensation is arrested after IVF or ICSI (FM, EM) Sperm head remains unexpanded and does not form a male PN (EM) Sperm decondense chromatin but do not release its centrosome (FM, EM) The ovum has proceeded to syngamy after rapid PN formation (FM, EM) The ovum has arrested at syngamy – centrosomal dysfunction (FM, EM).

(The acrosome has to be discarded before spermhead decondensation during ICSI.) 

HPE_Chapter01.qxp

7/13/2007

4:35 PM

Page 8


Table 1.1 Normal embryonic growth from day 2–6. An embryo that develops to this timetable is likely to be more viable than the one which shows delayed growth

A

B

Figure 1.10 Failed sperm incorporation – IVF and ICSI (TEM). Occasionally sperm heads fail to expand and decondense their nuclear chromatin, after incorporation – ‘silent fertilization’. The sperm (right) has an abnormal nucleus. Note egg mitochondria around heads. ⫻17 000, ⫻35 700. From Sathananthan22 and Menezes.25

Day

Embryo

Appearance/hours

1

Fertilized ovum

2 3

Cleaving embryo Cleaving embryo Compacting embryo

4

Compacted morula

5

Early cavitating Early blastocyst Mid-blastocyst

2 PN (12 hours) and syngamy (18–24 hours) 2–6 cells: rounded blastomeres 8–10 cells or rounded blastomeres Blastomeres show evidence of adhesion Blastomeres show increased adhesion Beginning of blastocoel formation Blastocoel formed ICM, trophoblast and blastocyst clearly seen Trophoblast expanding, zona thinning out Embryo growing, blastocoel much increased Expanded ∼150–200 cells ; diameter ∼215 ␮m Trophoblast hatching out of zona Trophoblast and ICM hatched out of empty zona

Expanding blastocyst

6/7

EMBRYO ASSESSMENT

The first week of preimplantation development begins at fertilization and proceeds to blastocyst hatching (Figure 1.1). The first 2 or 3 days are critical in assessing normal development for embryo transfer (ET) in routine ART. The rate and timing of cleavage and development are important in assessing normality (Table 1.1). On day 1 the pronuclear ovum is assessed for normal or abnormal fertilization. The most important morphological parameters to assess in the laboratory are blastomere appearance, fragmentation, and multinucleation. The latter can be assessed non-invasively by using superior optical lenses, combined with digital microphotography recorded on video. Those with equal blastomeres, minimal cytoplasmic fragmentation, and few multinucleated cells have a better prospect of implantation. The fate of each embryo could be monitored right up to blastocyst hatching. Embryos are graded accordingly for ET (Table 1.2). Occasionally, embryonic blocks may occur at the 1-cell, 8-cell, or at any stage depending on culture conditions and embryo quality. It is advisable to let early embryos continue to develop at their own pace to overcome blocks. Totally arrested embryos should be discarded as they will eventually degenerate. Arrests could be caused by mitotic 

Late blastocyst Hatching blastocyst Hatched blastocyst

PN, pronuclei; ICM, inner cell mass. Modified from Gunasheela.24

Table 1.2 Embryo grading for embryo transfer in the laboratory. Grades 1 and 2 have a greater potential of establishing a clinical pregnancy Grade

Appearance

1

Blastomeres of equal size and no cytoplasmic fragmentation Blastomeres of equal size and minor cytoplasmic fragmentation (⬍10%) Blastomeres of unequal size and variable fragmentation Blastomeres of equal or unequal size and significant fragmentation (⬎10%) Few blastomeres of any size and severe fragmentation (50%)

2 3 4 5 Modified from Veeck.4

disturbances involving both chromosomal and centrosomal dysfunction. Aneuploidy, polyploidy, and mosaicism are the chief causes of early embryonic loss,18 apart from extensive fragmentation, which is now regarded as an apoptotic phenomenon. Several

HPE_Chapter01.qxp

7/13/2007

4:35 PM

Page 9


laboratories are now culturing embryos to blastocysts to select the most viable and vigorous for ET on day 6 or 7. Blastocyst culture is expensive, timeconsuming, and suitable for larger IVF centers.19–22 Normal cleavage stage embryos (Figures 1.1, 1.11–1.13) usually have:2,4,12

●

●

●

● ●

●

●

●

Rounded equal-sized blastomeres, except when cells are dividing (LM) Blastomeres with well defined outlines or cell membranes (LM) Cells with centralized, single nuclei, or metaphases (LM, EM) No fragments or show minimal fragmentation (⬍10%) (LM, EM).

(Embryos should develop according to the time frame in Table 1.1.) Common abnormalities (Figures 1.14–1.17) in early embryos include:1,2,4 ●

Extensive cytoplasmic fragmentation of blastomeres (30–50%) (LM)

F

a

●

● ●

●

Dark granular blastomeres and aggregation of organelles (LM) Extensive vacuolation of blastomeres – increases density (LM, EM) Cells with eccentrically located nuclei (LM, FM, EM) Multinucleated cells, many fragments and uneven cells (LM, FM, EM) Lack of compaction in later embryos and morulae (LM, EM).

a

b

c

d

e

f

Sy

d

4C

f

8C

Ag

●

2PN

2C

e

Arrested, degenerating embryos (Figures 1.16 and 1.17) show:1,2

b

PNA

c

Spontaneous fragmentation of whole blastomeres – apoptosis? (LM) Some unequal or fused blastomeres with eccentric nuclei (LM) Multinucleation of blastomeres – polyploidy (LM, FM, EM) Micronuclei in blastomeres beside normal nucleus – aneuploidy (LM, FM, EM).

16C

h

B

Figure 1.11 Diagrams of normal (A) and abnormal (B) embryos. Both normal and abnormal blastomeres are seen in abnormal embryos. Cytoplasmic fragmentation, multinucleation, and micronucleation are the main abnormalities. Fragments may be internal or external in the PVS, few to many. F, fertilization; C, cell; PN, pronuclei. From Sathananthan et al 1993.8



HPE_Chapter01.qxp

7/13/2007

4:35 PM

Page 10


Normal blastocysts on day 5/6 (Figure 1.18) have a:2,4,19–21 ●

●

●

●

●

Figure 1.12 Two-cell embryos – normal and fragmented. Both normal and abnormal embryos are evident. Fragments appear over the cleavage furrow or a whole cell can fragment totally. ⫻400. From Sathananthan et al.25

(Healthy blastocysts usually have 150–250 cells after DAPI, 4⬘6⬘-diaminolino-2-phenylindole, staining.) Hatching blastocysts on day 6/7 (Figures 1.1 and 1.19) show:2,4,19–21 ● ● ●

1 cell

2 cell

6 cell

8 cell

Distinct trophoblast, ICM and blastocoel (LM) Well-defined, compact ICM with many cells and cell junctions (LM, EM) Trophoblast forming a continuous, flat epithelium with cell junctions (LM) A large fluid-filled blastocoel, when expanded (LM) Few cleavage stage fragments in the blastocoel and PVS (LM, EM).

A fully expanded trophoblast and blastocoel (LM) A thinned-out zona (LM) Evidence of early hatching – trophoblast emerging at one pole (LM, EM)

3 cell

Morula

Figure 1.13 Normal human embryos – one cell to morula (phase-contrast). The cleavage embryos have equal blastomeres and minimal cytoplasmic fragmentation, except the 3-cell embryo. ⫻400. From Menezes (2005).25



HPE_Chapter01.qxp

7/13/2007

4:36 PM

Page 11


A

B

C Zona Zona

Mitochondria

Mitochondria

Nucleoli Fragment

Degenerated fragment

Nucleus

Vacuoles

D

E

Figure 1.14 Fragmented dispermic embryos (LM and TEM). Fragmentation is a common occurrence in early human embryos. Cytoplasmic fragments are devoid of nuclear material (D) compared with a normal blastomere (E). Four–8-cell and 5-cell (left) and a 10-cell embryo (right). ⫻400, ⫻3500, ⫻6000. From Sathananthan et al. 1999b.8,9

Golgi Mitochondria

Mitochondria Vacuoles

Nucleus NUCLEI

Nucleolus

Nucleolus Micronucleus Annulate lamellae Fragment

A

B Zona Sperm

Nucleolus

Nucleus Nucleolus

Golgi

Golgi Mitochondria

C

Mitochondria

Chromatin

Nucleus

D

Figure 1.15 Abnormal multinucleated dispermic embryos (TEM). (A) 1-cell (fragmented), (B) 2-cell (micronucleated), and (C) and (D) 3-cell embryos (multinucleated). ⫻6000, ⫻4000. From Sathananthan 2004.6



HPE_Chapter01.qxp

7/13/2007

4:36 PM

Page 12


●

●

Plump ‘zona-breaker’ (trophoblast) cells at hatching point (LM, EM) A layer of endoderm cells beneath ICM (LM, EM).

(Apoptotic cells are usually found in the ICM associated with phagocytic cells.) Abnormal blastocysts (Figures 1.18 and 1.20) may:4,20,21 ●

●

Figure 1.16 Three to 6-cell dispermic embryos (LM). Both normal and abnormal embryos are shown. Vacuolated blastomeres are degenerating. Note variation in cell size, few fragments, and multinucleated blastomeres. ⫻100. From Sathananthan et al. 1999b.9

● ●

●

Have no ICM or have a small or dispersed ICM (LM) Fail to expand and hatch on day 6 – are moribund or unable to break zona (LM) Arrest in development and often degenerate (LM) Have cleavage stage fragments in PVS – interferes with hatching (EM) Show many multinucleated cells in ICM, trophoblast, and endoderm (EM).

Figure 1.17 Eight to 10-cell dispermic embryos (LM). Both normal and abnormal embryos are evident. Blastomeres with clear vacoules are degenerating. Note unequal-sized blastomeres and few fragments. ⫻200. From Sathananthan et al. 1999b.9



HPE_Chapter01.qxp

7/13/2007

4:36 PM

Page 13


T ICM B A

B

E

D

C

Figure 1.18 Normal and abnormal blastocysts after ICSI (LM). (A) Normal blastocyst with trophoblast (T), inner cell mass (ICM), and blastocoel (B); (B) disorganized ICM; (C) disorganized endoderm (E); (D) Failed hatching – degenerating. ⫻400. From Sathananthan et al. 2003a,b.20,21

Figure 1.20 Fragments in a blastocyst (TEM). Fragments (F) are found between the trophoblast and zona and are discarded at hatching or remain in the blastocoel (not shown). Note dense mitochondria in fragment of early cleavage embryo. ⫻3400. From Sathananthan et al. 2003b.21

Blastocoel ZB

Blastocyst ZB

Zona Trophoblast vesicle

A

B

Hatching point

Trophoblast vesicle

Figure 1.19 Hatching blastocyst – zona breakers (LM). The blastocyst has hatched halfway. The inner cell mass is elsewhere. Zona breakers (ZB) at hatching point. A ⫻400, B⫻1000. From Sathananthan et al. 2003b.21



HPE_Chapter01.qxp

7/13/2007

4:36 PM

Page 14


Chromosome abnormalities in embryos (LM, FISH, PGD): Are higher than 50% irrespective of maternal age Aneuploidy increases with maternal age (⬎35 years) Aneuploidy is unrelated to embryo dysmorphism Polyploidy, mosaicism, chaoticism, and haploidy Most arrested embryos are abnormal Slow developing embryos also show more abnormalities Mosaicism is common in blastocysts – will not implant Most dispermic ova are mosaics compared to digynous ova.

● ●

● ● ● ●

●

●

Morphological abnormalities correlate well with chromosomal aberrations and decreased implantation potential.9,10,18,22,23 REFERENCES 1. Sathananthan AH, Trounson AO, Wood C. Atlas of Fine Structure of Human Sperm Penetration, Eggs and Embryos Cultured in Vitro. Philadelphia: Praeger Scientific, 1986: 279. 2. Sathananthan AH, Ng SC, Bongso A et al. Visual Atlas of Early Human Development for Assisted Technology. Singapore: Serono, 1993: 209. 3. Sathananthan AH, ed. Visual Atlas of Human Sperm Structure and Function for Assisted Reproductive Technology. Singapore: Serono, 1996: 279. 4. Veeck L. An Atlas of Human Gametes and Conceptuses. London: Parthenon, 1999: 215. 5. Gianaroli L, Plachot M, Magli MC. Atlas of embryology. Hum Reprod 2000; 15 (Suppl 4):1–79. 6. Sathananthan AH. Early Human Development. [CD-ROMs of human sperm, oocyte maturation, fertilization and embryos (previewed)]. 2002–2006. www.sathembryoart.com. 7. Gasser RF, ed. Digitally Reproduced Embryonic Morphology (DREM stage 1–4) [CD-ROM database of human embryos (Stages 1–4)]. 2005. www.virtualhumanembryo.isuhsc.edu.



8. Sathananthan AH, Ratnam SS, Trounson A, Edwards RG. Human preimplantation development – CD-ROM. Hum Reprod Update 1999; 5:89. 9. Sathananthan AH, Tarin JJ, Gianaroli L et al. Development of the human dispermic embryo (CD-ROM). Hum Reprod Update 1999; 5:553–60. 10. Makabe S, Van Blerkom J, Nottola SA, Naguro T. Atlas of Human Female Reproductive Function: Ovarian development to early embryogenesis in vitro fertilization. London: Taylor & Francis, 2006: 180. 11. Sathananthan AH. Ultrastructure of the human egg. Hum Cell 1997; 10:21–38. 12. Sathananthan AH. Ultrastructure of human gametes, fertilization and embryo development. In: Trounson AO, Gardner DK, eds. Handbook of In Vitro Fertilization, 2nd edn. Boca Raton, Florida: CRC Press, 2000: 431–64. 13. Sathananthan AH. Morphology and pathology of the human oocyte. In: Trounson AO, Gosden RG, eds. Biology and Pathology of the Human Oocyte. Cambridge University Press, 2003: 185–208. 14. Sathananthan AH, Kola I, Ng SC et al. Centrioles in the beginning of human development. Proc Natl Acad Sci U S A 1991; 88:4806–10. 15. Sathananthan AH, Ratnam SS, Ng SC et al. The sperm centriole: its inheritance, replication and perpetuation in early human embryos. Hum Reprod 1996; 11:345–56. 16. Schatten G. The centrosome and its mode of inheritance: the reduction of the centrosome during gametogenesis and its restoration during fertilization. Dev Biol 1994; 165:299–335. 17. Sathananthan AH. Mitosis in the human embryo: the vital role of the sperm centrosome (centriole). Histol Histopathol 1997; 12:827–56. 18. Munne S. Chromosome abnormalities and their relationship to morphology and development of human embryos. Reprod.Biomed.Online 2006; 12:234–53. 19. Bongso A. Handbook on Blastocyst Culture. Singapore: Sydney Press Indusprint, 1999: 93. 20. Sathananthan AH, Gunasheela S, Menezes J. Critical evaluation of human blastocysts for assisted reproduction techniques and embryonic stem cell biotechnology. Reprod Biomed Online 2003; 7:219–27. 21. Sathananthan AH, Menezes J, Gunasheela S. Mechanics of blastocyst hatching in vitro. Reprod Biomed Online 2003; 7:228–34. 22. Sathananthan AH. Embryology morphology in repeated ART failures. In: Arora M, Konje JC, eds. Recurrent Pregnancy Loss. New Delhi: Jaypee, 2007; 274. 23. Sathananthan AH. Abnormal nuclear configurations encountered in human IVF: Possible genetic implications. Assisted Reprod Technol Androl. 1990; 1:115–33. 24. Gunasheela S. The A–Z Encyclopedia on Male and Female Infertility. New Delhi: Jaypee, 2005: 146. 25. Menezes J. Video of whole human embryos in vitro. 2005.

HPE_Chapter02.qxp

7/13/2007

5:24 PM

Page 15

2. The zona pellucida and markers of oocyte and embryo viability Anette Gabrielsen and Svend Lindenberg

INTRODUCTION

Mammalian immature and mature oocytes, and early stage embryos are surrounded by a multilayer glycoprotein coat, the zona pellucida.1–3 The zona contains two to four highly conserved zona proteins (ZP1, ZP2, ZP3, ZP4) that vary between species. Three genes that encode these proteins have been identified.4 Homologous genes in different species share at least 25% of their sequences. Fish extracellular coat proteins are synthesized in the liver, whereas in mammalian eggs this takes place in the ovary, requiring certain changes in the C-terminal region of the polypeptides.5 Targeted mutagenesis of endogenous mouse genes and transgenesis with human homologs has provided models to investigate the roles of individual zona structures. This work in progress was recently reviewed by Dean.6 The proteins appear to be actively secreted by oocytes during the process of follicle development. In the human ovary, the developing granulosa cell layer that supports the ova also contributes to the secretion of stage-dependent zona proteins during folliculogenesis.7,8 Ultrastructural studies of the developing human zona pellucida show that it consists of a series of highly ordered filaments.9,10 At the molecular level, the zona has a paracrystalline, three-dimensional structure composed of heterodimeric filaments of ZP2 and ZP3 proteins, crosslinked to ZP1 proteins.11 Other highly glycosylated proteins of unknown function are distributed within this orderly arranged structure of filaments.12 In terms of function, the mammalian zona pellucida forms a physical barrier between the oocyte and the follicular cells during oogenesis and folliculogenesis. However, during these processes it is essential that the zona can be penetrated by granulosa cell

projections that form junctions with the oocyte, thereby incorporating the oolemma into the syncytial meshes of surrounding granulosa cells. A direct communication between the oocyte and the somatic compartment thus exists during the follicular development of granulosa cells and the oocyte.13 These communications facilitate and ensure the arrest of meiosis.14 The granulosa cell protrusions are directly engaged in the architecture of the zona, and disturbance in ZP1 specifically disrupts granulosa cell– oocyte interaction; this has been shown to jeopardize the early stages of oocyte maturation.15 Thus, any disturbance in the structure of the zona during the early stages of oogenesis may result in an asynchrony in nuclear cytoplasmic maturation, and hence impair further embryonic development. The zona also plays an obvious role in vivo during fertilization and implantation16 in preventing polyspermy,17 a phenomenon first demonstrated in the human zona pellucida by Soupart and Strong18 and recently re-evaluated for mammalian eggs in an excellent review by Gardner and Evans.19 The zona pellucida also protects the embryos from mechanical stress up to the time of implantation.20 The zona pellucida is also involved in preventing an immune reaction to any foreign structure in the female reproductive tract, expressing genes both maternally and paternally – a phenomenon first recognized by Willadsen in 1979.21 Indeed, immune suppression therapy has been advocated during the application of certain zona pellucida opening procedures in human IVF.22 During in vivo development of the human embryo, embryo-derived enzymes as well as endometrial and fallopian tube secretions possibly modulate the zona, changing its morphology and structure. For obvious reasons, only the embryonic counterpart of this modulation may take place in vitro, and further 

HPE_Chapter02.qxp

7/13/2007

5:24 PM

Page 16


to this a process of zona hardening in vitro has been described, which might impede the blastocyst’s attempt to hatch. Failed implantation may be due at least in part to entrapment of the embryo as a consequence of impaired zona-hatching.23 Indeed, implementation of the assisted hatching (AH) technique was demonstrated to be effective in patients where chemical or physical changes in the zona pellucida had occurred as a consequence of advanced reproductive age or elevated basal follicle stimulating hormone (FSH) concentrations.24,25 From these considerations it might be speculated that the structure and function of the zona is associated with specific changes within the oocyte, and this may therefore also affect the oocyte’s potential for successful embryo development. The morphology of the zona pellucida can also be influenced by extrinsic factors. The periovulatory hormonal environment may have an impact on the thickness of the zona pellucida. Studies by Bertrand et al26 demonstrated that the thickness of the zona pellucida is also influenced by hormonal stimulation in assisted reproductive technology (ART) cycles. Therefore, ovulation induction might be an important modulator of zona function in the process of fertilization in vitro.26 There is also increasing evidence that suggests a further influence of ovulation induction on the hatching procedure. This will be discussed below.

MORPHOLOGICAL CHANGES IN THE ZONA AS PREDICTORS OF ART OUTCOME

Criteria that can be assessed by routine microscopy prior to fertilization or embryo transfer would be useful in selecting embryos for transfer during ART procedures; however, very few reliable, predictive and non-invasive markers for oocyte quality have thus far been identified. Nearly two decades ago it was described that the zona pellucida of a proportion of cleaving embryos had thinned areas in their zonae pellucidae.23 This could be expressed as a percentage variation of zona pellucida thickness and was considered an important factor associated with implantation of embryos. When the ‘best’ embryo had a zona



pellucida that varied more than 25%, 24 of 60 (40%) resulted in pregnancy; pregnancies were not induced (0/21) when the ‘best’ embryo had less than 10% variation. The thinned areas appeared to be part of a dynamic process that was already noticeable in zygotes, but can increase during preimplantation development, particularly in embryos with a good prognosis.27 Early studies showed that zona thinning of the expanding blastocyst is not a unique process, as it is preceded, at least in the human, by a gradual thinning process correlated with increased viability starting as early as the zygote and cleavage stages. Several other studies have since reported on the relationship between the thickness and morphology of the human zona pellucida and embryo quality, embryo development and pregnancy rates. Three major approaches have been used to assess morphological criteria in the human zona, including direct imaging with camera or video using an inverted microscope with Hoffman modulation optics followed by processing the film for manual measurements.23 This method has advantages in terms of reliability but is slow and not helpful in real-time prediction of viability. A second approach offered by other investigators27 uses a digitized imaging system to store images, which can be measured with the aid of a computer. Many different criteria have been measured, but the two that have been most commonly used are zona pellucida thickness (ZPT) and zona pellucida thickness variation (ZPTV). In the majority of studies, the zona was measured at three points as suggested by Cohen and co-workers,23 with the following calculations applied. Mean ZPT value and ZPTV value computed as: ZPTmean ⫽ (ZPT1 ⫹ ZPT2 ⫹ ZPT3 )/3 ZPTV ⫽ (ZPTmax ⫺ ZPTmean )/ZPTmean ⫻ 100%.

An example of this type of measurement is shown in Figure 2.1, with multiple data sets used to increase the accuracy of the measurements.28–30 Zona pellucida thickness variation is associated not only with better outcome in terms of pregnancy,

HPE_Chapter02.qxp

7/13/2007

5:24 PM

Page 17

THE ZONA PELLUCIDA AND MARKERS OF OOCYTE AND EMBRYO VIABILITY

Figure 2.1 The outer and inner zona is delineated and the zona pellucida thickness has been calculated.

but also with better overall embryo morphology.28 One might therefore speculate as to which variable might be the better predictor. We investigated this question in a smaller series, and found no significant impact of the ZPTV in embryos that already had optimal morphology measured as a function of development rate and fragmentation, among other parameters. The ZPTV did predict viability of embryos when other morphology criteria were inferior. In reports where pregnancy rate was the measured outcome,27–32 ZPT was not associated with an improved pregnancy rate for all groups of patients, but was found to be generally correlated with pregnancy in all of the studies, which included biometric analysis of the zona pellucida at the time of embryo

transfer. The data suggest that assessment of ZPTV should be included in criteria used to select embryos of optimal quality prior to transfer. A third exciting and real-time technique for evaluating zona pellucida morphology can be achieved with the use of a Polscope.33 With this system, the retardance magnitude and the thickness of the inner, middle and outer layers of the ZP are measured before embryo transfer. The authors found that the magnitude of light retardance by the inner layer of the zona pellucida appears to represent a unique non-invasive marker for the developmental potential of the oocyte. We were able to confirm this observation in a small series of 20 oocytes prepared for ICSI (Lindenberg, unpublished data).



HPE_Chapter02.qxp

7/13/2007

5:24 PM

Page 18


CONCLUSION

The human zona pellucida is not static, but is a highly dynamic structure that serves several known and unknown functions during oogenesis and embryonic development. Specifically, the zona seems to have a vital impact on co-ordinating the influence of the somatic cell compartment in the ovary on oocyte maturation. Fertilization and remodelling of the zona after cortical granule release are also classical features of the zona pellucida’s role. With the use of conventional microscopy to measure ZPTV, it has been found that remodelling of the zona after fertilization seems to influence both the immediate in vitro, and later in vivo development of the human embryo. ZPTV can be seen as early as the zygote stage and increases in a subset of embryos over time. It is our firm conviction that assessment of ZPTV should be an integral part of the selection criteria for embryo transfer. We therefore conclude that ZPTV measurement is a well documented but clearly underappreciated tool that can be used to select embryos for transfer on days 2–3 after fertilization. Furthermore, this criterion appears to be even more important when only embryos of suboptimal morphology are available for transfer. The ZPTV could also be used to select those embryos that might benefit from assisted hatching.26 These observations are a valuable addition to criteria that may be of benefit in a policy of selected single embryo transfer. REFERENCES 1. Baranska W, Konwinski M, Kujawa M. Fine structure of the zona pellucida of unfertilized egg cell and embryos. J Exp Zool 1975; 192: 193–202. 2. Herrler A, Beier HM. Early embryonic coats: morphology, function, practical applications. Cell Tissue Organs 2000; 166(2): 233–46. 3. Carino C, Prasad S, Skinner S et al. Localization of species conserved zona pellucida antigens in mammalian ovaries. Reprod Biomed Online 2002; 4: 116–26. 4. Spargo SC, Hope RM. Evolution and nomenclature of the zona pellucida gene family. Biol Reprod 2003; 68: 358–62. 5. Litscher ES, Wassarman PM. Egg extracellular coat proteins: from fish to mammals. Histol Histopathol 2007; 22: 337–47. 6. Dean J. Reassessing the molecular biology of sperm-egg recognition with mouse genetics. Bioessays 2004; 26: 29–38. 7. Sinowatz F, Topfer-Petersen E, Kolle S, Palma G. Functional morphology of the zona pellucida. Anat Histol Embryol 2001; 30: 257–63.



8. Gook D, Martic M, Borg J, Edgar DJ. Identification of zona pellucida proteins during folliculogenesis. Hum Reprod 2004; 19 (Suppl 1): 140. 9. Familiari G, Nottola SA, Macchiarelli G, Micara G, Aragona C, Motta PM. Human zona pellucida during in vitro fertilization; an ultrastructural study. Mol Reprod Dev 1992; 32: 51–61. 10. Oehninger S. Biomedical and functional characterization of the human zona pellucida. Reprod Biomed Online 2003; 7: 641–8. 11. Wasserman PM. Zona pellucida glycoproteins. Ann Rev Biochem 1988; 57: 415–42. 12. Bogner K, Hinsch KD, Nayudu P et al. Localization and synthesis of zona pellucida proteins in the mammoset mokey ovary. Mol Hum Reprod 2004; 10: 481–8. 13. Albertini DF, Riedler V. Patterns of intercellular connectivity in the mammalian cumulus–oocyte complex. Microsc Res Tech J 1994; 27: 125–33. 14. Eppig JJ. Intercommunication between mammalian oocyte and companion somatic cells. BioEssays 1991; 13: 569–74. 15. Rankin T, Talbot P, Lee E, Dean J. Abnormal zona pellucida in mice lacking ZP1 result in early embryonic loss. Development 1999; 126: 3847–55. 16. Rankin T, Soyal S, Dean J. The mouse zona pellucida. Mol Cell Endocrinol 2001; 163: 21–5. 17. Hoodbhoy T, Dean J. Insights into the molecular basis of sperm-egg recognition in mammals. Reproduction 2004; 127: 417–22. 18. Soupart P, Strong PA. Ultrastructural observations on polyspermic penetration of zona pellucida-free human oocytes inseminated in vitro. Fertil Steril 1975; 26: 523–37. 19. Gardner AJ, Evans JP. Mammalian membrane block to polyspermy: new insights into how mammalian eggs prevent fertilisation by multiple sperm. Reprod Fertil Dev 2006; 18: 53–61. 20. Nichols J, Gardner RL. Effect of damage to the zona pellucida on development of preimplantation embryos in the mouse. Hum Reprod 1989; 4: 180–7. 21. Willadsen SM. A method for culture of micromanipulated sheep embryos and its use to produce monozygotic twins. Nature 1979; 277(5694): 298–300. 22. Cohen J, Malter H, Elsner C et al. Immuno-suppression supports implantation of zona pellucida dissected human embryos. Fertil Steril 1990; 53: 662–5. 23. Cohen J, Inge KL, Suzman M, Wiker SR, Wright G.Videocinematography of fresh and cryopreserved embryos: a retrospective analysis of embryonic morphology and implantation. Fertil Steril 1989; 51: 820–7. 24. Cohen J, Alikani M, Reing AM et al. Selective assisted hatching of human embryos. Ann Acad Med Singapore 1992; 21: 565–70. 25. Cohen J. Assisted hatching of human embryos. J Assist Reprod Genet 1993; 17: 179–90. 26. Bertrand E, Van den Bergh M, Englert Y. Clinical parameters influencing human zona pellucida thickness. Fertil Steril 1996; 66(3): 408–11. 27. Wright G, Wiker S, Elsner C et al. Observations on the morphology of pronuclei and nucleoli in human zygotes and implications for cryopreservation. Hum Reprod 1990; 5: 109–15. 28. Gabrielsen A, Bhatnager PR, Petersen K, Lindenberg S. Influence of zona pellucida thickness of human embryos on clinical pregnancy outcome following in vitro fertilization treatment. J Assist Reprod Genet 2000; 17: 323–8. 29. Host E, Gabrielsen A, Lindenberg S, Smidt-Jensen S. Apoptosis in human cumulus cells in relation to zona pellucida thickness variation, maturation stage, and cleavage of the corresponding oocyte after intracytoplasmic sperm injection. Fertil Steril 2002 Mar; 77(3): 511–15.

HPE_Chapter02.qxp

7/13/2007

5:24 PM

Page 19

THE ZONA PELLUCIDA AND MARKERS OF OOCYTE AND EMBRYO VIABILITY

30. Gabrielsen A, Lindenberg S, Petersen K. The impact of the zona pellucida thickness variation of human embryos on pregnancy outcome in relation to suboptimal embryo development. A prospective randomized controlled study. Hum Reprod 2001; 16(10): 2166–70. 31. Garside WT, Loret D, Mola JR et al. Sequential analysis of zona thickness during in vitro culture of human zygotes: correlation with embryo quality, age, and implantation. Mol Reprod Dev 1997; 47: 99–104.

32. Palmstierna M, Murkes D, Csemiczky G, Andersson O, Wramsby H. Zona pellucida thickness variation and occurrence of visible mononucleated blastomeres in preembryos are associated with a high pregnancy rate in IVF treatment.J Assist Reprod Genet 1998; 15: 70–5. 33. Shen Y, Stalf T, Mehnert C, Eichenlaub-Ritter U, Tinneberg HR. High magnitude of light retardation by the zona pellucida is associated with conception cycles. Hum Reprod 2005; 20: 1596–606.



HPE_Chapter02.qxp

7/13/2007

5:24 PM

Page 20

HPE_Chapter03.qxp

7/13/2007

5:22 PM

Page 21

3. Morphology and kinetics of human pronuclei Jan Tesarik, Raquel Mendoza-Tesarik, Ermanno Greco, and Carmen Mendoza

INTRODUCTION

The process of fertilization is a complex sequence of events the signs of which are perceptible as early as several seconds after sperm–oocyte fusion.1 However, the demonstration of very early signs of fertilization requires recourse to special techniques, most of which are destructive for the fertilized oocyte. Hence, since the first successful clinical application of human in vitro fertilization (IVF),2 two signs of fertilization that are detectable by non-invasive inspection of inseminated oocytes have been used: the extrusion of the second polar body and the development of pronuclei. The extrusion of the second polar body is a much earlier sign of fertilization than is the development of pronuclei. However, the high incidence of first polar body division or fragmentation compromises the accuracy of second polar body detection in many cases. Consequently, the development of pronuclei after IVF has become the gold standard of fertilization assessment. The possibility of distinguishing between normal fertilization (2 pronuclei), parthenogenetic oocyte activation (1 pronucleus), and polyspermic oocyte penetration (⬎2 pronuclei) is an additional advantage of pronucleus-based evaluation systems, because parthenogenetically activated and polyspermically penetrated oocytes both show the same pattern of second polar body extrusion as does a normally fertilized oocyte. There are a few exceptions to these common observations: firstly, eggs can be activated by a sperm cell without decondensation. This is apparently quite common, and may be overlooked as abnormal fertilization, activation or failed fertilization.3 Secondly, a single pronucleus can be the result of the close association of male and female genomes in human zygotes.4 Thirdly, some eggs may activate without extruding the second polar body, rendering the zygote digynic.

The idea that analysis of pronuclei can provide something more than simply evidence of fertilization arose from studies performed in the late 1980s which showed that pronuclear development reflects the activity of developmentally important oocyte cytoplasmic factors,5 and is also related to an early period of RNA synthesis in the zygote.6,7 At the same time, the relationship between characteristics of pronuclear development, nucleolar distribution/ movement and embryo viability/implantation potential, was established by Wright et al;8 this was confirmed and expanded upon by our team and others some years later.9–11 Here we outline the biological basis that underlies pronuclear development, as well as the pathological conditions that lead to abnormal pronuclear development. This information should help to understand the relationship between pronuclear morphology and IVF outcomes. It can also serve as a background for further studies aimed at refining the existing pronuclear scoring systems and defining new clinical applications and relationships.

THE PHYSIOLOGY OF THE HUMAN ZYGOTE

The beginning of the zygote’s existence is marked by sperm–oocyte fusion. At this point in time, the fertilizing spermatozoon triggers a cascade of cell signalling events in the oocyte, collectively termed oocyte activation. A series of repetitive increases in free intracellular calcium concentration (calcium oscillations) has become the most well studied aspect of human oocyte activation.12 Relatively less is known about downstream elements of the oocyte-activating signal transduction cascade, which can be expected to affect regulatory elements that control the oocyte’s cell cycle checkpoints and the function of cytoskeletal elements; these promote the exit of the oocyte 

HPE_Chapter03.qxp

7/13/2007

5:22 PM

Page 22


from metaphase II (MII) arrest and its entry into the regular mitotic cell cycle.13,14 Studies performed in different mammalian species have shown that early embryonic development occurs in the absence of active gene expression during a period which is species dependent. In the human, a major activation of nuclear RNA synthesis occurs at the 4-cell stage,15,16 and the first biochemical17 and morphological18 signs of embryonic gene expression are detectable between the 4- and the 8cell stage. The messages required to guide developmental processes that take place prior to embryonic gene activation are derived from a pool of mRNAs that have been synthesized during oocyte growth and are stored in the oocyte cytoplasm as polyadenylated mRNA. It is tempting to speculate that the drawn-out and spatially differentiated series of periodic calciumdriven signalling events observed throughout human zygote development19 may determine the timing of deadenylation and expression of specific maternal mRNAs stored in different regions of the oocyte and zygote. In fact, spontaneous12 and drug-induced20 disturbances of calcium oscillations in the human zygote were shown to be associated with abnormalities of pronuclear morphology and developmental arrest. From the developmental point of view, the main function of the zygote is to ensure a timely start for the embryonic cell cycle after the long period of arrest that both the sperm- and the oocyte-derived genomes have experienced in their respective gametes. This function includes the correct timing of DNA synthesis during S-phase and the subsequent fusion of paternal and maternal genomes to merge into a unique genome for the future embryo.

THE DEVELOPMENT AND FUNCTION OF PRONUCLEI EXPERIMENTAL APPROACHES

Most of the basic knowledge about the structure and function of human pronuclei is derived from invasive studies in which either normal zygotes were sacrificed, or polypronuclear eggs were formed for 

research purposes by inseminating zona-free human oocytes. These studies used electron microscopy and autoradiography to analyze changes in pronuclear ultrastructure and nucleic acid synthesis.5–7,21 More recently, a number of non-invasive studies were carried out, based on observations of pronuclei in living human zygotes. These studies, nicely reviewed by Scott,22 could not use the same structural detail and experimental rigor as the earlier invasive studies. However, they have the merit of having established the existence of relationships between the appearance of pronuclei and further embryonic development, including data about implantation and post-implantation events. Both types of studies are thus complementary. THE ESTABLISHMENT OF PRONUCLEI

Soon after sperm–oocyte fusion, the nuclear envelope of the fertilizing spermatozoon disintegrates, and both sperm- and oocyte-derived chromatin is thus directly accessible to oocyte cytoplasmic factors.1 However, sperm chromatin differs from the oocyte telophase chromosomes by its higher degree of condensation, due to its association with spermspecific nuclear protamines. Sperm nuclear protamines are removed and replaced with oocyte-derived histones before the male pronucleus is formed.1 This step is obviously dependent on the availability of histones in the oocyte. Relative insufficiency of oocyte histones may thus slow down the transformation of the sperm nucleus into the male pronucleus, and consequently cause asynchrony of pronuclear development. The establishment of pronuclei is not completed until a new nuclear envelope is formed around the sperm- and oocyte-derived chromatin. This process has been studied in human polyspermically penetrated zona-free oocytes,21 and was shown to involve alignment and stepwise fusion of vesicles and tubules of endoplasmic reticulum around the chromatin (Figure 3.1). Once nuclear envelope formation is completed, the pronuclei can be easily visualized in living zygotes with the use of phase contrast, Nomarski differential interference contrast or Hoffman modulation contrast optics.

HPE_Chapter03.qxp

7/13/2007

5:22 PM

Page 23

MORPHOLOGY AND KINETICS OF HUMAN PRONUCLEI

Figure 3.1 Electron micrograph of a fully decondensed sperm nucleus in the course of nuclear envelope assembly, occurring in an early phase of male pronucleus development, 4 hours after in vitro insemination. Chromatin (ch) is surrounded by a discontinuous series of vesicles (v) and cisternae (c) of endoplasmic reticulum. Arrows indicate intranuclear dense bodies. ⫻20 000. Reprinted from Tesarik et al.18

DEVELOPMENTAL CHANGES IN PRONUCLEAR STRUCTURES

Newly formed pronuclei contain finely dispersed chromatin with disseminated small electron-dense bodies.21 The latter progressively aggregate and fuse with each other, and are closely associated with large chromatin clusters (Figure 3.2). This process leads to the formation of still larger aggregates, consisting of material derived from the electron dense intranuclear bodies and chromatin, merged in compact structures that develop into functionally active nucleoli three cell cycles later.16,23 Thus, these structures were given the name of nucleolar precursor bodies (NPBs).23 In the experimental system based on polyspermically penetrated human zona-free oocytes, the process of NPB formation began as early as 4 hours after in vitro insemination, and the first fully developed NPBs appeared 12 hours after in vitro insemination.21 This transformation coincides with the development of typical nuclear pores all around the pronuclei, and with the appearance of characteristic

Figure 3.2 Electron micrograph showing a part of a developing male pronucleus with nucleolar precursors (np) at different phases of their assembly. ⫻44 000. Reprinted from Tesarik et al.18

vesicles between the two membranes of the nuclear envelope.21 The function of these vesicles, also observed in blastomeres of cleaving human embryos,24 is unknown. DEVELOPMENTAL CHANGES IN PRONUCLEAR FUNCTION

One of the main functions of pronuclei is to ensure adequate conditions for the first phase of DNA synthesis after fertilization so that the sperm- and oocyte-derived chromatin is eventually mixed. An experiment in which 3H-thymidine was incorporated into the pronuclei of polyspermically penetrated human zona-free oocytes, evaluation and analysis by autoradiography, showed the first signs of DNA synthesis no earlier than 12 hours after in vitro insemination, and only in those pronuclei that had completed ultrastructural differentiation of NPBs.6 Interestingly, human paternal pronuclei can only complete their NPB differentiation when they have previously carried out a limited RNA synthesis, detected in polyspermically penetrated human zona-free oocytes 

HPE_Chapter03.qxp

7/13/2007

5:22 PM

Page 24


as early as 4 hours after in vitro insemination.5 Consequently, delayed NPB development is likely to signal a failure or delay of the early pronuclear RNA synthesis, a condition which will in turn cause a delay in the onset of DNA synthesis. The nature of the transcripts resulting from pronuclear RNA synthesis is largely unknown. However, it has been shown that paternal Y-linked transcripts are expressed in paternal pronuclei of human zygotes.25 Early RNA synthetic activity has also been observed in mouse pronuclei, similar to that observed in humans.26 Interestingly, chromatin-mediated repression of promoter activity that is almost total is imposed on maternal pronuclei, but not on paternal pronuclei.27,28 Consequently, the ongoing rate of transcription by endogenous genes in the mouse zygote is four to five times greater in the paternal pronucleus than in the maternal pronucleus.26 The existence of similar differences between transcriptional activity in male and female pronuclei in the human zygote still remains to be determined, although the male pronucleus in general contains more nucleoli than the female pronucleus.29 PRONUCLEAR MOVEMENT

The male and female pronuclei are initially separate spatially, and they then move in the zygote cytoplasm to approach each other. This process culminates in a close apposition of both pronuclei.1 It appears that apposition of the two pronuclei is achieved near the spindle if a spermatozoon has penetrated close by, and in the center of the zygote if the spermatozoon has entered the oocyte through a region opposite the spindle.30,31 It was hypothesized that the movement of the pronuclei is integrated within an overall cytoplasmic movement (rotation) in the oocyte and zygote.32,33 Direct evidence for zygote cytoplasmic rotation comes from time-lapse video recordings of human sperminjected oocytes showing that the sperm centrosome can organize contraction waves, resulting in clockwise rotations of cortical granulated cytoplasm. This was noted in 95% of mature human MII oocytes, beginning 2–3 hours before second polar body extrusion following ICSI, and ending as the 

second polar body was extruded.34 Indirect evidence for oocyte cytoplasmic rotation near the time of fertilization comes from the observation that second polar body extrusion often occurs distant from the first polar body after fertilization by ICSI.33 The correct function of the sperm-derived centrosome, acting as a microtubule-organizing center (MTOC), is a necessary prerequisite for pronuclear apposition.33 If a sperm-derived centrosome (aster) has an inherent defect that prevents it from nucleating the formation of microtubules, the apposition of pronuclei and syngamy in the human zygote fail.34 This mechanism also functions between the pronuclei of the same gamete (sperm) origin in polyspermically penetrated human zona-free oocytes (Figure 3.3). A similar mechanism may be in place in parthenogenetic human embryos, where mosaicism is commonly observed.35 POLARIZATION OF INTRAPRONUCLEAR STRUCTURES AND INTERPRONUCLEAR SYNCHRONY

Concomitantly with the apposition of the male and female pronuclei, chromatin and NPBs in both pronuclei polarize, and rotate to face the adjacent pronucleus. Moreover, in both pronuclei, chromatin that faces the other pronucleus becomes highly condensed during apposition.31 This mechanism seems to work only when pronuclei originate from different gametes (male and female), because only an insignificant, if any, intrapronuclear polarization can be observed in multiple male pronuclei developing in polyspermically penetrated human oocytes, even though the pronuclei lie in close apposition to each other (Figure 3.3). Circumstantial evidence suggests that the paternal pronucleus imposes, or collaborates in, the formation of polar axes in mammalian eggs.33 Chromatin located near the sperm tail becomes polarized facing the oocyte interior as the differentiating paternal pronucleus rotates, and this occurs even while the sperm head lies distant from the maternal pronucleus.31 The failure of pronuclei to come into apposition is usually associated with a failure of cleavage and

HPE_Chapter03.qxp

7/13/2007

5:22 PM

Page 25


arrest at the zygote stage. Zygotes with pronuclei of uneven size nearly all develop into mosaic embryos.36 On the other hand, failure of NPB polarization can occur even if pronuclei are in apposition, and this usually involves only one of the two pronuclei. Such interpronuclear asynchrony was shown to be associated with abnormalities of further embryonic development.10 Based on the finding that intrapronuclear RNA synthesis is required for NPB development and growth,7 together with the observation that the failure of NPBs to polarize is typically associated with a reduced NPB size,10 interpronuclear asynchrony seems to be caused by abnormal gene expression in the pronucleus that is lagging behind.

ANALYSIS OF PRONUCLEI IN LIVING HUMAN ZYGOTES VISUALIZATION OF PRONUCLEI

Using Hoffman modulation contrast optics, the most widely used optical system in the current IVF and ICSI practice, pronuclei can be easily observed in living human zygotes only after completion of pronuclear envelope formation (Figure 3.4). According to electron microscopic studies,21 this occurs about 10–12 hours after fertilization. When pronuclear differentiation and DNA synthesis are completed, the pronuclear envelopes disintegrate, the chromatin of both pronuclei forms a single spindle, and the zygote enters the first cleavage division. This can occur as early as 18–20 hours after fertilization. The optimal time window for pronuclear observation is thus between 12 and 16 hours after in vitro insemination or ICSI. Both immature and fully developed NPBs can be visualized by Hoffman modulation contrast optics. Since the fully developed NPBs are formed by fusion between several NPBs that are immature with the participation of chromatin,6,21 with advancing pronuclear development, fewer NPBs can be visualized, and the size of each NPB is greater. This process coincides with NPB migration towards the area of interpronuclear contact. Hence, the number, size, and position of NPBs are key elements of any pronuclear scoring system used in living human zygotes. Changes in the intrapronuclear distribution of chromatin, which is also a part of pronuclear developmental changes (see above), are inaccessible to observation in the living state. POSSIBILITIES AND LIMITS OF NON-INVASIVE PRONUCLEAR EVALUATION

Figure 3.3 Electron micrograph showing multiple sperm-derived pronuclei in a polyspermically penetrated human oocyte. The pronuclei are in close apposition with each other. ⫻12 000.

The development of pronuclear scoring systems was initially largely motivated by the need to select the best embryos for fresh transfer and to determine which embryos were to be cryopreserved, as early as the zygote stage.8 The first clinical study to describe a coherent pronuclear scoring system9 was performed 

HPE_Chapter03.qxp

7/13/2007

5:22 PM

Page 26


Figure 3.4 Micrographs showing the size and distribution of nucleolar precursor bodies (NPB) in pronuclei of human zygotes at different phases of pronuclear development. (A) Relatively early phase of pronuclear development, characterized by a high number of NPB in both pronuclei (only part of them is visible at this focal level). The NPB are relatively small at this phase and are distributed randomly in the pronuclei. (B) Later phase of pronuclear development, characterized by a low number of NPB in both pronuclei. The NPB are larger at this phase and show a polarized distribution, with accumulation near that pole of each pronucleus at which it makes contact with the other one. (Hoffman modulation contrast, original magnification ⫻200.) Reprinted from Tesarik and Kopecny.7

in relation to embryo transfer at the zygote stage. Although transfer at the zygote stage is rather unusual nowadays, the issue of pronuclear scoring was revisited in studies aiming at zygote selection for fresh transfer and for cryopreservation. The policy of freezing embryos at the zygote stage was considered as an alternative to freezing embryos at 

cleavage stages in the context of legal restrictions prohibiting embryo cryopreservation beyond the zygote stage.37–39 More recently, it was noted that neither zygote evaluation nor cleaving embryo evaluation in isolation can yield an optimal prediction of embryo developmental potential. In combination, however, zygote and cleaving embryo evaluation yield a highly efficient embryo selection system which offers an implantation rate after embryo transfer on day 3 after ICSI that is comparable to transfer of day-5 embryos at the blastocyst stage.40 The potential usefulness of pronuclear scoring as a means of selection against embryos carrying chromosomal abnormalities has also been reported.36,41–43 However, pronuclei are highly dynamic structures, and essential features may sometimes escape attention due to the limited time period during which they can be analyzed by non-invasive means. Moreover, the ability of a zygote showing various pronuclear abnormalities to recover and develop normally may be conditioned by independent factors, such as maternal age, ovarian reserve or sperm quality, which still remain to be analyzed in this context. These caveats may be responsible for the inconsistencies sporadically reported in the literature as to the value of pronuclear scoring for prediction of embryo developmental potential.43 PRONUCLEAR SCORING SYSTEMS

Most of the current studies evaluating pronuclear morphology in living human zygotes use one of the two most popular scoring systems, described by Scott and Smith9 and by Tesarik and Greco.10 The former scoring system for evaluation of living human pronuclear-stage zygotes6 was based on the definition of five patterns, graded 1–5 according to nuclear size, nuclear alignment, nucleolar alignment and distribution, and position of the nuclei within the zygote. Grade 1 zygotes had equal numbers of nucleoli aligned at the pronuclear junction. The absolute number was not counted but was between three and seven. Grade 2 had equal numbers of nucleoli of equal sizes in the same nuclei but with one nucleus having alignment at the pronuclear junction and the other with scattered

HPE_Chapter03.qxp

7/13/2007

5:22 PM

Page 27


nucleoli. Grade 3 zygotes had equal numbers and sizes of nucleoli (between three and seven) which were equally scattered in the two nuclei. Grade 4 zygotes had unequal numbers (a difference of more than one nucleolus) and/or sizes of nucleoli. Grade 5 zygotes were those with pronuclei that were not aligned, were of grossly different sizes or were not located in the central part of the zygote.9 This system was later refined and simplified by grouping grade 2 and 4 together (Z3 score). The most desirable pronuclear patterns in this new Z-scoring system are Z1 and Z2.45 The appearance of cytoplasm around the pronuclei, forming a characteristic ‘halo’ was considered an additional quality parameter in this scoring system. The other widely used pronuclear scoring system10 has the advantage of simplicity, as it requires only a single static observation of the zygotes, 12–16 hours after in vitro insemination or ICSI. It is also very simple because it does not include cytoplasmic appearance as an additional zygote evaluation parameter. In fact, a subsequent study has shown that cytoplasmic appearance has little importance in the prediction of zygote and embryo developmental potential.46 The scoring system proposed by Tesarik and Greco10 is based on the evaluation of the synchrony of developmental changes (with regard to the size and position of NPBs) in both pronuclei rather than the timing of these changes. Accordingly, zygotes in which differences between both pronuclei can be observed at this stage have a greater risk of cleavage arrest, poor cleaving-embryo morphology, and the appearance of blastomere multinucleation as compared with zygotes in which pronuclear differentiation progresses in synchrony.10

ETIOLOGY OF PRONUCLEAR DEVELOPMENTAL PERTURBATIONS PATERNAL FACTOR

In view of the complex transformations the sperm nucleus must undergo in the oocyte cytoplasm before it forms the male pronucleus, the paternal contribution to pronuclear developmental perturbations

is highly probable. However, an unequivocal demonstration of such a contribution was made difficult by the frequent co-existence of sperm and oocyte abnormalities in current IVF and ICSI programs. The first direct evidence for the role of a sperm factor in the etiology of pronuclear developmental perturbations came from studies in which donor oocytes were shared.47 Consequently, it is evident that ICSI with spermatozoa from some men repeatedly results in the formation of zygotes with pronuclear abnormalities, and these patients achieve lower pregnancy and implantation rates than couples who share oocytes from the same donors. This adverse paternal effect is often associated with abnormal basic sperm parameters, but is also observed in men with normal sperm.47 OOCYTE FACTORS AND OVARIAN STIMULATION

The finding that poor oocyte quality can also cause abnormalities of pronuclear development came from a study analyzing the relationship between serum luteinizing hormone (LH) levels and embryo quality in an oocyte donation program.48 When serum LH levels during ovarian stimulation were too low or too high, the frequency of pronuclear abnormalities in the resulting zygotes was higher as compared with oocytes from donors whose LH level was between 0.5 and 1.0 IU per liter.48 Further studies are needed to obtain more information about the effect of different aspects of ovarian stimulation on pronuclear development.

SUMMARY

The development of pronuclei is a complex sequence of events that involves the nuclear envelope, chromatin, and nucleolar precursor bodies (NPBs). The timely occurrence of these events is a prerequisite for correct pronuclear function and further embryonic development. These events were analyzed with the use of electron microscopy and autoradiography in polyspermically penetrated human zona-free oocytes. Some of these events, namely those concerning the development of NPBs, can also be visualized 

HPE_Chapter03.qxp

7/13/2007

5:22 PM

Page 28


by non-invasive examination of living human zygotes with the use of Hoffman modulation contrast optics. Pronuclear scoring systems, based on the determination of the number, size, and distribution of NPBs, have been suggested to predict further embryo development and implantation potential. The etiology of abnormal pronuclear development can be related to both sperm and oocyte quality. REFERENCES 1. Yanagimachi R. Mammalian fertilisation. In: Knobil E, Neill J, eds. The Physiology of Reproduction, 2nd edn. New York: Raven Press, 1994: 189–317. 2. Steptoe PC, Edwards RG. Birth after the reimplantation of a human embryo. Lancet 1978; 2: 366. 3. Bedford JM, Kim HH. Sperm/egg binding patterns and oocyte cytology in retrospective analysis of fertilization failure in vitro. Hum Reprod 1993; 8: 453–63. 4. Levron J, Munne S, Willadsen S, Rosenwaks Z, Cohen J. Male and female genomes associated in a single pronucleus in human zygotes. Biol Reprod 1995; 52: 653–7. 5. Tesarik J, Kopecny V. Developmental control of the human male pronucleus by ooplasmic factors. Hum Reprod 1989; 4: 962–8. 6. Tesarik J, Kopecny V. Nucleic acid synthesis and development of human male pronucleus. J Reprod Fertil 1989; 86: 549–58. 7. Tesarik J, Kopecny V. Assembly of the nucleolar precursor bodies in human male pronuclei is correlated with an early RNA synthetic activity. Exp Cell Res 1990; 191: 153–6. 8. Wright G, Wiker S, Elsner C et al. Observations on the morphology of pronuclei and nucleoli in human zygotes and implications for cryopreservation. Hum Reprod 1990; 5: 109–15. 9. Scott LA, Smith S. The successful use of pronuclear embryo transfers the day following oocyte retrieval. Hum Reprod 1998; 13: 1003–13. 10. Tesarik J, Greco E. The probability of abnormal preimplantation development can be predicted by a single static observation on pronuclear stage morphology. Hum Reprod 1999; 14: 1318–23. 11. Tesarik J, Junca AM, Hazout A et al. Embryos with high implantation potential after intracytoplasmic sperm injection can be recognized by a simple, non-invasive examination of pronuclear morphology. Hum Reprod 2000; 15: 1396–9. 12. Tesarik J, Mendoza C. In vitro fertilization by intracytoplasmic sperm injection. Bioessays 1999; 21: 791–201. 13. Jones KT. Mammalian egg activation: from Ca2⫹ spiking to cell cycle progression. Reproduction 2005; 130: 813–23. 14. Whitaker M. Calcium at fertilization and in early development. Physiol Rev 2006; 86: 25–88. 15. Tesarik J, Kopecny V, Plachot M, Mandelbaum J. Activation of nucleolar and extranucleolar RNA synthesis and changes in the ribosomal content of human embryos developing in vitro. J Reprod Fertil 1986; 78: 463–70. 16. Tesarik J, Kopecny V, Plachot M et al. Nucleologenesis in the human embryo developing in vitro: ultrastructural and autoradiographic analysis. Dev Biol 1986; 115: 193–203. 17. Braude P, Bolton V, Moore S. Human gene expression first occurs between the four- and eight-cell stages of preimplantation development. Nature 1988; 332: 459–61.



18. Tesarik J, Kopecny V, Plachot M, Mandelbaum J. Early morphological signs of embryonic genome expression in human preimplantation development as revealed by quantitative electron microscopy. Dev Biol 1988; 128: 15–20. 19. Sousa M, Barros A, Tesarik J. Developmental changes in calcium dynamics, protein kinase C distribution and endoplasmic reticulum organization in human preimplantation embryos. Mol Hum Reprod 1996; 2: 967–77. 20. Sousa M, Barros A, Mendoza C, Tesarik J. Effects of protein kinase C activation and inhibition on sperm-, thimerosal-, and ryanodineinduced calcium responses of human oocytes. Mol Hum Reprod 1996; 2: 699–708. 21. Tesarik J, Kopecny V. Development of human male pronucleus: ultrastructure and timing. Gamete Res 1989; 24: 135–49. 22. Scott L. Pronuclear scoring as a predictor of embryo development. Reprod Biomed Online 2003; 6: 201–14. 23. Tesarik J, Kopecny V, Plachot M, Mandelbaum J. High resolution autoradiographic localization of DNA-containing sites and RNA synthesis in developing nucleoli of human preimplantation embryos: a new concept of embryonic nucleologenesis. Development 1987; 101: 777–91. 24. Sundström P, Nilsson O, Liedholm P. Cleavage rate and morphology of early human embryos obtained after artificial fertilization and culture. Acta Obstet Gynecol Scand 1981; 60: 109–20. 25. Ao A, Erickson RP, Winston RML, Handyside AH. Transcription of paternal Y-linked genes in the human zygote as early as the pronucleate stage. Zygote 1994; 2: 281–7. 26. Aoki F, Worrad DM, Schultz RM. Regulation of transcriptional activity during the first and second cell cycles in the preimplantation mouse embryo. Dev Biol 1997; 181: 296–307. 27. Van Blerkom J. Structural relationships and post-translational modification of stage-specific proteins synthesised during early preimplantation development in the mouse. Proc Natl Acad Sci USA 1981; 78: 7629–33. 28. Nothias JY, Majumder S, Kaneko KJ, DePamphilis ML. Regulation of gene expression at the beginning of mammalian development. J Biol Chem 1995; 270: 22077–80. 29. Payne D, Flaherty SP, Barry MF, Matthews CD. Observations on polar body extrusion and pronuclear formation in human oocytes using time-lapse cinematography. Hum Reprod 1997; 12: 532–41. 30. Van Blerkom J. Developmental failure in human reproduction associated with preovulatory oogenesis and preimplantation embryogenesis. In: Van Blerkom J, Motta P, eds. Ultrastructure of Human Gametogenesis and Early Embryogenesis. Boston: Kluwer Acad. Pub., 1989: 125–80. 31. Van Blerkom J, Davis P, Merriam J, Sinclair J. Nuclear and cytoplasmic dynamics of sperm penetration, pronuclear formation and microtubule organisation during fertilization and early preimplantation development in the human. Hum Reprod Update 1995; 1: 429–61. 32. Sathananthan AH, Ratnam SS, Ng SC et al. The sperm centriole: its inheritance, replication and perpetuation in early human embryos. Hum Reprod 1996; 11: 345–56. 33. Edwards RG, Beard HK. Oocyte polarity and cell determination in early mammalian embryos. Mol Hum Reprod 1997; 3: 863–905. 34. Asch R, Simerly C, Ord T, Schatten G. The stages at which human fertilization arrests: microtubule and chromosomal configurations in inseminated oocytes which failed to complete fertilization and development in humans. Mol Hum Reprod 1995; 1 see Hum Reprod 1995; 10: 1897–906. 35. Palermo G, Munne S, Cohen J. The human zygote inherits its mitotic potential from the male gamete. Hum Reprod 1994; 9: 1220–25.

HPE_Chapter03.qxp

7/13/2007

5:22 PM

Page 29


36. Sadowy S, Tomkin G, Munne S, Ferrara-Congedo T, Cohen J. Impaired development of zygotes with uneven pronuclear size. Zygote 1998; 6: 137–41. 37. Ludwig M, Schopper B, Al-Hasani S et al. Clinical use of a pronuclear stage score following intracytoplasmic sperm injection: impact on pregnancy rates under the conditions of the German Embryo Protection Law. Hum Reprod 2000; 15: 325–9. 38. Montag M, van der Ven H. Evaluation of pronuclear morphology as the only selection criterion for further embryo culture and transfer: results of a prospective multicenter study. Hum Reprod 2001; 16: 2384–9. 39. Zollner U, Zollner KP, Hartl G, Dietl J, Steck T. The use of a detailed zygote score after IVF/ICSI to obtain good quality blastocysts: the German experience. Hum Reprod 2002; 17: 1327–33. 40. Rienzi L, Ubaldi F, Iacobelli M et al. Day 3 embryo transfer with combined evaluation at the pronuclear and cleavage stages compares favourably with day 5 blastocyst transfer. Hum Reprod 2002; 17: 1852–5. 41. Coskun S, Hellani A, Jaroudi K et al. Nucleolar precursor body distribution in pronuclei is correlated to chromosomal abnormalities in embryos. Reprod Biomed Online 2003; 7: 86–90. 42. Gámiz P, Rubio C, de los Santos MJ et al. The effect of pronuclear morphology on early development and chromosomal abnormalities in cleavage-stage embryos. Hum Reprod 2003; 18: 2413–19.

43. Balaban B, Yakin K, Urman B, Isiklar A, Tesarik J. Pronuclear morphology predicts embryo development and chromosome constitution. Reprod Biomed Online 2004; 8: 695–700. 44. Salumets S, Hydén-Granskog C, Suikkari AM, Tiitinen A, Tuuri T. The predictive value of pronuclear morphology of zygotes in the assessment of human embryo quality. Hum Reprod 2001; 16: 2177–81. 45. Scott L, Alvero R, Leondires M, Miller B. The morphology of human pronuclear embryos is positively related to blastocyst development and implantation. Hum Reprod 2000; 15: 2394–403. 46. Ebner T, Moser M, Sommergruber M et al. Presence, but not type or degree of extension, of a cytoplasmic halo has a significant influence on preimplantation development and implantation behaviour. Hum Reprod 2003; 18: 2406–12. 47. Tesarik J, Greco E, Mendoza C. Paternal effects acting during the first cell cycle of human preimplantation development after ICSI. Hum Reprod 2000; 17: 184–9. 48. Tesarik J, Mendoza C. Effects of exogenous LH administration during ovarian stimulation of pituitary down-regulated young oocyte donors on oocyte yield and developmental competence. Hum Reprod 2002; 17: 3129–37.



HPE_Chapter03.qxp

7/13/2007

5:22 PM

Page 30

HPE_Chapter04.qxp

7/13/2007

4:45 PM

Page 31

4. Human pronuclei as a mode of predicting viability Aidita N James

INTRODUCTION

Age-corrected implantation rates with the use of assisted reproductive technologies (ART) have been on the increase over the past 25 years. Despite this success, supernumerary embryos are still transferred simultaneously, with the risk of increased multiple pregnancy rates. Although enforced restrictive policies of single embryo transfer have yet to be enacted in most countries, the desire to select the best embryos and predict implantation has aided the development of several evaluation methods to enhance selection of embryos with optimal prognosis for implantation. Embryo selection is most commonly based on morphological criteria that are assessed postfertilization, including both pronuclear and cleavage stage embryo evaluation. Studies of phenomena related to development rate, pronuclear and nucleolar behavior, blastomere fragmentation, and multinucleation have been of particular interest.1–8 Events related to pronuclear and nucleolar movements were first described by Wright et al2 and Van Blerkom.9 Recently, these phenomena have been expressed as more distinct pronuclear scoring systems that can be used for the purpose of selecting embryos. The scores have been correlated with improved embryo development5,10–13 as well as with increased pregnancy and implantation rates.5,10,12–20 A number of different pronuclear scoring systems have been proposed by different ART laboratories, and are in use to select high quality embryos; however, no standardized scoring system for zygote grading is currently in use. Unfortunately, the same holds true for other systems used to evaluate embryo morphology. As a result, comparing success rates between laboratories while controlling for embryo quality is a challenge. A consistent scoring system, such as a

pronuclear scoring system in combination with other embryo development markers and patient status, may be useful in determining the optimal number of embryos to transfer. This could lead to an increase in pregnancy and implantation rates, while decreasing the number of high order multiple pregnancies. PRONUCLEAR SCORING METHODOLOGY

Pronuclear scoring is a non-invasive examination that relies upon one static observation of simple morphological parameters, and is routinely performed at 16–18 hours postinsemination with both standard IVF and ICSI. However, pronuclear examination for scoring purposes can be more time-consuming than the usual cursory examination used for determining fertilization. Caution should always be used whenever embryos are removed from incubators for prolonged periods to perform any type of examination. Cumulus cells must first be removed from zygotes that result from oocytes that were inseminated with standard IVF procedures, to allow an unobstructed view of the pronuclei during analysis. All pronuclear scoring systems attempt to classify zygotes based on the characteristics of the two pronuclei as well as the nucleoli. Pronuclei are scored based on the following criteria: symmetry (equal vs unequal), position (in apposition vs at a distance), and location (central vs non-central). The nucleoli are scored based on number (3 to 7), symmetry (equal vs unequal sizes), and location (polarized or aligned vs non-polarized or non-aligned). Other observations factored into the score might include polar body morphology and alignment, cytoplasmic morphology (presence vs absence of halo; i.e., cytoplasmic contraction), and finally a second embryo evaluation at 25–27 hours postinsemination for pronuclear 

HPE_Chapter04.qxp

7/13/2007

4:45 PM

Page 32


morphology (presence vs absence of pronuclei) and/or early cell division. Evaluation of pronuclear morphology is most efficient or accurate under phase-contrast microscopy utilizing either Hoffman modulation contrast (HMC) or differential-interference contrast (DIC) optics. During this assessment, the technician should visualize the zygote through several focal planes to ensure the best possible evaluation of morphology. Visualization in only one focal plane can be misleading and result in an incorrect assessment of zygote morphology. Pronuclear morphology is not a static event, and therefore all embryos should be assessed at the same time point postinsemination in order to avoid incorrect assessment due to a bias based on timing.2

PRONUCLEAR SCORING PARAMETERS AND THEIR IMPORTANCE

A suspicion that major abnormalities of pronuclear development (i.e. pronuclei of unequal sizes, at a distance or not centrally located within the zygote cytoplasm) are incompatible with normally progressing development is not a novel concept. A number of studies have confirmed that zygotes exhibiting unequal pronuclear sizes are associated with an increased risk of embryonic chromosomal anomalies.4,8,21 The sperm-derived centriole and associated microtubule organizing region function to reposition the pronuclei into apposition,2,22,23 and visual changes in the pronuclei may thus correlate with functional abnormalities in chromosome segregation. While parameters of pronuclear size, location, and position are obvious and easily assessed, other morphological parameters are less distinct, including polarization of the nucleoli or nucleolar precursor bodies (NPB) along the pronuclear junction. Nucleolar precursor body alignment has been related to the chromatin rotation that occurs in developing pronuclei,14,24 and therefore anomalies in this parameter potentially have severe consequences. Chromatin rotation plays an important role in establishing the embryonic axis, an essential part of cell determination for preimplantation embryos.25 Tesarik and Kopecny24 reported that NPB alignment is not evident at the 

beginning of pronuclear formation but progresses over time as the NPB coalesce to form fewer, larger nucleolar structures within the pronuclei. This again demonstrates the fluidity of pronuclear morphology. Different numbers and dimensions of nucleoli are compatible with development, although assessments may be inaccurate because of light microscopic limitations. It is likely that there is considerable biological variation within normal development. The alignment of the pronuclei and polar bodies is another phenomenon in pronuclear morphology assessment that has been less well studied. This relationship is critical, because it eventually correlates with the polar axis of the first cleavage division as defined by the second polar body.14 The alignment that is currently hypothesized to be most ideal is for the pronuclei to be parallel to the polar axis (Figure 4.1). Perpendicular placement of the pronuclei would instead position one pronucleus on the opposite side of the cleavage plane. This misalignment could subsequently result in incorrect nuclear division followed by chromosomally abnormal embryo development.13 The appearance of a cytoplasmic halo has been associated with the redistribution of mitochondria in the human zygote after fertilization, and has been positively correlated with blastocyst quality.26–28 This redistribution should place the majority of the

Perpendicular configuration

Parallel configuration

Pronuclei First and second polar body

Figure 4.1 Graphical representation of pronuclear orientation as it relates to the second polar body.

HPE_Chapter04.qxp

7/13/2007

4:45 PM

Page 33

HUMAN PRONUCLEI AS A MODE OF PREDICTING VIABILITY

mitochondria around the pronuclei, producing a cytoplasmic clearing around the periphery of the ooplasm and thereby creating the appearance of a halo.29 Incorrect distribution of mitochondria and associated ATP could lead to unequal cellular division and abnormal embryo metabolism. Finally, early embryonic cleavage at 25–27 hours postinsemination has been associated with increased embryo implantation potential and live birth rates.7,30,31 This correlation may be due to cytoplasmic and/or nuclear maturity. It is plausible that oocytes with optimal cytoplasmic/nuclear maturity may lead to more appropriate metabolism (ATP, mitochondria, etc.) in the embryo, so that it is better able to perform cellular functions.32

PRONUCLEAR SCORING SYSTEMS

Scott and Smith15 proposed a Z score system for assessing pronuclear morphology, and this was later modified by Scott et al5 and Scott.13 Tesarik and Greco33 developed another system for scoring pronuclei, which was further modified by Tesarik et al18 to create scores based on pattern 0 vs non-pattern 0 groups. In response to pressure from German and Swiss law to aid in the prediction of pregnancy and blastocyst development Ludwig et al,16 Zollner et al12 and Senn et al34 developed further systems for scoring pronuclear morphology. Both Germany and Switzerland have extremely restrictive IVF laws that allow culture only of embryos that will be transferred,

which thereby limits embryo selection to choosing zygotes prior to the first cellular division. All of the different scoring systems are based on a common theme, that is the value of symmetry in all of the morphological parameters that are examined. THE Z SCORE

One of the early pronuclear scoring systems, proposed by Scott and Smith,15 was based on a total score derived from values awarded for specific parameters of morphology, and the sum of these values was used to determine which embryos would have a higher implantation potential. The parameters assessed were pronuclear size and alignment, alignment of nucleoli within the pronuclei, and cytoplasmic morphology (Table 4.1). An additional score of 10 points was awarded to embryos for early cell division observed at a later embryo evaluation (24–26 hours postinsemination). The total sum yields a score ranging from 7 to 25, with a higher score related to greater implantation potential. Scott and Smith15 reported that a pronuclear morphology score of ⱖ15 was associated with a pregnancy rate of 71%. Scott et al5 then revised the scoring system to include only parameters that could be observed at the time of fertilization assessment. This system classified zygotes into five basic groups, denoted as a Z score (Figure 4.2). The system was subsequently revised to classify zygotes into four basic categories (Figure 4.2, Table 4.2). Numerous reports have confirmed the zygote scoring system known as Z scoring

Table 4.1 The pronuclear scoring system created by Scott and Smith.15 A cumulative point system with values ranging from 3 to 35 with an ideal of ⱖ15 correlating to increased pregnancy rates Points awarded for various morphological parameters Parameters

5

4

3

1

Pronuclear position

In apposition

NA

NA

NPB position

Aligned

Unaligned

Cytoplasmic morphology

Clear halo

Coming into alignment NA

At a distance or very unequal in size NA

Non-homogenous and/or darkened

NA

NPB, nucleolar precursor bodies (nucleoli); NA, not applicable.



HPE_Chapter04.qxp

7/13/2007

4:45 PM

Page 34


Scott et al5 Series 1

Scott et al5 Series 2

Scott13

Tesarik and Greco33

Tesarik et al18

Grade 1

Z1

Z1

Pattern 0

Pattern 0

Grade 3

Z2

Z2

Pattern 0

Pattern 0

Grade 2

Z3

Z3-2

Pattern 2

Non-pattern 0

Grade 4

Z3

Z3-1

Pattern 5

Non-pattern 0

Grade 4

Z3

Z3-4

Pattern 1

Non-pattern 0

Grade 4

Z3

Z3-4

Pattern 3

Non-pattern 0

Grade 5

Z3

Z3-1

Pattern 4

Non-pattern 0

Grade 5

Z4

Z4-2

Pattern 4

Non-pattern 0

Grade 5

Z4

Z4-1

NA

Non-pattern 0

Pronuclear morphology

NA, not applicable.

Figure 4.2 Comparison of the various pronuclear morphology scoring systems by Scott’s group and by Tesarik’s group and a representative illustration of pronuclei in each scoring group.



HPE_Chapter04.qxp

7/13/2007

4:45 PM

Page 35


to be useful in the selection of embryos with improved implantation rates5 and therefore this system has been widely accepted in ART. The Z score system was later revised13 in order to incorporate further data and new evaluation parameters, the presence of a cytoplasmic halo and the orientation of the pronuclei in relation to the second polar body, as outlined in Figure 4.2. This version was then adapted to define Z3 and Z4 scores, which included these parameters.5 The presence of a halo was correlated to faster development and better morphological scores on day 3 and to higher blastulation rate on day 5 of culture. Pronuclear/polar body orientation was not observed to correlate with embryo development. Consistent with the older versions of the Z scoring system, embryos that were derived from zygotes scored as Z1 or Z2 had a better blastocyst development rate and a higher implantation potential. 0 PATTERNS

Tesarik and Greco33 proposed yet another pronuclear scoring system, based upon one static observation made on an unobstructed view of the entire volume

of both pronuclei at the time of fertilization assessment; this assessment noted the number and distribution of the NPBs within each pronucleus (aligned vs unaligned). The size (equal vs unequal) and position (in apposition vs at a distance) of both pronuclei was also evaluated during this assessment. The zygotes were categorized into six different groups (patterns 0 to 5) (Figure 4.2, Table 4.3). This system did show some difference in the percentage of embryos that developed into good morphology embryos on day 3 of culture, based upon their respective pattern grouping. Tesarik et al18 then further simplified their scoring system to one that placed embryos into two groups, pattern 0 and non-pattern 0. Non-pattern 0 comprised all of the other pattern groupings, from 1 to 5. The use of this scoring system increased the prediction of pregnancy and implantation rates with the use of a single, static observation at the zygote stage. The authors reported pregnancy and implantation rates of 45% and 30%, respectively, for embryos derived from pattern 0 zygotes, and 22% and 11%, respectively, for those derived from non-pattern 0 zygotes. CUMULATIVE POINT SCORING

In response to the constraints of the German IVF laws, Ludwig et al16 adapted the pronuclear scoring Table 4.2 The simplified pronuclear scoring system developed by Scott et al5 with scores of Z1 and Z2 considered morphologically normal zygotes Z score

Morphological description

Z1

Equal PN Equal number and size of nucleoli (ranging between 3 and 7) All nucleoli aligned at the pronuclear junction in both PN Equal PN Equal number and size of nucleoli (ranging between 3 and 7) Nucleoli unaligned in both PN Equal PN Equal number and even and/or uneven size of nucleoli (ranging between 3 and 7) Nucleoli aligned at the pronuclear junction in one PN and unaligned in the other PN Unequal or separated or not centrally located PN

Z2

Z3

Z4

PN, pronuclei.

Table 4.3 The pronuclear scoring system developed by Tesarik and Greco,33 with normal zygotes assigned to the pattern 0 group Pattern

Morphological description

0

Equal number and size of NPB (ranging between 3 and 7) All NPB either polarized or non-polarized in both pronuclei Unequal number (difference ⬎ 3) of NPB between both pronuclei Small number (⬍ 7) of NPB Non-polarized in at least one pronuclei Large number (⬎ 7) of NPB Polarized in at least one pronuclei Very small number (⬍ 3) of NPB in at least one pronuclei Polarized NPB in one pronucleus and non-polarized NPB in the other pronucleus

1 2 3 4 5

NPB, nucleolar precursor bodies (nucleoli).



HPE_Chapter04.qxp

7/13/2007

4:45 PM

Page 36


number of nucleoli, and nucleolar polarization. Lower CPNS scores were positively correlated with embryo implantation. They found that polarization of nucleoli and the appearance of a halo were the most important predictive factors of implantation potential. Zollner et al12 developed a pronuclear scoring system that is also similar to that of Scott and Smith.15 Zygotes were given points for ten different criteria, with a lower total morphology score (ranging from 10 to 30) correlating to better zygote quality. The ten criteria used to assess zygote quality were the number of pronuclei, position of pronuclei, size of pronuclei, halo effect, alignment of nucleoli in pronucleus number one, alignment of nucleoli in pronucleus number two, number of nucleoli in pronucleus number one, number of nucleoli in pronucleus number two, appearance of vacuoles, and appearance of ooplasm. Points awarded for each characteristic are described in Table 4.4. Zollner et al12 reported that a mean zygote score of ⱕ15 was positively correlated to increased blastocyst quality for ICSI derived embryos. Interestingly, this was not found to be the case with IVF derived embryos.

system developed by Scott and Smith,15 using the same point scoring system. However, further assessment of pronuclear membrane breakdown and the first cellular division were excluded because of the constraints of the IVF laws. They reported that the use of this scoring system provided a negative predictive value of 92% when the threshold pronuclear score was set at 13. Embryos transferred from zygotes that had earned a score of ⱖ13 produced a pregnancy rate of 22%, whilst those with a score of ⬍13 resulted in a pregnancy rate of 4%.16 The pronuclear morphology scoring system suggested by Senn et al34 is also based upon a point system to determine the implantation potential of zygotes prior to assignment of embryo status (for transfer or for cryopreservation). According to Swiss law, embryo selection is prohibited, forcing practitioners to select zygotes for immediate cryopreservation or fresh transfer. Like ART facilities in Germany, a scoring system at the pronuclear stage is beneficial in aiding practitioners to select the zygotes with the greatest implantation potential. Senn et al34 developed the cumulative pronuclear score (CPNS) as a sum of points assigned to six morphological parameters, scored from 1 (best) to 3 (worst) with a CPNS, ranging from 6 to 18: pronuclear position, pronuclear orientation in relation to the second polar body, pronuclear location (centering), cytoplasmic halo,

PRONUCLEI MORPHOLOGY SCORING SYSTEMS

Kahraman et al35 created a pronuclei morphology scoring (PNMS) system that considered the effect of

Table 4.4 The pronuclear scoring system developed by Zollner et al.12 A cumulative point system with values ranging from 10 to 30 with a lower value indicative of better morphology Points awarded for various morphological parameters Parameters

1

2

3

4

PN number PN position PN size Halo effect NPB polarity (PN1) NPB polarity (PN2) NPB number (PN1) NPB number (PN2) Vacuoles Ooplasm

2 PN IA Equal Normal Aligned Aligned 3–5 3–5 No vacuoles Homogenous Gr

NA Nearby Unequal Light CIA CIA ⬎5 ⬎5 Lightly Vac Strong Gr

NA Not IA NA Extreme Unaligned Unaligned ⬍3 ⬍3 Strongly Vac NA

1 PN NA NA None No NPB/no PN No NPB/no PN No NPB/no PN No NPB/no PN NA NA

PN, pronucleus; IA, in apposition; NPB, nucleolar precursor bodies (nucleoli); CIA, coming into alignment; NA, not applicable; Vac, vacuolated; Gr, granulation.



HPE_Chapter04.qxp

7/13/2007

4:45 PM

Page 37


sperm source (testicular sperm and round spermatids vs ejaculated sperm) on corresponding score and subsequent chromosomal status. The morphological parameters evaluated were pronuclear position and size, nucleolar polarity and size, and the halo effect (Table 4.5). The zygotes were placed into eight groups labeled as A, B, C, D, E, F, G, and J, and after further review categorized as group I (A–D) and group II (E–J). Group I zygotes from this scoring system are similar in morphology to Z scores of 1 and 2 as well as pattern 0 from previously discussed scoring systems, and these would be considered the more normal of the zygote morphologies. This study found a correlation between sperm source and resulting pronuclear morphology, as well as with PNMS score and further embryo development and chromosomal abnormalities. Zygotes in group II were more likely to be derived from testicular sperm or round spermatids (32.1%) versus ejaculated sperm (22.7%). A higher proportion of good quality blastocysts were derived from group I (50.4%) versus group II (28.2%). Finally, chromosomal abnormalities were lower in embryos from group I (37.6%) versus embryos from group II (56.2%). Gianaroli et al36 proposed another PNMS system based on three parameters, and used this to correlate chromosomal status with zygote morphology. The three parameters were location and position of pronuclei (A through E), size and polarity of nucleoli (1 through 4), and pronuclear orientation in relation to the second polar body (␣, ␤, and ␥) (Table 4.6).

Upon completion of chromosomal analysis it was determined that four configurations had a greater chance of producing euploid embryos (A1␣, A2␤, A3␤, and A3␣) which correlated to better embryonic development.

STUDIES UTILIZING ZYGOTE SCORING SYSTEMS

Testing of these pronuclear morphology scoring systems by various ART facilities reported inconsistent results in their usefulness in predicting embryonic implantation potential. This inconsistency is twofold; first, it can be attributed to the multifactorial situations in which nearly all of these studies were performed, and second, a static observation cannot be used as the sole predictive source of embryonic implantation potential. The predictive value of zygote scoring can be increased by combining several morphological factors, evaluated at different time points during embryonic development. Tesarik and Greco33 reported a 50% (22/44) pregnancy rate following the transfer of at least one embryo that was derived from a zygote of good morphology (pattern 0) versus a 9% (2/23) pregnancy rate for the transfer of embryos derived from abnormal zygotes (patterns 1–5). Another study correlated pattern 0 zygotes with higher quality embryos of greater implantation potential than embryos

Table 4.5 Pronuclear morphology scoring (PNMS) system created by Kahraman et al.35 Zygotes in subgroups A through D are considered to have normal morphology that correlates to embryo development and the incidence of chromosomal abnormalities Group

Sub-group

Halo

PN position

PN size

I

A B C D E F G

Distinct Distinct Distinct Distinct Distinct Distinct Blurred

IA IA IA IA IA IA IA

J

Blurred

Not IA

Equal Equal Equal Equal Equal Equal Equal or unequal Equal

II

Nucleolar alignment

Nucleolar size

Aligned Aligned Scattered Scattered Aligned Scattered Unknown

Large Small Large Small 1 Large, 1 Small PN 1 Large, 1 Small PN Unknown

Syngamic

Large or Small

PN, pronucleus; IA, in apposition.



HPE_Chapter04.qxp

7/13/2007

4:45 PM

Page 38


Table 4.6 The pronuclear morphology scoring (PNMS) system developed by Gianaroli et al,36 with normal morphology configurations being A1␣, A2␤, A3␤, and A3␣ Parameter

Configuration

Pronuclei (location, position, aother)

A B

Nucleoli (size, alignment)

Pronuclear orientation

C D E 1 2 3 4

␣ ␤ ␥

Parameter description Central, in apposition Non-central, in apposition Central, at a distance Unequal sizesa Fragmenteda Large, aligned Large, unaligned Large, 1 PN aligned Small, unaligned Parallel Perpendicular Other

PN, pronucleus.

originating from abnormal zygote patterns.19 Pattern 0 embryos33 correspond to Z1 and Z2 embryo scores as developed by Scott et al5 and Scott13 (Figure 4.2). Scott et al5 associated clinical pregnancies with day 3 transfers only when embryos derived from a zygote with a score of Z1 or Z2 were included in the transfer. In another study, Scott37 reported no live births from transfers involving only Z3- or Z4-derived embryos. Unlike Tesarik and Greco,33 Scott et al5 or Wittemer et al19, James et al38 demonstrated an equal chance of pregnancy and live birth from transfers consisting of only Z3-derived embryos replaced on day 3. Taking this one step further, Jaroudi et al39 reported absolutely no correlation between the selection of higher quality zygotes and pregnancy rates when embryos were selected for transfer using Tesarik and Greco’s system,33 or when retrospectively assessed using the Scott et al5 revised Z scoring system. Similar results were observed by Salumets et al40 who also found no difference in pregnancy or implantation rates between pattern 0 and non-pattern 0 zygotes in single embryo transfer patients. Payne et al41 reported a similar finding for embryo morphology and pregnancy rate from Z1- and Z3-derived embryos. A multicenter study completed in Germany showed that embryos selected for transfer based 

upon pronuclear morphology exhibited some differences in pregnancy and implantation rates exhibited according to the morphological scoring group.6 Germany is one of the few countries in which embryo transfer is limited by legislation so that embryos must be selected prior to development beyond the pronuclear stage. Despite the reported differences, the study showed that pregnancy and implantation still occur even if the pronuclear morphology was suboptimal, or less than a Z1 or Z2 score5 or a pattern 0.18 Balaban et al10 reported that embryos were more likely to develop into high quality blastocysts as well as achieve higher implantation and pregnancy rates if they derived from a zygote with an ideal pronuclear pattern as compared with those embryos derived from a non-ideal pronuclear pattern. Despite the better developmental prognosis of the embryos with ideal pronuclear pattern, they also found that blastocysts from abnormal pronuclear patterns would produce pregnancies, but at a lower rate. On the other hand, Scott et al5 reported that when only Z3derived blastocysts were available for transfer on day 5, there were no pregnancies (n ⫽ 6). In direct contrast to this study, James et al38 reported a live birth rate of 100% when only Z3-derived blastocysts were transferred (n ⫽ 4). Kattera and Chen42 reported on the predictive value of pronuclear orientation in relation to the second polar body upon embryo development and implantation and pregnancy rates. There was a positive correlation with early embryo cleavage when the pronuclei were perpendicular to the second polar body. However, there was no clear difference for pronuclear orientation in relation to implantation or pregnancy rates. A number of investigators have correlated chromosomal status to zygote pronuclear morphology. Sadowy et al21 reported an increased incidence of mosaicism associated with zygotes that showed pronuclei of unequal sizes. Kahraman et al35 also showed an increased number of chromosomally abnormal embryos derived from zygotes with suboptimal morphology. Another study also reported that zygotes with normal pronuclear patterns, evaluated using the method of Tesarik and Greco,33 had a greater chance of producing genetically normal embryos.43

HPE_Chapter04.qxp

7/13/2007

4:45 PM

Page 39


Gamiz et al44 stated that in patients ⱕ37 years of age, zygotes with equal sized pronuclei and polarized nucleoli were positively correlated with embryo development and a lesser incidence of chromosomal abnormalities.

CONCLUSIONS

Timing is a critical issue in the evaluation of early embryos, and this creates a challenge when assessing a dynamic event such as pronuclear morphology. To date, laboratories have evaluated this parameter within a time range that is based on the timing of the insemination procedure, with the rationale that there is a limited time range during which pronuclei can be visualized for evaluation. Furthermore, the timing of fertilization can depend upon the method of insemination as well as the actual oocyte undergoing treatment. The exact time of insemination is known with ICSI because each sperm is injected into individual oocytes at a precise time. This is not the case with IVF, in which the fertilization time can only be speculated. The amount of time sperm and oocytes are together during the insemination procedure for standard IVF can vary between laboratories, and can range from 4 to 18 hours. There is no way of knowing the time of sperm–oocyte fusion within this window. Montag and van der Ven6 reported that a greater number of higher quality embryos were derived from zygotes with a high score after ICSI versus IVF. This difference is probably related to a difference in the timing of insemination between the two groups and not to inherent differences in IVF itself. Pronuclear development is a dynamic event14,20,24,33 and to disregard an embryo based on a single time assessment can be misleading. Zygotes that are assigned a poor pronuclear score are not necessarily chromosomally aberrant, and have been shown to become high grade zygotes when assessed at a later time. Sadowy et al21 analyzed chaotic mosaicism in zygotes with uneven pronuclei and found no indication that all such zygotes were abnormal. This was further demonstrated by Wright et al2 and Demirel et al45 who observed morphological changes over time. They reported that the nucleoli were small,

numerous, and distributed in a random pattern early in pronuclear development, but that as time proceeded, the nucleoli coalesced and began to align near the furrow between the pronuclei. Adding to the previous literature showing the dynamic nature of pronuclear development, they demonstrated that timing of embryonic evaluation can have a drastic effect on the assigned quality when static observations are used.6,14,39,45 Jaroudi et al39 suggested that zygote scoring used as the only selection criteria may not be sufficient to select the most viable embryos. This has been reinforced by recent studies stating that evaluation of pronuclear morphology is useful for embryo selection only when used in conjunction with other methods of evaluation (i.e. day 3 morphology).46,47 Thus, a single observation may not reveal the true status of a zygote, because this cannot determine whether nuclear development is in progress, or has yet to progress when syngamy is initiated. It is important to use pronuclear morphology scoring carefully when determining the fate of embryos, because their potential can easily be misdiagnosed based upon a single point in time. REFERENCES 1. Tesarik J, Kopecny V, Plachot M et al. Ultrastructural and autoradiographic observations on multinucleated blastomeres of human cleaving embryos obtained by in-vitro fertilization. Hum Reprod 1987; 2: 127–36. 2. Wright G, Wiker S, Elsner C et al. Observations on the morphology of pronuclei and nucleoli in human zygotes and implications for cryopreservation. Hum Reprod 1990; 5: 109–15. 3. Winston NJ, Braude PR, Pickering SJ et al. The incidence of abnormal morphology and nucleocytoplasmic ratios in 2-, 3- and 5-day human pre-embryos. Hum Reprod 1991; 6: 17–24. 4. Munne S, Cohen J. Unsuitability of multinucleated human blastomeres for preimplantation genetic diagnosis. Hum Reprod 1993; 8: 1120–5. 5. Scott L, Alvero R, Leondires M et al. The morphology of human pronuclear embryos is positively related to blastocyst development and implantation. Hum Reprod 2000; 15: 2394–403. 6. Montag M, van der Ven H. Evaluation of pronuclear morphology as the only selection criterion for further embryo culture and transfer: results of a prospective multicentre study. Hum Reprod 2001; 16: 2384–9. 7. Lundin K, Bergh C, Hardarson T. Early embryo cleavage is a strong indicator of embryo quality in human IVF. Hum Reprod 2001; 16: 2652–7. 8. Manor D, Drugan A, Stein D et al. Unequal pronuclear size – a powerful predictor of embryonic chromosome anomalies. J Assist Reprod Genet 1999; 16: 385–9. 9. Van Blerkom J. Occurrence and developmental consequences of aberrant cellular organization in meiotically mature human oocytes after



HPE_Chapter04.qxp

7/13/2007

4:45 PM

Page 40


10.

11.

12.

13. 14.

15. 16.

17.

18.

19. 20. 21. 22.

23.

24. 25. 26.

27.

28.



exogenous ovarian hyperstimulation. J Electron Microsc Techn 1990; 16: 324–46. Balaban B, Urman B, Isklar A et al. The effects of pronuclear morphology on embryo quality parameters and blastocyst transfer outcome. Hum Reprod 2001; 16: 2357–61. Rienzi L, Ubaldi F, Iacobelli M et al. Day 3 embryo transfer with combined evaluation at the pronuclear and cleavage stages compares favourably with day 5 blastocyst transfer. Hum Reprod 2002; 17: 1852–5. Zollner U, Zollner KP, Hartl G et al. The use of a detailed zygote score after IVF/ICSI to obtain good quality blastocysts: the German experience. Hum Reprod 2002; 17: 1327–33. Scott L. Pronuclear score as a predictor of embryo development. Reprod BioMed Online 2003; 6: 201–14. Payne D, Flaherty SP, Barry MF et al. Preliminary observations on polar body extrusion and pronuclear formation in human oocytes using time-lapse video cinematography. Hum Reprod 1997; 12: 532–41. Scott L, Smith S. The successful use of pronuclear embryo transfers the day following oocyte retrieval. Hum Reprod 1998; 13: 1003–13. Ludwig M, Schopper B, Al-Hasani S et al. Clinical use of a pronuclear stage score following intracytoplasmic sperm injection: impact on pregnancy rates under the conditions of the German embryo protection law. Hum Reprod 2000; 15: 325–9. Ludwig M, Schopper B, Katalinic A et al. Experience with the elective transfer of two embryos under the conditions of the German embryo protection law: results of a retrospective data analysis of 2573 transfer cycles. Hum Reprod 2000; 15: 319–24. Tesarik J, Junca AM, Hazout A et al. Embryos with high implantation potential after intracytoplasmic sperm injection can be recognized by a simple, non-invasive examination of pronuclear morphology. Hum Reprod 2000; 15: 1396–9. Wittemer C, Bettahar-Lebugle K, Ohl J et al. Zygote evaluation: an efficient tool for embryo selection. Hum Reprod 2000; 15: 2591–7. Scott L. The biological basis of non-invasive strategies for selection of human oocytes and embryos. Hum Reprod Update 2003; 9: 237–49. Sadowy S, Tomkin G, Munne S et al. Impaired development of zygotes with uneven pronuclear size. Zygote 1998; 6: 137–41. Schatten G. The centrosome and its mode of inheritance: the reduction of the centrosome during gametogenesis and its restoration during fertilization. Dev Biol 1994; 165: 299–335. Asch R, Simerly C, Ord T et al. The stages at which human fertilization arrests: microtubule and chromosomal configurations in inseminated oocytes which failed to complete fertilization and development in humans. Hum Reprod 1995; 10: 1897–906. Tesarik J, Kopecny V. Development of human male pronucleus: ultrastructure and timing. Gamete Res 1989; 24: 135–49. Edwards RG, Beard HK. Oocyte polarity and cell determination in early mammalian embryos. Mol Hum Reprod 1997; 3: 863–905. Van Blerkom J, Davis P, Alexander S. Differential mitochondrial distribution in human pronuclear embryos leads to disproportionate inheritance between blastomeres: relationship to microtubular organization, ATP content and competence. Hum Reprod 2000; 15: 2621–33. Van Blerkom J, Davis P, Mathwig V et al. Domains of high-polarized and low-polarized mitochondria may occur in mouse and human oocytes and early embryos. Hum Reprod 2002; 17: 393–406. Ebner T, Moser M, Sommergruber M et al. Presence, but not type or degree of extension, of a cytoplasmic halo has a significant influence on preimplantation development and implantation behaviour. Hum Reprod 2003; 18: 2406–12.

29. Bavister BD and Squirrell JM. Mitochondrial distribution and function in oocytes and early embryos. Hum Reprod 2000; 15: 189–98. 30. Shoukir Y, Campana A, Farley T et al. Early cleavage of in-vitro fertilized human embryos to the 2-cell stage: a novel indicator of embryo quality and viability. Hum Reprod 1997; 12: 1531–6. 31. Sakkas D, Shoukir Y, Chardonnens D et al. Early cleavage of human embryos to the two-cell stage after intracytoplasmic sperm injection as an indicator of embryo viability. Hum Reprod 1998; 13: 182–7. 32. Grisart B, Massip A, Dessy F. Cinematographic analysis of bovine embryo development in serum-free oviduct-conditioned medium. J Reprod Fertil 1994; 101: 257–64. 33. Tesarik J, Greco E. The probability of abnormal preimplantation development can be predicted by a single static observation on pronuclear stage morphology. Hum Reprod 1999; 14: 1318–23. 34. Senn A, Urner F, Chanson A et al. Morphological scoring of human pronuclear zygotes for prediction of pregnancy outcome. Hum Reprod 2006; 21: 234–9. 35. Kahraman S, Kumtepe Y, Sertyel S et al. Pronuclear morphology scoring and chromosomal status of embryos in severe male infertility. Hum Reprod 2002; 17: 3193–200. 36. Gianaroli L, Magli MC, Ferraretti AP et al. Pronuclear morphology and chromosomal abnormalities as scoring criteria for embryo selection. Fertil Steril 2003; 80: 341–9. 37. Scott L. Embryological strategies for overcoming recurrent assisted reproductive technology treatment failure. Hum Fertil 2002; 5: 206–14. 38. James AN, Hennessy S, Reggio B et al. The limited importance of pronuclear scoring of human zygotes. Hum Reprod 2006; 21: 1599–604. 39. Jaroudi K, Al-Hassan S, Sieck U et al. Zygote transfer on day 1 versus cleavage stage embryo transfer on day 3: a prospective randomized trial. Hum Reprod 2004; 19: 645–8. 40. Salumets A, Hyden-Granskog C, Suikkari AM et al. The predictive value of pronuclear morphology of zygotes in the assessment of human embryo quality. Hum Reprod 2001; 16: 2177–81. 41. Payne JF, Raburn DJ, Couchman GM et al. Relationship between preembryo pronuclear morphology (zygote score) and standard day 2 or 3 embryo morphology with regard to assisted reproductive technique outcomes. Fertil Steril 2005; 84: 900–9. 42. Kattera S, Chen C. Developmental potential of human pronuclear zygotes in relation to their pronuclear orientation. Hum Reprod 2004; 19: 294–9. 43. Balaban B, Yakin K, Urman B et al. Pronuclear morphology predicts embryo development and chromosome constitution. Reprod BioMed Online 2004; 8: 695–700. 44. Gamiz P, Rubio C, de los Santos MJ et al. The effect of pronuclear morphology on early development and chromosomal abnormalities in cleavage-stage embryos. Hum Reprod 2003; 18: 2413–9. 45. Demirel LC, Evirgen O, Aydos K et al. The impact of the source of spermatozoa used for ICSI on pronuclear morphology. Hum Reprod 2001; 16: 2327–32. 46. Lan K-C, Huang F-J, Lin Y-C et al. The predicitive value of using a combined Z-score and day 3 embryo morphology score in the assessment of embryo survival on day 5. Hum Reprod 2003; 18: 1299–306. 47. Nagy ZP, Dozortsev D, Diamond M et al. Pronuclear morphology evaluation with subsequent evaluation of embryo morphology significantly increases implantation rates. Fertil Steril 2003; 80: 67–74.

HPE_Chapter05.qxp

7/13/2007

4:45 PM

Page 41

5. Multinucleation and mosaicism in the human preimplantation embryo Renee Walmsley

INTRODUCTION

Human embryos that are cultured in vitro sometimes have two or more nuclei visible within a single blastomere, even after apparently normal fertilization. This phenomenon, referred to as multinucleation, occurs after the completion of meiosis, and can be seen during interphase throughout the stages of in vitro culture. Multinucleated blastomeres (MNB) may be observed most readily on day 2 and 3 of development, but on day 3 definitive identification may be compromised by overlap between cells, compaction and/or fragmentation. Although multinucleation continues to occur through blastocyst development, it is very difficult to observe microscopically after day 3.1 A high incidence of multinucleation is associated with other types of embryo dysmorphism, including fragmentation, impaired embryo development and arrest,2–5 but it does not always correlate with poor embryo morphology.2,6 Consequently, care should be exercised when deciding the fate of multinucleated embryos (MNE) that are otherwise apparently of good quality. MNE do have the potential to implant, but their implantation and subsequent pregnancy rates are lower than those of non-MNE.4,7–10 Several studies have confirmed the birth of healthy babies that originated from MNE,4,8,10 and therefore, if no other embryos are available for transfer, it is preferable to transfer these embryos unless other developmental information such as the presence of chromosomal anomalies is known. The use of fluorescence in situ hybridization (FISH) has allowed a greater understanding of the developmental potential of MNE,2,11,12 and several studies have demonstrated that multinucleation is associated with higher rates of chromosomal

abnormalities.13–15 At our center, the Institute for Reproductive Medicine and Science at Saint Barnabas (IRMS), a large retrospective study using FISH to analyze 498 non-arrested MNE demonstrated abnormal chromosome compositions in 71.9% of MNE. It is important to bear in mind that the accuracy of FISH is limited by the number of chromosomes that can be analyzed for a given cell, and therefore rates of abnormality for MNE could potentially be higher if the entire complement of chromosomes were to be assessed. TYPES OF MULTINUCLEATION

Multinucleation can be observed in one or more of the blastomeres of two- to eight-celled embryos (Figure 5.1). Blastomeres can be binucleated, with two nuclei (Figure 5.1B and 5.1E), or micronucleated (Figure 5.1C, 5.1D and 5.1F) with ⱖ3 nuclei in a single cell. The impact of specific types of multinucleation on clinical outcomes has recently been studied and is discussed in detail later in this chapter.16 FREQUENCY OF MULTINUCLEATION

The frequency of multinucleation has been analyzed in several comprehensive studies, where MNE were observed in 14–79% of all patient cycles, with 15–33% of all embryos multinucleated.4,8–10,13,16 The differences found between studies may be due to variation in hormonal stimulation protocols, as well as to diverse laboratory culture conditions.17 At IRMS, 57 015 monospermic non-arrested embryos from 6012 standard IVF/ICSI patient cycles were assessed, and at least one MNE in the cohort of embryos was found in 63.6% of these cycles. In agreement with previous studies,10 a higher rate of 

HPE_Chapter05.qxp

7/13/2007

4:45 PM

Page 42


A

B

C

D

E

F

Figure 5.1 Types of multinucleation commonly observed in human embryos cultured in vitro: (A) day 2 non-MNE; (B) day 2 MNE, one binucleate blastomere (1w2); (C) day 2 MNE, one micronucleate blastomere (1w3); (D) day 2 MNE, two blastomeres with multinucleation (2wⱖ2); (E) day 3 MNE, one binucleated blastomere (1w2); (F) day 3 MNE, one micronucleated blastomere (1w3).

MNE was observed on day 2 than on day 3: multinucleation first presented on day 2 of development in 12.6% of embryos, whereas only 5.0% of embryos demonstrated multinucleation for the first time on day 3 of development (Table 5.1). By 72 hours of culture, 17.5% of embryos showed signs of multinucleation. Maternal age has been shown to correlate with the frequency of multinucleation, although the data are conflicting: younger patients had slightly higher rates of multinucleation than older patients,4 and a slight correlation was also seen at IRMS, specifically for patients ⱕ30 years old (Table 5.1). In contrast, other studies found no correlation between MNE and maternal age.8,10 The role of maternal age with respect to the frequency of multinucleation is therefore still unclear, although these contradicting data sets may simply be related to differences in stimulation protocols for the centers studied. Those protocols which ‘push’ younger patients into multifollicular 

development with high doses of gonadotropins equivalent to those administered to older patients may have a possible correlation related to increased estrogen levels in younger patients. Culture conditions within the laboratory may influence the incidence of multinucleation. For example, embryos from IRMS patients that were cultured in human tubal fluid (HTF) supplemented with maternal serum had a significantly higher rate of multinucleation than those cultured in HTF supplemented with human serum albumin (HSA) (Table 5.2). HSA was used as a protein supplement during culture either for poor prognosis patients (as an effort to change culture conditions in the hope of encouraging better overall embryo development), or when maternal serum was not available. Interestingly, patients who exhibited multinucleation in their first cycle showed lower rates of multinucleation in a subsequent second cycle (Table 5.3). This finding supports the suggestion that multinucleation

HPE_Chapter05.qxp

7/13/2007

4:45 PM

Page 43

MULTINUCLEATION AND MOSAICISM IN THE HUMAN PREIMPLANTATION EMBRYO

Table 5.1 The incidence of multinucleation in relation to age Age group

Cycles (n)

Total 2PN

Total day 2 MNE

Total day 3 MNE

Total MNE

All ⱕ30 years ⱖ31 years *p ⫽ 0.022

6012 719 5293

57 015 7912 49 103

7178 (12.6%) 1068 (13.4%) 6110 (12.4%)

2901 (5.0%) 403 (5.0%) 2498 (5.0%)

10 079 (17.6%) 1471 (18.5%) 8608 (17.5%)

PN, pronuclei; MNE, multinucleated embryos.

Table 5.2 Incidence of multinucleation in relation to culture media supplementation Protein HSA Maternal serum *p⬍ 0.001

Cycles (n)

Total 2PN

Total day 2 MNE

Total day 3 MNE

Total MNE

133 2350

1069 20 884

102 (9.5%) 3066 (14.6%)

46 (4.3%) 1342 (6.6%)

148 (13.8%) 4408 (21.1%)

PN, pronuclei; MNE, multinucleated embryos; HSA, human serum albumin.

Table 5.3 Incidence of multinucleation in patients with repeated IVF cycles Attempt 1 2

No. of cycles

Age (years)

Total 2PN

Total day 2 MNE

Total day 3 MNE

617 617

35.3 36.3

5611 5879

1079 (19.2%) 893 (15.1%)

483 (8.6%) 344 (6.0%)

Total MNE 1562 (27.8%) 1237 (21%)

Implantation rate (%)

Pregnancy rate (%)

12.8 28.0

31.7 55.4

PN, pronuclei; MNE, multinucleated embryos.

is not exclusively a patient-specific phenomenon, but may be influenced by culture and/or stimulation protocols. The lower rates of multinucleation for repeat cycles at IRMS were encouraging, suggesting that modification of cycle protocols and/or culture conditions were effective in improving patient treatment. Indeed, pregnancy and implantation rates were collectively higher for second cycles of repeat patients where multinucleation was observed during initial cycles (Table 5.3). MECHANISMS OF MULTINUCLEATION

The mechanisms that underlie embryo multinucleation during early cleavage divisions are still relatively unclear; it may be a result of karyokinesis (specifically

DNA synthesis) continuing in the absence of cytokinesis.6,18 Cells of this type were thought to be arrested, without participating in further embryo development, and therefore multinucleated embryos were felt to be developmentally (possibly terminally) compromised by multinucleation. Because cleavage stage embryos have on average more cells on day 3 than on day 2, the transfer of multinucleated embryos on day 3 in preference to those where multinucleation was witnessed on day 2 was recommended when no other embryos were available.13,19 In contrast, higher pregnancy and implantation rates for embryos where multinucleation appeared first on day 2 rather than on day 3 were recently observed at IRMS (Figure 5.2). Furthermore, in this study, 15 patient cycles in which all embryos transferred were 

HPE_Chapter05.qxp

7/13/2007

4:45 PM

Page 44


60 non-MNE (n = 5573)

50

MNE (n = 86)

40

Day 2 MNE (n = 63) 30

Day 3 MNE (n = 23)

20 10

) y nc na

Pr eg

ta tio n

(%

(%

) en

Im pl an

ta tio n

(%

el l# C m

IMPACT OF MULTINUCLEATION ON EMBRYO MORPHOLOGY

Fr

ag

)

0

Nuclear fragmentation and dysfunctional spindle tubules can cause improper separation of chromosomes during mitotic divisions, thereby leading to multinucleation.6,22,23 The oocyte cytoskeleton can be damaged by substandard laboratory conditions and suboptimal culture conditions, such as sudden changes in temperature; this can result in improper cleavage, inducing multinucleation,20,24,25 and may also play a role in causing multinucleation during later cleavage divisions.6,23

Figure 5.2 Comparison of clinical outcomes for transfer of non-multinucleated embryos (MNE) exclusively, all MNE, day 2 MNE and day 3 MNE exclusively. Cell#, the average number of cells in the embryo.

MNE exhibiting multinucleation in all cells on day 2 (i.e. every blastomere in all transferred embryos displayed multinucleation on day 2) resulted in four pregnancies. In view of this, multinucleated cells must have the potential to give rise to normal diploid daughter cells at some stage in their development. A recent and significant study using time-lapse photography showed that multinucleated blastomeres (MNB) undergo normal cell division, providing visual evidence that MNB are capable of normal mitotic cell division, and are not necessarily arrested.16 The mechanisms that are involved in cleavage stage multinucleation may occur even prior to oocyte retrieval or ovulation, and may be related to the response to stimulation. Specifically, follicular underoxygenation has been correlated with multinucleation seen at the two-cell stage of development.20 This finding supports the evidence for higher rates of multinucleation in patients who responded aggressively to hormonal stimulation during their IVF cycles,4,10,16 a trend that was also confirmed in our center (Table 5.4). Another interesting study demonstrated that oocytes matured in vitro (IVM) frequently presented with multinucleation during later development,21 suggesting that incomplete nuclear and cytoplasmic remodeling in an oocyte may result in higher incidences of multinucleation.16 

An embryo’s potential to establish a viable pregnancy can be diagnosed to a certain degree by assessing gross morphology and identifying anomalies present within the embryo.26–29 Total blastomere count and phenotypic dysmorphisms, including degree and type of fragmentation,30 appearance and severity of multinucleation, blastomere organization, contracted or granular cytoplasm, vacuoles, dense bodies and abnormal zona pellucida are recorded during morphological assessments on day 2 and 3, and the majority of studies show a clear tendency for MNE to be compromised morphologically. The presence of nucleoli in the nucleus distinguishes multinucleation from vacuoles. On average, MNE have lower cell counts, higher rates of fragmentation, uneven blastomere sizes and division, and lower rates of blastocyst formation.1,4,5,8,29,31 Micronucleated embryos are more frequently associated with poor pronuclear morphology than binucleated embryos, and they have lower rates of blastocyst formation.16 In support of this finding, IRMS observed higher implantation and pregnancy rates from the exclusive and mixed transfer of binucleated embryos than the exclusive and mixed transfer of micronucleated embryos (Table 5.5). CHROMOSOMAL STATUS OF MULTINUCLEATED EMBRYOS

Preimplantation genetic diagnosis is an important tool for analyzing the chromosomal status of embryos in vitro. When applied to specific types of

HPE_Chapter05.qxp

7/13/2007

4:45 PM

Page 45


Table 5.4 Incidence of multinucleation in relation to length of stimulation Days on GND 6–7 8–11 *p ⬍ 0.001

No. of cycles

Total 2PN

Total day 2 MNE

Total day 3 MNE

Total MNE

775 4811

7647 46 051

1050 (13.7%) 5721 (12.4%)

457 (5.9%) 2261 (4.9%)

1507 (19.7%) 7982 (17.3%)

GND, gonadotropin; PN, pronuclei; MNE, multinucleated embryos.

Table 5.5 Comparison of the severity of multinucleated embryos (MNE) in exclusive transfers after transfer of 2 embryos only MNB severity None Any MNE Any MNE 1w2 1 w2 1wⱖ3 1wⱖ3 2wⱖ2 2wⱖ2

MNE repl (%) 0 50 100 50 100 50 100 50 100

Age Total embryo Procedures (years) repl 1023 163 36 87 11 23 11 15 14

34.3 35.7 36.0 35.4 36.2 36.1 35.7 36.1 35.8

806 326 72 174 22 46 22 30 28

Total FHB

577 75 9 44 4 10 1 4 4

Procedures FHB ⬎0

Embryos repl

Implantation (%)

Pregnancy (%)

541 59 8 36 4 7 0 3 4

2 2 2 2 2 2 2 2 2

35.3 23.0 12.5 25.2 18.1 21.7 4.5 13.3 14.2

52.8 36.2 22.2 41.3 36.3 30.4 9.1 20.0 28.5

MNB, multinucleated blastomere; FHB, fetal heart beat; 1w2, one binucleated blastomere; 1wⱖ3, one micronucleated blastomere; 2wⱖ2, two blastomeres with multinucleation; Repl, replaced.

dysmorphic embryos, FISH analysis revealed higher rates of chromosomal abnormalities in MNE than in non-multinucleated embryos (non-MNE). Specifically, extensive mosaicism and/or polyploidy was observed in 44–75% of MNE embryos.13,14,32 Mosaicism in an embryo can only be accurately detected when at least two cells are analyzed;19 when a single cell is analyzed, there is a risk that a mosaic embryo may be misdiagnosed as aneuploidy.33 Mosaicism, haploidy and polyploidy occur during mitotic cell division after fertilization, and although rates of aneuploidy in cleavage stage embryos are known to increase with maternal age, this is not the case for embryo mosaicism.3,34 Polyploidy in monospermic MNE probably results from continued DNA synthesis (and possibly karyokinesis) without cytokinesis.35 In the studies carried out in our center, the embryos were analyzed on day 3, 4 or 5, and some may have been continuing DNA replication in the absence of cleavage, leading to polyploidy.22,36

We investigated correlations between chromosome anomalies and the incidence and severity of multinucleation by fixing and analyzing disaggregated blastomeres of MNE according to previously published protocols.1,37,38 Individual blastomeres from multinucleated embryos were analyzed by FISH, simultaneously using probes specific for chromosomes 13, 16, 18, 21 and 22. In the majority of cases, chromosomes X, Y, 15, and 17 were also analyzed with additional probes. The chromosomal status of all MNE and non-MNE embryos was analyzed and recorded in the EggCyte™ database, and the results were separated into groups according to the following criteria: aneuploid (ANE), aneuploid/mosaic (ANE-MOS), aneuploid/polyploid (ANE-POL), haploid (HAP), normal (NORM), polyploid (POL), high rate of mosaicism (MOS-HI), and low rate of mosaicism (MOS-LO). In order to compensate for the effect of age on FISH results, wholly aneuploid embryos were then discounted. 

HPE_Chapter05.qxp

7/13/2007

4:45 PM

Page 46


All calculations were re-analyzed for chromosome abnormalities in multinucleated embryos, excluding aneuploidy and the influence of maternal age. However, if an embryo analyzed as aneuploid also possessed another chromosome abnormality (i.e. ANE-MOS or ANE-POL), these were included in the calculations. Rates of mosaicism were classified according to parameters previously published.19 Embryos from translocation patients were not included in this study. CHROMOSOMAL STATUS OF NON-MNE

As a control reference, a group of supernumerary embryos that had not displayed multinucleation on

day 2 or day 3 (non-MNE) were analyzed. These embryos were either previously rejected for transfer or cryopreservation, or had previously been diagnosed as abnormal during preimplantation genetic diagnosis (PGD) cycles. Excluding aneuploidy, 38.5% of all non-MNE were chromosomally normal (Table 5.6), and 27% were MOS-HI (majority of the cells of the embryo were abnormal). The total percentage of non-MNE exhibiting mosaicism was 46.3%, which was the lowest incidence in all of the types of embryo analyzed in this study (Table 5.7). The low percentage of normal embryos observed in non-MNE (38.5% excluding aneuploidy) was not surprising, considering their source as detailed above. Furthermore, the number of chromosomal

Table 5.6 Incidence (%) and severity of multinucleation in relation to chromosomal abnormalities (excluding aneuploidy) MNB type 2 cell day 2 MNE 2wⱖ2 All non-MNE 2 cell, day 2 MNE Day 2 MNE All MNE 2 cell, day 2 MNE 1w2 All MNE 1w3 All MNE 1w2 Day 3 MNE

Total no. embryos 89 1630 186 356 498 45 72 118 215

Normal Abnormal 44 39 36 32 28 27 26 25 21

56 61 64 68 72 73 74 75 79

Total MOS

Total POL

Total HAP

46 46 54 49 50 64 57 59 55

5 11 5 14 18 7 15 15 22

6 4 4 6 4 2 2 1 2

ANE-MOS MOS-HI 8 11 8 6 7 7 10 8 9

34 27 42 39 39 56 44 46 41

MOS-LO ANE-POL POL 4 9 4 3 4 2 3 5 5

1 1 1 1 1 0 0 2 2

3 10 5 14 16 7 15 14 20

MNB, multinucleated blastomere; MOS, mosaic; POL, polyploidy; HAP, haploid; HI, high rate; LO, low rate; ANE, aneuploid.

Table 5.7 Incidence (%) and severity of multinucleation in relation to chromosomal abnormalities (excluding aneuploidy), sorted by highest rate of total mosaicism (MOS) MNB type 2 cell, day 2 MNE 1w2 All MNE 1w2 All MNE 1w3 Day 3 MNE 2 cell, day 2 MNE All MNE Day 2 MNE 2 cell day 2 MNE 2wⱖ2 All non-MNE

Total no. embryos

Normal

Abnormal

Total MOS

Total POL

Total HAP

45 118 72 215 186 498 356 89 1630

27 25 26 21 36 28 32 44 39

73 75 74 79 64 72 68 56 61

64 59 57 55 54 50 49 46 46

7 15 15 22 5 18 14 5 11

2 1 2 2 4 4 6 6 4

ANE-MOS MOS-HI MOS-LO 7 8 10 9 8 7 6 8 11

56 46 44 41 42 39 39 34 27

2 5 3 5 4 4 3 4 9




ANE-POL POL 0 2 0 2 1 1 1 1 1

7 14 15 20 5 16 14 3 10

HPE_Chapter05.qxp

7/13/2007

4:45 PM

Page 47


anomalies for re-analyzed embryos of PGD cycles is expected to be artificially high, since normally dividing euploid embryos are transferred or cryopreserved and not subjected to further analysis. CHROMOSOMAL STATUS OF ALL MNE (REGARDLESS OF SEVERITY/DAY OF MULTINUCLEATION)

The chromosomal status of embryos that had displayed MNB on day 2 or day 3 was determined by fixation and FISH on days 3, 4 or 5. Excluding aneuploidy, only 28.1% of MNE were chromosomally normal, compared to 38.5% for non-MNE (Table 5.6; pⱕ 0.001). The total percentage of MNE exhibiting mosaicism was 50.1%, which was not significantly different from the 46.3% incidence of mosaicism observed for non-MNE. However, a significant difference was seen between the number of MOS-HI embryos found in MNE and non-MNE (Table 5.7): 38.9% MOS-HI for MNE vs. 27% for non-MNE (p ⱕ 0.001). Notably higher rates (p ⱕ 0.001) of total polyploidy were observed for MNE (17.5%) when compared to rates of total polyploidy for non-MNE (11%; Table 5.8). CHROMOSOMAL STATUS OF DAY 2 VERSUS DAY 3 MNE (REGARDLESS OF SEVERITY OF MULTINUCLEATION)

Day 2 and day 3 MNE were assessed for chromosomal status. Excluding aneuploidy, only 20.5% of day 3

MNE were normal compared to 31.5% of day 2 MNE (Table 5.6; p ⱕ 0.002). Virtually the same rate of MOS-HI (40.2%) was observed for day 3 MNE as for day 2 MNE (39.0%; Table 5.7). The rate of total mosaicism did not differ significantly between day 3 MNE (54.9%) and day 2 MNE (50.1%). There was an interesting disparity in rates of total polyploidy between day 2 and day 3 embryos, with a significantly higher rate (p ⱕ 0.01) of total polyploidy (22.3%) observed for day 3 MNE compared to the rate for day 2 MNE (14.4%; Table 5.8). CHROMOSOMAL STATUS OF MNE WITH ONE BINUCLEATED VERSUS ONE MICRONUCLEATED BLASTOMERE (REGARDLESS OF THE DAY ON WHICH MULTINUCLEATION WAS FIRST OBSERVED)

MNE with only one binucleated blastomere (i.e. 1w2), or only one micronucleated blastomere (i.e. 1ⱖ3), irrespective of the day on which these were observed, were assessed for their chromosomal status. All of the other cells in these embryos had either a single nucleus, or no nucleus visible. Excluding aneuploidy, the total number of embryos that were chromosomally normal did not differ significantly between binucleated embryos (25.4%) and micronucleated embryos (26.3%; Table 5.6). The percentage of binucleated and micronucleated embryos exhibiting extensive mosaicism was 45.7% and 44.4%, respectively, and the percentage of embryos that were completely mosaic were comparable, 58.4% and 56.9%

Table 5.8 Incidence (%) and severity of multinucleation in relation to chromosomal abnormalities (excluding aneuploidy), sorted by highest rate of total polyploidy MNB type Day 3 MNE All MNE All MNE 1w2 All MNE 1w3 Day 2 MNE All non-MNE 2 cell, day 2 MNE 1w2 2 cell, day 2 MNE 2 cell, day 2 MNE 2wⱖ2

Total no. embryos Normal Abnormal 215 498 118 72 356 1630 45 186 89

21 28 25 26 32 39 27 36 44

79 72 75 74 68 61 73 64 56

Total MOS

Total POL

Total HAP

55 50 59 57 49 46 64 54 46

22 18 15 15 14 11 7 5 5

2 4 1 2 6 4 2 4 6

ANE-MOS MOS-HI MOS-LO ANE-POL POL 9 7 8 10 6 11 7 8 8

41 39 46 44 39 27 56 42 34

5 4 5 3 3 9 2 4 4

2 1 2 0 1 1 0 1 1

20 16 14 15 14 10 7 5 3




HPE_Chapter05.qxp

7/13/2007

4:45 PM

Page 48


(Table 5.7). Approximately 15% of binucleate or micronucleated embryos were polyploid (Table 5.8). CHROMOSOMAL STATUS OF TWO-CELL DAY 2 MNE WITH BOTH CELLS MULTINUCLEATED

Non-arrested two-cell embryos with both cells multinucleated (i.e. 2wⱖ2) on day 2 were assessed for chromosomal status after further cleavage on days 3, 4 or 5 (Figure 5.1D); when aneuploidy was excluded, 43.8% of the 89 embryos analyzed were normal (Table 5.6). Interestingly, this category of embryos had the highest proportion of chromosomally normal embryos, comparable to the 38.5% of chromosomally normal embryos observed for nonMNE; however, it is important to note that these results may be due to the relatively small dataset of this particular subgroup of MNE. Results further demonstrated that two-cell day 2 MNE with both cells multinucleated had the lowest proportion of MOS-HI (33.7%), total MOS (46.1%) and total-POL (4.5%) embryos among all types of multinucleation analyzed (Tables 5.7 and 5.8). These embryos had the same rates of Tot-MOS (46.1%) as non-MNE and were second to non-MNE in terms of lowest rates of extensive mosaicism (MOS-HI) (Table 5.7). These findings support the conclusion that MNB, despite showing multiple nuclear membranes, may still be capable of undergoing normal spindle formation and cleavage divisions. Unambiguous evidence of this phenomenon is the birth of a healthy baby derived from a completely multinucleated embryo where all cells were multinucleated on day 2.

the majority of these transfers were carried out in younger patients, in whom only two embryos are transferred as a routine, in order to reduce the likelihood of multiple pregnancy. In contrast, transfer of MNE exclusively occurs only in cycles with few embryos or poor overall embryo development. Embryos were selected for transfer and/or cryopreservation based on their assessment as having the best prognosis for establishing viable pregnancies.3,5,26,39,40 Assisted hatching was performed as indicated, based on criteria published by Cohen et al.41 Embryo transfers were usually performed on day 3 of development (76–79 hours post oocyte retrieval). Clinical pregnancy was defined as fetal heart activity in the presence of an appropriate rise in serum ␤hCG levels. A positive rise in serum ␤hCG level with a gestational sac, but where fetal heart was absent, was classified as a blighted ovum and not considered as a clinical pregnancy for this large study of MNE. Exclusive transfers of embryos where multinucleation was observed first on day 2 had much higher implantation rates than exclusive transfers of embryos where multinucleation was first observed on day 3 (Figures 5.2 and 5.3). This would indicate that multinucleation that occurs initially on day 2 is less detrimental to embryos than when it is initiated on

25

Non-MNE (n = 290) All MNE (n = 60)

20

Day 2 MNE (n = 39) 15

IMPACT OF MULTINUCLEATION ON

Day 3 MNE (n = 21)

10

CLINICAL OUTCOMES 5



(% )

) eg n

an

cy

tio n Pr

ta an pl Im

gm

en t

at io

C

n

el l

(%

(%

)

#

0

Fr a

Comprehensive studies have reported inferior clinical outcomes after transfer of MNE compared to transfer of non-MNE.4,8–10 Implantation and pregnancy rates at IRMS after transfer of MNE exclusively were approximately half of the rates after transfer of nonMNE exclusively (Figure 5.2). However, in analyzing cycle outcomes when only two embryos were transferred, the higher implantation and pregnancy rates for controls are to some extent exaggerated, because

Figure 5.3 Comparison of clinical outcomes for transfer of non-MNE exclusively, all MNE, day 2 MNE and day 3 MNE exclusively, limited to transfer of one embryo only.

HPE_Chapter05.qxp

7/13/2007

4:45 PM

Page 49


day 3. Indeed, when chromosome constitutions of day 2 and day 3 MNE were investigated and compared, there were higher percentages of normal embryos from day 2 (32%) than day 3 (21%), which supports the above clinical findings (i.e. higher implantation and pregnancy rates for exclusive transfers of embryos whose multinucleation was first observed on day 2 when compared to exclusive transfers of embryos whose multinucleation was first observed on day 3). Since a greater percentage of the embryo is affected by multinucleated cells on day 2 than on day 3, these observations are unexpected, and suggest the possibility that the process and consequence of creating multiple nuclei within a blastomere may differ according to the developmental stage of the embryo. Implantation and pregnancy rates (18% and 36%, respectively) for exclusive transfers of binucleated embryos (regardless of the day on which multinucleation was first observed) were the highest for any type of multinucleation patterns analyzed (Table 5.5). It was surprising, therefore, to find that only 25% of binucleated MNE were normal after chromosome analysis. Although they had better positive clinical outcomes, they also had one of the highest rates of chromosome abnormality (Table 5.6). In contrast, the embryos considered to contain the most severe type of multinucleation (i.e. both cells multinucleated on day 2) had the highest percentage of chromosomally normal embryos (44%); however, this may again be simply a function of different sample sizes when comparing the groups.

birth. Multinucleation is most easily observed on day 2, and can also be seen, to a lesser extent, on day 3. Embryos that display multinucleation first on day 2 have a better prognosis for implantation than do those that develop this feature on day 3. MNE were noted to have higher rates of extensive mosaicism, haploidy and polyploidy than non-MNE, with mosaicism as the most common chromosomal abnormality observed. Rates of mosaicism within the MNE themselves appear to depend on the type and severity of multinucleation. After transfer of MNE exclusively, those with binucleated cells were associated with a better clinical outcome, although these embryos overall had the highest rates of chromosome abnormalities. In contrast, embryos considered to contain the most severe type of multinucleation (i.e. all cells of the embryo were multinucleated on day 2) had the highest percentage of normal embryos (44%) when FISH was performed at later cleavage stages, which indicates that multinucleated cells may undergo some mechanism which lends itself to normal spindle function and cleavage division (despite multiple nuclear membranes) and eventually give rise to diploid daughter cells. This comprehensive review of multinucleated cleavage stage human embryos has yielded some unexpected results, which give us a better understanding regarding the incidence and effects of multinucleation. This knowledge will hopefully result in more effective embryo selection for uterine transfer and/or cryopreservation during IVF cycles. ACKNOWLEDGMENTS

CONCLUSION

Embryos with multinucleated blastomeres are commonly observed during the development of preimplantation embryos cultured in vitro. Factors that have been found to correlate with the prevalence of multinucleation include an aggressive and/or high response to ovarian stimulation, and features of in vitro culture systems. Although the implantation and pregnancy rates for these dysmorphic embryos are lower than those for non-MNE, nonetheless they still have the potential to lead to a healthy live

Many thanks to Giles Tomkin for his aid with Eggcyte™ database queries as well as Santiago Munne, Jacques Cohen, John Garrisi, and Mina Alikani for their help in formulating and interpreting the data. Thanks to the embryologists at IRMS: Toni Congedo, Kathleen Ferry, Sheri Klein, Elena Kissin, Adrienne Reing, and Kenny Smalls. REFERENCES 1. Sandalinas M, Sadowy S et al. Developmental ability of chromosomally abnormal human embryos to develop to the blastocyst stage. Hum Reprod 2001; 16: 1954–8.



HPE_Chapter05.qxp

7/13/2007

4:45 PM

Page 50


2. Munne S, Cohen J. Unsuitability of multinucleated human blastomeres for preimplantation genetic diagnosis. Hum Reprod 1993; 8: 1120–25. 3. Munne S, Alikani M et al. Embryo morphology, developmental rates and maternal age are correlated with chromosome abnormalities. Fertil Steril 1995; 64: 382–91. 4. Jackson K, Ginsburg E et al. Multinucleation in normally fertilized embryos is associated with an accelerated ovulation induction response and lower implantation and pregnancy rates in in vitro fertilization-embryo transfer cycles. Fertil Steril 1998; 70: 60–6. 5. Alikani M, Calderon G et al. Cleavage anomalies in early human embryos and survival after prolonged culture in vitro. Hum Reprod 2000; 15: 2634-43. 6. Hardy K, Winston RML, Handyside AH. Binucleate blastomeres in preimplantation human embryos in vitro: failure of cytokinesis during early cleavage. J Reprod Fertil 1993; 98: 549–59. 7. Mohr LR, Trounson AO et al. Evaluation of normal and abnormal human embryo development during procedures in vitro 1983. In: Beier HM, Lindner HR, eds. Fertilization of the Human Egg In-Vitro. Heidelberg: Springer Verlag 1983; 211–22. 8. Balakier H and Cadesky K. The frequency and developmental capability of human embryos containing multinucleated blastomeres. Hum Reprod 1997; 12: 800–4. 9. Pelinck MJ, De Vos M et al. Embryos cultured in vitro with multinucleated blastomeres have poor implantation potential in human in-vitro fertilization and intracytoplasmic sperm injection. Hum Reprod 1998; 13: 960–3. 10. Van Royen E, Mangelschots K, Vercruyssen M et al. Multinucleation in cleavage stage embryos. Hum Reprod 2003; 18: 1062–9. 11. Handyside AH, Pattinson JK et al. Biopsy of human pre-implantation embryos and sexing by DNA amplification. Lancet 1989; 1: 347-9. 12. Grifo JA, Boyle A et al. Preimplantation biopsy and analysis of blastomeres by in situ hybridization. Am J Obstet Gynecol 1990; 163: 2013–19. 13. Kligman I, Benadiva C et al. The presence of multinucleated blastomeres in human embryos is correlated with chromosomal abnormalities. Hum Reprod 1996; 11: 1492–8. 14. Staessen C, Van Steirteghem AC et al. The genetic constitution of multinucleated blastomeres and their derivative daughter blastomeres. Hum Reprod 1998; 13: 1625–31. 15. Magli MC, Gianaroli L, Ferrareti AP. Chromosomal abnormalities in embryos. Mol Cell Endocrinol 2001; 22(Suppl 1): S29–34. 16. Meriano J, Clark C, Cadesky K et al. Binucleated and micronucleated blastomeres in embryos derived from human assisted reproduction cycles. RBM online 2004; 9: 511–20. 17. Munne S, Magli, C et al. Treatment-related chromosome abnormalities in human embryos. Hum Reprod 1997; 12: 780–4. 18. Tesarik J, Kopecny V et al. Ultrastructural and autoradiographic observations on multinucleated blastomeres of human cleaving embryos obtained by in-vitro fertilization. Hum Reprod 1987; 2: 127–36. 19. Munne S, Cohen J. Chromosome abnormalities in human embryos. Hum Reprod Update 1998; 4: 842–55. 20. Van Blerkom J, Freytag M et al. The developmental potential of the human oocyte is related to the dissolved oxygen content of follicular fluid: association with vascular endothelial growth factor levels and perifollicular blood flow characteristics. Hum Reprod 1997; 12: 1047–55. 21. Nogueira D, Staessen C et al. Nuclear status and cytogenetics of embryos derived from in-vitro matured oocytes. Fertil and Steril (2000); 74: 295–8.



22. Winston NJ, Braude PR et al. The incidence of abnormal morphology and nucleocytoplasmic ratios in 2-, 3-, and 5-day human pre-embryos. Hum Reprod 1991; 6: 17–44. 23. Pickering SJ, Taylor A et al. An analysis of multinucleated blastomere formation in human embryos. Mol Hum Reprod 1995; 10: 1912-22. 24. Angell RR, Sumner AT, West JD et al. Post-fertilization polyploidy in human preimplantation embryos fertilized in vitro. Hum Reprod 1996; 2: 721–7. 25. Pickering SJ, Braude PR et al. Transient cooling to room temperature can cause irreversible disruption of the meiotic spindle in the human oocyte. Fertil Steril 1990; 54: 102–8. 26. Puissant F, Van Rysselberge M et al. Embryo scoring as a prognostic tool in IVF treatment. Hum Reprod 1987; 2: 705–8. 27. Giorgetti C, Terriou P et al. Embryo score to predict implantation after in vitro fertilization: based on 957 single embryo transfers. Hum Reprod 1995; 10: 2427–31. 28. Ziebe S, Petersen K et al. Embryo morphology or cleavage stage: how to select the best embryos for transfer after in vitro fertilization. Hum Reprod 1997; 12: 1545–49. 29. Alikani M, Cohen J et al. Human embryo fragmentation in-vitro and its implications for pregnancy and implantation. Fertil Steril 1999; 71:836–42. 30. Alikani M, Cohen J. Patterns of cell fragmentation in the human embryo in vitro. J Assist Reprod Genet 1995; 12(Supp 1): 28s. 31. Hardarson T, Hanson C et al. Human embryos with unevenly sized blastomeres have lower pregnancy and implantation rates: indications for aneuploidy and multinucleation. Hum Reprod 2001; 16: 313–8. 32. Laverge H, Sutter P et al. Triple colour fluorescent in-situ hybridization for chromosomes X, Y and 1 on spare human embryos. Hum Reprod 1997; 12: 809–14. 33. Almeida P, Bolton V. The relationship between chromosomal abnormality in the human preimplantation embryo and development in-vitro. Reprod Fertil Dev 1996; 8: 235–41. 34. Marquez C, Sandalinas M et al. Chromosome abnormalities in 1255 cleavage-stage human embryos. RBM Online 2000; 1: 17–27. 35. Munne S, Grifo J et al. Chromosome abnormalities in human arrested preimplantation embryos: a multiple-probe FISH study. Am J Hum Genet 1994A; 55: 150–9. 36. Artley J, Braude P, Johnson M. Gene activity and cleavage arrest in human pre-embryos. Hum Reprod 1992; 7: 1014–21. 37. Munne S, Weier U et al. Reduction in signal overlap results in increased FISH efficiency: implications for preimplantation genetic diagnosis. J Assist Reprod Genet 1996; 13: 149–56. 38. Velilla E, Escudero T et al. Blastomere fixation techniques and risk of misdiagnosis for preimplantation genetic diagnosis of aneuploidy. RBM Online 2002; 4: No 3. 39. Cohen J, Inge K et al. Video-cinematography of fresh and cryopreserved embryos: a retrospective analysis of embryonic morphology and implantation. Fertil Steril 1989; 51: 820. 40. Bolton VN, Hawes SM et al. Development of spare human preimplantation embryos in vitro: an analysis of the correlations among gross morphology, cleavage rates, and development to the blastocyst. J In Vitro Embryo Transf 1989; 7: 186. 41. Cohen J, Alikani M et al. Rescuing abnormally developing embryos by assisted hatching. In: Mori T, Aono T, Tominage T, Hiroi M, eds. Frontiers in Endocrinology, Perspectives on Assisted Reproduction. Rome: Ares Serono Symposia, 1994; 4: 536–44.

HPE_Chapter06.qxp

7/13/2007

4:46 PM

Page 51

6. The origins and consequences of fragmentation in mammalian eggs and embryos Mina Alikani

INTRODUCTION

The typical course of development following normal fertilization of the human egg in vitro begins with completion of second meiosis, extrusion of a polar body (PB2), and the emergence of one male and one female pronucleus some 9–11 hours later (day 1). Development continues with three to four mitotic divisions in the following 48 hours, producing an embryo with 2–4 (on day 2), and 8–10 (on day 3) roughly equal-sized mononucleated cells; division continues followed by compaction of the cells on day 4, and cavitation and differentiation of the embryo into two distinct cell types – inner cell mass and trophectoderm – between days 4 and 6 of development (Figure 6.1). However, in vitro fertilization (IVF) and culture of human eggs often leads to atypical development. A recent survey of the EggCyte clinical embryology database1 indicates that roughly three-quarters of over 79 000 human embryos generated in our laboratory were affected by cell fragmentation, that is, they contained non-nucleated cell fragments of varying size and number along with nucleated cells. The survey also revealed that fragmentation is often accompanied by one or more other abnormalities, including blastomere multinucleation, size discrepancy, and disorganization, as well as reduced cell–cell adhesion and abnormally thick zonae pellucidae (Figure 6.1). Fragmentation was described in human IVF embryos as early as 1970,2 but the phenomenon is neither unique to IVF, nor to human embryos. Fragments have been found in in vivo developed embryos in the human3–6 and in in vivo and in

vitro produced embryos of practically all other mammalian species studied so far.7–9 It has long been known that human embryo fragmentation is associated with decreased embryo viability10 and given its prevalence, embryo fragmentation presents a formidable challenge to the success and efficiency of human IVF. It is also clear that fragmentation hampers the efficiency of somatic cell nuclear transplantation in mammals since many eggs fragment following enucleation or after reconstitution with a donor nucleus.11–13 While many of the early studies on fragmentation in the human viewed this ‘anomaly’ in absolute terms, i.e. whether it was present or absent, more recently, detailed observations have led to a better understanding of the relationship between fragmentation, developmental competence, and viability. There is particular interest in determining the causes and mechanisms of fragmentation. Earlier studies identified a number of potentially contributing factors, including adverse follicular conditions, in vitro culture conditions, suboptimal culture medium composition, reactive oxygen species, and chromosomal abnormalities.14–18 More recently, a number of interesting observations have been made during detailed microscopic examination of embryos at 8–12 hour intervals.19 Many of these observations are relevant in the context of the origins of fragmentation. For instance, little change was noted in the volume and distribution of fragments in embryos that were mitotically inactive. This is at least suggestive that fragmentation does not occur in arrested cells. Over time, some fragments were seen to swell and 

HPE_Chapter06.qxp

7/13/2007

4:46 PM

Page 52


Day 1

Day 2

Day 3

Day 4

Day 5

pb

A

B

C

D

F

G

H

I

E

J

Figure 6.1 Human embryos showing normal (A)–(E) and abnormal (F)–(J) development on days 1 through 5 in culture. (A) At the zygote stage, pronuclei are evenly sized, the cytoplasm is slightly contracted, and two polar bodies (pb) are visible. (B) On day 2, the embryo has gone through two divisions and has four evenly sized cells, each with one nucleus. (C) On day 3 of development, the embryo is at the 8-cell stage with evenly sized, mononucleate cells and minimal fragmentation (arrowhead). (D) On day 4, the embryo has undergone full compaction, a prerequisite for normal blastulation. (E) On day 5, a blastocyst has formed with a distinct inner cell mass (arrowhead) and a cohesive trophectoderm. Embryos in (F)–(J) show various morphological abnormalities, including (F) uneven pronuclear size (pronuclei are outlined); (G) slow development and multinucleation of blastomeres (multiple nuclei are outlined), and fragmentation (arrowhead); (H) extensive fragmentation (arrowheads) and disorganization of blastomeres; (I) abnormal compaction with exclusion of cells and fragments (arrowheads); and (J) abnormal blastulation and a thick zona pellucida (arrow).

lyze, raising the possibility of a ‘toxic’ microenvironment within the confines of the zona pellucida. Many ‘fragments’ present at the 2–4-cell stage were undetectable at the 8–10-cell stage and the absence of residual cellular debris suggested resorption rather than lysis of these structures. This was reported to be accompanied by ‘turbulence’ or movement in the underlying cytoplasm, and representative of the ‘transient’ nature of fragments in some cases. Time-lapse video images19,20 showed that the initial fragmentation event occurred over a 30 minute interval in about 90% of the embryos. At the same time, some embryos with extensive fragmentation during early cleavage stages were noted to have returned to apparently normal morphology by day 3 of development. In some embryos, fragmentation was described as ‘episodic,’ occurring at



both the 2- and 4-cell stages. This would imply that fragmentation is not restricted to the first cleavage division. The observations regarding ‘disappearance’ of fragments following completion of division point to normally occurring changes in cell shape that are cell cycle specific. They highlight the importance of distinguishing between the different types of fragments and defining fragmentation in relation to the cell cycle.

DEFINING FRAGMENTATION AND DELINEATING ITS CONSEQUENCES IN HUMAN EMBRYOS

The extent to which fragmentation occurs, along with cell number, are primary criteria for selecting embryos for transfer during clinical IVF, but the

HPE_Chapter06.qxp

7/13/2007

4:46 PM

Page 53

ORIGINS AND CONSEQUENCES OF FRAGMENTATION

adoption of a universal and reproducible fragmentation classification system has been slow. In some laboratories, loosely defined morphological terms (such as minor or major fragmentation) are still being used for embryo characterization and incorporated into ‘embryo grading’ systems that fail to convey adequate detail. Other laboratories including our own use somewhat more specific, albeit still subjective, descriptions of fragmentation in their evaluation schemes. These descriptions entail a close approximation of the volume of anucleate fragments, expressed as a percentage of the total volume of the embryo, and referred to as the ‘degree’ of fragmentation. Apart from fragmentation degree, non-invasive morphological evaluation of large numbers of embryos led us to define several distinct patterns of fragmentation on day 3 of development (Figure 6.2).21 The patterns are distinguished on the basis of the size and location of fragments relative to the size and position of nucleated cells: few small fragments typically associated with only one blastomere (type I), many small and localized fragments associated with one or more cells (type II), small and scattered fragments associated with multiple cells (type III), large and scattered fragments associated with several unevenly sized cells (type IV), and fragments appearing with a characteristic granularity associated with cytoplasmic contraction in intact blastomeres (type V). Some embryos, particularly those with extensive fragmentation, do not show a specific pattern, but a non-distinct combination of patterns. These embryos were classified as having no distinct pattern (NDP) of fragmentation. Type III is the most frequently occurring pattern of fragmentation, while type V is rare (Figure 6.3). The patterns do not appear to be related to maternal age, but the degree of fragmentation, contrary to general belief, appears to decrease with increasing maternal age.6 The reason for this is not immediately clear, particularly since, according to our database, average cell number is not affected by maternal age. Our original study21 also showed that embryos with localized fragments (type II) had on average

A

A⬘

B

B⬘

C

C⬘

D

D⬘

Figure 6.2 Human embryos showing different patterns of fragmentation, before (left panel) and after (right panel) microsurgical fragment removal. (A) and (B) Type II; (C) type III; (D) type IV. Note the gross discrepancy in size of the blastomeres in (D) and (D⬘).

fewer cells on day 2 than embryos with all other patterns of fragmentation. On day 3, embryos with both types II and IV fragmentation had fewer cells than those with other patterns. The differences in cell number are both interesting and important, and are likely to be related to how the different patterns arise. In particular, in the case of type II and IV, reduced cell number suggests that these patterns may result from loss of one or more cells, and/or



HPE_Chapter06.qxp

7/13/2007

4:46 PM

Page 54


30

45 40 35

20

30 25

15

20

10

15

Percentage

Percentage

25

10

5

5

0

0 Type I

Type II Type III Type IV Type V

NDP

No fragments

Fragmentation pattern frequency Degree of fragmentation according to pattern Figure 6.3 The frequency of fragmentation patterns (bars, left Y axis) and the average fragmentation degree for each pattern (red line, right Y axis). Types III and IV are the most commonly occurring patterns.

slow division or arrest of cells following loss of large volumes of cytoplasm. A similar fragmentation classification system was developed by Antczak and Van Blerkom,22 which considered the temporal as well as spatial characteristics of fragmentation: type I fragmentation was defined as a monolayer carpet of very small fragments that coated only a modest portion of the cell surface with minor or no apparent reduction in cell size; multiple layers of fragments involving a considerable portion of the cell surface and accompanied by a significant reduction in cell size was designated as type II; type III was defined as complete fragmentation of one blastomere in an embryo showing no fragmentation on a previous day; type IV was defined as occasional fragments of variable size scattered over several blastomeres in otherwise normal appearing and developmentally progressive embryos. These patterns, in ascending order, are roughly equivalent to fragmentation types III, IV, II, and I as defined by Alikani et al.21 Both the degree and the pattern of fragmentation determine the viability of fragmented embryos following intrauterine transfer 

and their survival in extended culture to the blastocyst stage.

THE IMPACT OF FRAGMENTATION ON IMPLANTATION, PREGNANCY, AND NEONATAL OUTCOME

The relationship between fragmentation and embryo viability has been analyzed in a number of studies. With the exception of one study23 in which neither implantation nor multiple birth correlated with fragmentation, all these studies indicate that cleavage stage embryos with no fragmentation establish pregnancy more readily than those with fragmentation.21,22,24–28 Our own study21 included a large number of ‘homogeneous’ cases – those in which more than half of transferred embryos were in the same morphological category, with respect to degree and/or type of fragmentation. This was an important (and unique) aspect of the study, and its aim was to eliminate or minimize correlative uncertainties when

HPE_Chapter06.qxp

7/13/2007

4:46 PM

Page 55


multiple embryos were transferred, but without restricting the analyses to too few cases. Homogeneous transfers in this study included all single embryo transfers, two-embryo transfers in which both embryos were in the same category, threeembryo transfers in which two or three of the embryos were in the same category, etc. The majority of transfers involved two to four embryos. The study established a clear relationship between the degree of fragmentation and embryo viability. Fewer fetuses developed (85 fetuses/420 transferred embryos; 20.2%) following transfer of embryos the majority of which had ⬎15% fragmentation than when the majority had ⱕ15% fragmentation (1681/5496; 30.1%) at the time of evaluation, on the morning of day 3 of development.

We recently reanalyzed 3322 transfers (regular and oocyte recipient cycles) in which all embryos (a total of 8064 transferred embryos) were in the same category with respect to ‘fragmentation degree’ (Tables 6.1 and 6.2). We ran contingency table analyses to test whether the proportions of embryos implanted and of pregnancies achieved were the same among the five fragmentation transfer groups. The log-likelihood ␹2 values generated in these analyses were significant (p ⬍0.01) in all cases, suggesting that the frequency of implantation and pregnancy in fact differed among the embryo fragmentation categories. Because of this, we followed up by testing ten pair-wise contrasts among the five fragmentation groups and evaluated the probabilities associated with the ␹2 values from these contrasts

Table 6.1 Implantation and pregnancy outcome in homogeneous transfer groups including oocyte recipients Fragmentation degree transfer group (1) 0–5% (2) 6–15% (3) 16–25% (4) 26–35% (5) over 35%

No. of procedures

No. of pregnancies

Pregnancy ratea (%)

No. of embryos transferred

No. of FHB

Implantation rateb (%)

2295 745 172 67 43

1270 341 47 15 6

55.3 45.8 27.3 22.4 14.0

5864 1730 307 92 71

1940 503 54 18 6

33.1 29.1 17.6 19.6 8.5

FHB, fetal heart beat. aPair-wise contrasts (Bonferroni corrected) of pregnancy results in groups 1–5 show statistically significant differences between all except group 3 vs groups 4 and 5; and group 4 vs group 5. bPair-wise contrasts of implantation results in groups 1–5 show statistically significant differences in all except group 2 vs 4; group 3 vs groups 4 and 5; and group 4 vs group 5.

Table 6.2 Other characteristics of the embryos in homogeneous fragmentation degree transfer groups Fragmentation degree transfer group (1) 0–5% (2) 6–15% (3) 16–25% (4) 26–35% (5) over 35% aDifferences

Average day-3 cell numbera (all embryos) 6.7 6.0 5.3 4.9 4.3

⫾SD

Average day-3 fragmentationa (all embryos) (%)

1.2 1.3 1.5 1.3 1.3

8.4 19.6 29.0 38.9 47.8

⫾SD

Average day-3 cell numbera (transferred embryos)

6.8 9.6 11.7 11.7 14.2

7.6 7.2 6.4 6.2 5.6

⫾SD

Average day-3 fragmentationa (transferred embryos) (%)

⫾SD

1.2 1.4 1.5 1.2 1.0

2.1 12.3 22.1 33.0 45.8

2.0 2.0 2.2 2.3 7.2

among the groups are significant at the 0.05 significance level (Newman–Kuels multiple comparisons test).



HPE_Chapter06.qxp

7/13/2007

4:46 PM

Page 56


with the sequential Bonferroni procedure. The outcome of these analyses is indicated in the tables. The highest implantation and pregnancy rates were indeed obtained following transfer of embryos with 0–5% fragmentation. The embryos in this group had significantly more cells on average than the embryos in all other groups, demonstrating an association between fragmentation and total cell number (Table 6.2). The original study21 had also shown a general decline in implantation and pregnancy rates when the majority of the transferred embryos had type I, type II, type III, or type IV fragmentation, but the largest decrease in implantation rate occurred when embryos with fragmentation type IV were transferred. Embryos with type IV fragmentation also showed the highest degree of fragmentation,

on average about 25%, which is likely to have contributed to their low viability. An updated analysis of transfers in which all embryos were in the same category with respect to the pattern of fragmentation (1263 transfers and 2703 transferred embryos) (Tables 6.3 and 6.4; same statistical analyses as described above) showed that the highest implantation rate was achieved following exclusive transfer of embryos with type I fragmentation, while the poorest implantation and pregnancy outcomes were obtained with embryos with type IV fragmentation. Embryos with type II fragmentation implanted as frequently as those with type III fragmentation. On average, these two patterns showed similar degrees of fragmentation, but type II embryos had significantly fewer cells than type III embryos (5.4 ⫾ 1.5 vs 6.1 ⫾ 1.3 cells,

Table 6.3 Pregnancy and implantation outcome in homogeneous transfer groups, including oocyte recipients Fragmentation pattern transfer group (1) Type I (2) Type II (3) Type III (4) Type IV

No. of procedures

No. of pregnancies

Pregnancy ratea (%)

No. of embryos transferred

No. of FHB

Implantation rateb (%)

130 76 840 217

72 25 412 50

55.4 32.9 49.0 23.0

242 133 1957 371

101 37 613 61

41.7 27.8 31.3 16.4

FHB, fetal heart beat. aPair-wise contrasts of pregnancy results in groups 1–4 show statistically significant differences between all except group 1 vs group 3; and group 2 vs group 4. bPair-wise contrasts of implantation results in groups 1–4 show statistically significant differences in all except group 2 vs 3.

Table 6.4 Other characteristics of embryos in the homogeneous fragmentation pattern transfer groups Fragmentation pattern transfer group (1) Type I (2) Type II (3) Type III (4) Type IV

Average day-3 cell numbera (all embryos) 6.5 5.4 6.1 5.2

⫾SD

Average day-3 fragmentationa (all embryos) (%)

⫾SD

1.4 1.5 1.3 1.4

13.1 21.7 20.5 30.1

9.3 13.6 10.7 15.0

Average day-3 cell numbera (transferred embryos) ⫾SD 7.5 6.6 7.3 6.3

Average day-3 fragmentationb (transferred embryos) (%)

⫾SD

5.3 13.5 13.3 24.6

2.7 6.2 5.7 10.9

1.3 1.7 1.3 1.3

a) Differences among the groups are significant at the 0.05 significance level (Newman–Kuels multiple comparisons test). b) Difference in average day 3 fragmentation in transferred embryos in group 2 vs group 3 is not significant.



HPE_Chapter06.qxp

7/13/2007

4:46 PM

Page 57


respectively, overall; 6.6 ⫾ 1.7 vs 7.3 ⫾ 1.3 cells, respectively, in transferred embryos). Pregnancy rate, however, was significantly lower following transfer of type II embryos than type III embryos (32.9 vs 49%, respectively). Compared to embryos with all other patterns of fragmentation, those with type IV fragmentation had fewer cells (5.2 ⫾ 1.4) on day 3 and led to far fewer implantations and pregnancies. Fluorescence in situ hybridization (FISH) analysis has shown a significantly higher incidence of chromosomal abnormality in embryos with type IV fragmentation than those with types I, II, or III fragmentation (M Alikani, G Tomkin and S Munné, unpublished data). The same fragmentation pattern classification system21 was adopted by others29 who studied the incidence of different patterns among transferred embryos in pregnant and non-pregnant cycles. Prior to embryo transfer, the zona pellucida in all embryos was partially opened with acidified Tyrode’s solution but fragments were not removed. The degree of fragmentation was not noted in these embryos. The results showed that in pregnancy cycles, 48% of transferred embryos had type I fragmentation, while in non-pregnant cycles, a generally lower incidence of type I and a generally higher incidence of type II fragmentation had occurred. Implantation failed in seven cases in which embryos with fragmentation types III and IV were transferred exclusively. The latter outcome is not consistent with our own experience with these fragmentation patterns, but the discrepancy may be partly explained by the very low number of observations in the study of Desai et al.29 Moreover, since the classification of the patterns still has an element of subjectivity, observer variation may also account for these differences. In another interesting study,27 embryos with more than 25% fragmentation showed very low viability (0.8%). When cell number, degree of fragmentation, and cell asymmetry (the latter defined as ‘none, some, and severe’) were considered together, there was a measurable negative impact of asymmetry on viability of 8-cell embryos, regardless of the degree of fragmentation (⬍10% or 10–25%). Transfer of

7-cell and ⬍7-cell embryos with severe asymmetry led to complete failure of implantation.27 Collectively, these data point to a definition of ‘top quality’ embryos as those with less than 20% fragmentation, at least seven blastomeres on day 3 of development30,31, little asymmetry (attributable to asynchronous division of cells), and no major size discrepancy attributable to fragmentation or uneven division of cells.21,27,32 The impact of fragmentation on neonatal outcome is less clear. One study33 has suggested that transfer of embryos with 25% to ⬎50% fragmentation leads to significantly higher rates of fetal abnormalities than transfer of embryos with ⬍25% fragmentation. Four minor malformations occurred among the 180 children born following 309 transfers in the ⬍25% fragmentation group. In another group consisting of 75 transfers that included embryos with 25–50% fragmentation, one case of trisomy 21 and one of fibroma were seen among 13 newborns. Among 19 children born following 76 transfers that included embryos with ⬎50% fragmentation, two cases of trisomy 18, one of hydrocephalus with anal atresia, and one of hydrocele were found.33 Several aspects of this study are puzzling. For example, the large proportion of transfers that reportedly included embryos with ⬎50% fragmentation (75/460 or 16%) brings into question the accuracy of the fragmentation estimates. In our database of 3322 transfers homogeneous with respect to the degree of fragmentation, only 1.3% (43/3322) had embryos with ⬎35% fragmentation. The total number of embryos with 50% or more fragmentation within this group was 25. When the entire database was considered, 96 of 25 372 (0.38%) transferred embryos had 50% or more fragmentation. These figures reflect the deliberate exclusion of such embryos from transfer and the relatively low frequency with which such extensive fragmentation occurs in the first 2–3 days of culture (5893/79 936 or 7.4%). Another point of debate is the implied association of aneuploidy – in this case, trisomies – with fragmentation. So far, such an association has not been established by chromosomal analysis of large numbers of IVF embryos. The incidence of 

HPE_Chapter06.qxp

7/13/2007

4:46 PM

Page 58


aneuploidy (single chromosome loss or gain) does not seem to be related to embryo morphology, but to maternal age; the predominant abnormalities among fragmented embryos are polyploidy and extensive mosaicism which are often incompatible with development to term.34–37 The question of a potential relationship between early embryo morphology and structural abnormalities in fetuses should perhaps be more carefully examined in larger series of cases where several morphological parameters have been well defined and thoroughly examined. In the study of Ebner et al.,33 the embryos were evaluated by one static observation on day 2 of development, at a magnification of ⫻200. We have not systematically studied neonatal outcome following transfer of fragmented embryos. Although it has been suggested that embryo morphology on the day of transfer does not predict first trimester pregnancy loss,38 our data indicate a high incidence of early pregnancy loss in cases where extensively fragmented embryos were transferred. Among 43 cases in which all embryos had ⬎35% fragmentation at the time of evaluation (before fragment removal), 13 pregnancies resulted, but only three babies were born (two male and one female). Seven pregnancies were classified as biochemical, having shown at least three consecutive rises in ␤hCG levels but having failed to show a gestational sac with or without fetal heart activity. Three other pregnancies were lost between 7 and 9 weeks of gestation, after detection of a gestational sac. This is an early loss rate of 77% in an IVF program where the overall loss rates after a positive ␤hCG pregnancy test and detection of cardiac activity are approximately 25% and 11%, respectively. Such an extraordinarily high loss rate is an argument against transfer of embryos with extensive fragmentation (as also argued in ref. 33). Cytogenetic analysis was not performed on the aborted fetuses, so it is not known for certain whether chromosomal or other abnormalities were the cause of spontaneous abortion. However, such abnormalities have been found in 50–65% of very early pregnancy losses following IVF.39,40 Moreover,



transcervical embryoscopy (direct visualization of the fetus) combined with cytogenetic analysis of missed abortions has shown chromosomal abnormalities in 75%, and normal karyotype but structural defects in 18% of the cases.41 The observed incidence of implantation failure of embryos with extensive fragmentation reflects a persistent negative effect of fragmentation on embryo development in utero associated with fragmentation. The question is whether these effects are manifested during prolonged culture in vitro?

THE IMPACT OF FRAGMENTATION ON PREIMPLANTATION DEVELOPMENT

The relationship between embryo morphology and viability has become more evident following assessment of embryos in elective single embryo transfers.38 It is also of interest that a large study including a data set of 10 000 embryo transfers42 recently concluded that embryo quality (assessed on day 2) is the best predictor of pregnancy (following day 2 transfer) even when considered together with 16 other possible treatment cycle variables that included maternal age, duration and type of infertility, ovarian stimulation protocol, number of IVF attempts, progesterone level at hCG administration, sperm count, motility and morphology, number of retrieved and mature oocytes, number of embryos, and number of transferred embryos. The highest embryo quality score was assigned to 4-cell embryos with no fragmentation or fragmentation ⬍20% and evenly sized cells. Thus the suggestion that embryo morphology, including fragmentation, has no correlation with blastocyst formation or quality during extended culture43,44 seems both counter intuitive and generally unsupported. Our assessment of the impact of fragmentation on blastulation in over 1200 surplus embryos45 showed that with increasing fragmentation (0–15% vs ⬎15%), fewer embryos compacted, cavitated, and formed normal blastocysts. When fragmentation exceeded 35%, all processes were severely compromised, even though in some cases, fragments

HPE_Chapter06.qxp

7/13/2007

4:46 PM

Page 59


were excluded in the perivitelline space by the embryo (see Figure 6.4). When the origin of all the blastocysts that were selected for transfer (i.e. the best available embryos) was considered, we found that nearly 90% had developed from embryos with ⱕ15% fragmentation on day 3. Interestingly, fragmentation has been found to influence allocation of cells during differentiation.46 According to this study, increasing fragmentation resulted not only in reduced blastocyst formation but also in lower cell numbers when blastocysts did form. Moreover, in the case of minimal to moderate fragmentation, the reduction in cell number was apparently confined to the trophectoderm, while a steady number of inner cell mass cells was maintained. However, when fragmentation exceeded 25%, cell numbers in both lineages were reduced.46 These results are difficult to reconcile with the suggestion that blastocyst viability on day 5 is not affected by the degree of fragmentation on day 3 of development.27 Perhaps this can be partly attributed to the relatively small groups of patients and embryos examined.27 In the case of embryos with

A

B

⬎25% fragmentation, for instance, only five blastocysts (two of which implanted) and possibly two patients were included in the study.27 Thus the study’s conclusion cannot be considered definitive. Observational studies also suggest that the pattern of fragmentation has an influence on blastulation. The presence of large scattered fragments in embryos with uneven cell size is associated with a significant reduction in normal blastocyst formation compared with the other patterns.45 The reduced ability to form morphologically normal blastocysts often becomes obvious at compaction, which normally occurs on day 4 of development in the human. 47 ‘Regional’ or partial compaction, with exclusion of a number of cells and fragments from the morula, occurs frequently in fragmented embryos (Figure 6.4). In our study,45 nearly half of normally compacting embryos formed blastocysts, but this incidence was reduced to only one third in regionally compacted embryos and to 10% in embryos that did not show compaction on day 4 in culture. These data demonstrate that evidence of compaction on

C

D

E

I

J

M

F

G

H

Figure 6.4 Serial 4 micron thick optical sections through a day 5 human embryo, obtained on the laser scanning confocal microscope, showing ‘regional’ compaction. The excluded cells and fragments (A)–(E) as well as a small morula (M in (I)) are visible. The embryo has been stained with antibodies against E-cadherin (green). The nuclei are stained with propidium iodide (red).



HPE_Chapter06.qxp

7/13/2007

4:46 PM

Page 60


day 4 is highly prognostic for normal blastocyst formation. The relationship between fragmentation pattern and blastocyst formation was assessed in another study,48 in which the progress of 1566 surplus embryos was monitored until day 5 of development. Embryos with no fragmentation, or with type I fragmentation21 had a higher blastocyst formation rate than embryos with type II or III fragmentation, while embryos that were assessed to have type IV fragmentation failed to form any normal blastocysts.48 Based on the fragmentation patterns described by Antczak and Van Blerkom,22 the majority of cleavage stage embryos with types II (multiple layers of fragments with cell size reduction) or III (complete fragmentation of one blastomere) fragmentation continued to divide beyond the 4-cell stage, and about half reached the blastocyst stage, albeit in some cases with a delay of 12–24 hours. It was also observed that fragmentation at the 8-cell stage did not preclude development to the blastocyst stage, suggesting that the occurrence of fragmentation early in development was more likely to be detrimental to development. How does fragmentation interfere with normal embryonic development in vitro or in vivo?

THE IMPACT OF FRAGMENTS ON THE FRAGMENTING BLASTOMERE AND ITS SISTER BLASTOMERES

Cytoplasmic blebs that emerge often during the process of division can become ‘reincorporated’ into blastomeres.19,49 Furthermore, evidence suggests that small apical fragments may lyze, while larger fragments may swell and burst, possibly due to depletion of mitochondria and ATP stores in these structures.19 However, most fragments, once they are clearly formed, obviously persist throughout the early cleavage stages and it would be incorrect to imply that all fragments are transitory structures. Fragments may be excluded during compaction or even found in the blastocoel cavity of the 

blastocyst. According to one study, they are not static and can ‘move in concert’ with the underlying blastomeres.19 Clearly, fragmentation can affect the size of the fragmenting blastomeres, albeit not always appreciably. It has been shown that mean blastomere size decreases significantly with increasing degree of fragmentation.50 Highly fragmented embryos show a 43–67% reduction in blastomere volume. It is possible that the reduction in size, if substantial can lead to arrest of the affected blastomeres.21 Examination of individual ‘cells’ from fragmented embryos suggests that ‘cells’ ⬍45 ␮m in diameter in day 2 embryos and those ⬍40 ␮m in day 3 embryos should be considered fragments since they never contain a nucleus and only seldom contain chromosomal DNA.51 It has been suggested that certain patterns of fragmentation can result in ‘partial or near total loss’ of several regulatory proteins from specific blastomeres, with developmental consequences for both the affected blastomere and the embryo as a whole.22 The products of cell lysis or degeneration might cause deterioration of neighboring blastomeres or interfere otherwise in normal embryo development. This appears to be the case in the mouse, since the presence of deliberately lyzed cells among other viable cells reduces the incidence of blastocyst hatching, and removal of the lyzed cells restores hatching ability.52 At the same time, preliminary evidence suggested that removal of cryo-damaged cells from frozen-thawed human embryos was not only feasible but potentially beneficial.52 This technique has since been shown in clinical trials to improve pregnancy outcome.53,54 We investigated the question of fragment ‘toxicity’ in a mouse model,55 taking advantage of mouse blastomere totipotency at first cleavage.56 Two-cell embryos were dissociated (generating two blastomeres), and each blastomere (or half-embryo) was placed in a host zona pellucida (ZP; mouse, bovine, or human) either alone or aggregated with mouse or human egg/embryo fragments. The development of the half embryo-fragment aggregates and the control half embryos was monitored

HPE_Chapter06.qxp

7/13/2007

4:46 PM

Page 61


over the course of the following 3–4 days in culture. The fragments were not spontaneously generated by the half embryos, and therefore the experiments measured the impact of fragments or the microenvironment created by their presence on division of the original single blastomere (or half embryo) and its descendent cells. The presence of large numbers of fragments (in bovine or human ZP) appeared to be disruptive to development in some cases, but the overall results suggest that at least in the absence of an intact ZP, mouse blastomeres are neither overtly sensitive to close physical association with heterologous (often degenerative) cytoplasmic fragments nor responsive to the intrazonal microenvironment created by them (Figure 6.5). The loss of fragments through large slits in the zona pellucida presented a problem in these experiments and may have affected the outcome, and therefore the results should be interpreted with caution. However, mouse embryos are not negatively affected by exposure to cytoplasts, either removed from the zygote and reinserted in the perivitelline space, generated by enucleation of two blastomeres of an 8-cell embryo,57 or by enucleation of one or two blastomeres of a 2-cell embryo.58 Thus, in the mouse, there is no clear evidence of fragment ‘toxicity’ except under extreme circumstances. In the human, the picture is somewhat different. Clinical data suggest that removal of fragments from some fragmented embryos prior to intrauterine transfer leads to higher frequency of implantation. In the past 10 years, we have routinely used microsurgery to remove the majority of cytoplasmic fragments when fragmented embryos were to be transferred. This was initially based on finding a 4% overall increase in implantation rate when zona drilling (assisted hatching) and fragment removal were applied simultaneously, in comparison with embryos which were zona-drilled only.59 The analysis of a larger series of cases provided more support for the suggestion that the removal of the fragments contributes to the survival of fragmented embryos following intrauterine transfer: embryos with 10–35% fragmentation showed similar implantation rates, and a very low implantation rate (6%)

occurred only when fragmentation exceeded ⬎35% before fragment removal.21 In a more recent study,60 327 non-donor cycles in which embryos with ⬎10% fragmentation were subjected to assisted hatching and microsurgical fragment removal prior to transfer were retrospectively evaluated. Three groups were identified: 74 cycles in which at least one embryo required fragment removal, 39 cycles in which all embryos were subjected to this procedure, and a control group of 234 cycles that included embryos with ⬍10% fragmentation subjected to assisted hatching only. The data showed that the rates of implantation, live birth, spontaneous abortion, and fetal abnormalities in the first two groups were equivalent to those in the control group, suggesting a beneficial effect of fragment removal on pregnancy outcome. Thus it appears that the reduced viability of fragmented human embryos is at least partly attributable to the presence of the fragments per se. However, the beneficial effects of fragment removal are obviously limited. A recent analysis of a larger set of homogeneous transfers (described above) suggests that even after removal of fragments, the implantation rate of embryos with ⬎15% but ⬍35% fragmentation is still lower than that of embryos with 0–15% fragmentation (32% vs 18%, respectively). Moreover, removal of fragments from embryos with minimal fragmentation (0–15%) or from those in which fragmentation exceeds 35%, has little or no influence on the transfer outcome in a majority of cases. In the case of 0–15% fragmentation, viability is not substantially reduced in the first place, and in the case of ⬎35% fragmentation even though removal of fragments may occasionally lead to survival of the affected embryo (as mentioned above), it is unlikely to address the underlying cause of the abnormality or its associated developmental problems. Further insight into the possible impact of fragments on the neighboring cells was provided by examining the developmental capacity of blastomeres isolated from fragmented embryos. This was done either by individual culture of these blastomeres61,62 or their artificial aggregation in a host zona pellucida.61 

HPE_Chapter06.qxp

7/13/2007

4:46 PM

Page 62


Day 2

Day 3

Day 4

Day 5

A

a3

a4

a5

B

b3

b4

b5

C

c3

c4

c5

D

d3

d4

d5

E

e3

e4

e5

Figure 6.5 (Continued)



HPE_Chapter06.qxp

7/13/2007

4:46 PM

Page 63


In the first series of experiments, nucleated blastomeres and fragments were biopsied from discarded embryos with 1–10 cells and 20–75% (primarily type IV) fragmentation on day 3 of development. Roughly 40% of these arrested when placed in culture in isolation but the remaining 60% divided during the course of culture. Of those that divided, a substantial proportion went on to form a ‘cavity’ and blastulate, albeit with fewer than four cells. These observations suggest that the mixing of potentially normal and abnormal cells and cell fragments may reduce the development potential of the normal cells by reducing the likelihood of normal blastulation. The second set of experiments lends support to this proposal: non-viable embryos with extensive fragmentation and other abnormalities were disaggregated and blastomeres of normal appearance from two or more embryos were combined into ‘chimeric’ aggregates.61 Roughly one-third of these aggregates formed blastocysts with distinct inner cell masses and relatively high cell numbers, ranging from 31 to 56 cells (Figure 6.6). Of these cells, 52–90% were diploid. Chaotic mosaicism was the most common abnormality found in these embryos. This outcome demonstrates the developmental potential and regulatory capacity of a proportion of cells derived from non-viable embryos, again suggesting that poor development of the latter is at least partly attributable to the presence of the fragments and/or abnormal nucleated cells. Does the presence of fragments influence other aspects of early development?

THE IMPACT OF FRAGMENTS ON CELL–CELL INTERACTIONS/EMBRYO ORGANIZATION

It has been speculated that disruption of the spatial arrangement of cells through fragmentation may be a cause for reduced cell–cell contact and communication.21 Transmission electron microscopy does not support this proposal. According to Van Blerkom et al.19 (2001), although ‘lysed, ‘prelytic’ and intact fragments interposed between blastomeres seemed to prevent close apposition of adjacent plasma membranes’.19 Serial section analysis of images ‘demonstrated that these separated zones were focal and that close contact between opposed plasma membranes existed in other regions of these blastomeres.’ Whether this is a general phenomenon or one that is restricted to certain cases of fragmentation remains to be seen. It is nonetheless clear that during the course of culture in vitro, fragmented human embryos often fail to compact or they undergo abnormal compaction, excluding a number of cells and fragments (Figures 6.3–6.5).45 We investigated this phenomenon further by examining the localization of E-cadherin, a vital cell adhesion protein, in a large number of non-viable human embryos with normal and abnormal morphology.63 In all other mammalian species studied so far, E-cadherin is actively relocated in the course of embryogenesis.64,65 Relocation first occurs at the time of compaction, and involves the cells that form the outer layer of the embryo. In these cells,

Figure 6.5 The development of experimental and control half embryos from day 2 through day 5 in culture.In panel A, Cavitation and blastulation appear to be abnormal in both control and the experimental half embryos. In panel B (second row from top), on day 2 of development, five large fragments (arrowheads) aggregated with one blastomere are visible, while the sister blastomere from the same 2-cell embryo is seen in a host zona pellucida alone. One day later, on day 3 of development (b3), the five fragments can be seen surrounding three blastomeres (1⫻1/4 and 2⫻1/8). The control half embryo has exactly the same number of cells (1⫻1/4 and 2⫻1/8). On day 4 (b4), the fragments in the experimental half embryo have moved to one side of a compacted mass that has begun to cavitate. The control half embryo has also compacted and cavitated but appears better organized. By day 5, both half embryos have formed blastocysts. The fragments (arrows), now degenerated, are loosely attached to the experimental half embryo. Development of control and experimental blastomeres depicted in panels C, D, and E is comparable. Scale bar is 50 ␮m.



HPE_Chapter06.qxp

7/13/2007

4:46 PM

Page 64


A

B

A⬘

C

B⬘

D

C⬘

E

Figure 6.6 Development of a human aggregate. (A–C) Three discarded non-viable embryos on day 3 of development. (A⬘–C⬘) Dissociated cells of the respective non-viable day 3 human embryos in A–C, comprising mononucleated, multinucleated, and anucleate blastomeres/fragments. (D) Eleven of the mononucleated cells from the three embryos were used on day 3 to construct an aggregate. (E) Following 2 days of culture, the aggregate formed a blastocyst.

E-cadherin is transported from the cytoplasm to the cell membrane in areas of contact between cells. This relocation is a preliminary to the formation of junctional complexes between the trophectodermal cells, which are responsible for the integrity of the blastocyst. It is therefore reasonable to expect that fragmentation may be associated with the failure of proper expression, localization, or distribution of E-cadherin, all of which are required for normal compaction. Laser scanning microscopy (LSM) images suggest that the characteristic distribution pattern of E-cadherin is perturbed and erratic in abnormally cleaving human embryos (Figure 6.7).



Although it is not clear whether the erratic distribution is a cause or an effect of abnormal development, including fragmentation, these disturbances can nonetheless lead to failure of compaction, which in turn leads to failed or abortive blastulation. Moreover, the presence of non-interacting cells and fragments and interactive cells in the same embryo may contribute to the problem by disrupting cell signaling processes that are mediated by Ecadherin. It is therefore plausible that fragmentation, regardless of its underlying cause, disrupts early development via a mechanism that involves cell–cell interaction or communication.

HPE_Chapter06.qxp

7/13/2007

4:47 PM

Page 65


A

B

a

b

c

d

a

b

c

d

M

C

D

Figure 6.7 The distribution of E-cadherin (green) in human embryos at different stages of development. (A) Hatched day 7 blastocyst from a normally fertilized egg. The cells of the trophectoderm (both polar and mural) show an intense ‘belt’ of fluorescence indicating localization of E-cadherin in the membranes. Cells within the inner cell mass (arrows in panels a and b) show diffuse cytoplasmic E-cadherin. (B) An embryo showing abnormal blastulation and ‘belt’ staining of trophoectoderm cell membranes; many excluded cells and fragments which do not show any staining are visible (arrows in panels B, a, and b). (C) A day 5 human embryo with excluded cells and fragments (arrows) and a small morula (M) consisting of about 6 cells. (D) Compacting 8-cell embryo showing weak staining in areas of cell–cell contact (arrowheads). The images in A–D are projections of multiple 4–5 ␮m thick optical sections obtained on the laser scanning confocal microscope. Images in a–d are single optical sections.



HPE_Chapter06.qxp

7/13/2007

4:47 PM

Page 66


CAUSES AND MECHANISMS OF FRAGMENTATION APOPTOSIS AS A CAUSE OF FRAGMENTATION IN EGGS AND EMBRYOS OF NON-HUMAN SPECIES

A number of studies on the mechanisms and causes of fragmentation in mammalian eggs and embryos have viewed the phenomenon from the perspective of cell death, even referring to it as an ‘unequivocal example of apoptosis.’66 Apoptosis67 is a form of genetically regulated cell death ‘responsible for deletion of unwanted cells in processes as diverse as immunological tolerance, embryological remodelling, neoplasia, inflammation and normal tissue turnover.’68 Apoptosis has also been implicated in follicular atresia and germ cell loss during fetal development as well as in postnatal life.69 Although apoptotic and necrotic death express similar characteristics in early phases,70 they generally have different biological implications for neighboring viable cells. Necrosis results in the loss of plasma membrane integrity and an inflammatory response, while in apoptosis (in situ) membrane integrity is preserved and the cells are disassembled and phagocytosed without an inflammatory response.71 Thus apoptosis is considered a ‘physiological’ form of cell death while necrosis is considered ‘pathophysiological’ and a result of insult or injury to the cell.71 The ‘induction’ and ‘execution’ of the apoptotic pathway involve surface signaling, internal signal transduction mechanisms, and a complex interplay of proteins and other factors. Microtubule disruption, DNA damage, and a rapid increase in cytosolic Ca2⫹, all of which may be induced by chemotherapeutic drugs, can trigger this form of cell death.72 Sequential events following drug-induced apoptosis have been recently studied in a bladder cancer cell line.72 The particular drug used in the study was docetaxel, a taxane which inhibits microtubule depolymerization, preventing cell cycle completion. In this case, the sequence began within a few hours of drug treatment, with cell cycle arrest in the G2/M



phase; the disruption of microtubules within this phase is known to lead to degradation of the antiapoptotic protein B cell/lymphoma-2 or Bcl-2, bringing about a succession of other events. These include release of cytochrome C from mitochondria, the collapse of mitochondrial transmembrane potential, activation of caspase proteases, DNA fragmentation at internucleosomal sites (identified by electrophoretic laddering of DNA on agarose gels), and cell surface changes including exposure of phosphatidylserine to the outer surface (identified by binding of annexin V). Under the conditions of the above study, these events preceded massive DNA strand breaks which may be identified by terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL).72 One of the first studies to suggest apoptosis as the underlying mechanism of spontaneous fragmentation in eggs73 was based on a simple observation of a gradual increase in the number of eggs with ‘shrinkage of ooplasm and spontaneous cytoplasmic fragmentation’ after 24–40 hours of culture in vitro, and the association of this with a positive TUNEL reaction in the aging eggs.16 A subsequent study74 found a high incidence of DNA fragmentation in eggs from young (7–24 weeks old) and aged (40–48 week old) mice following 60 hours of culture in vitro; the incidence appeared to increase with increasing maternal age and time in culture. These results were disputed by Van Blerkom and Davis75 in a study in which TUNEL and annexin V assays were used to examine DNA fragmentation and phosphatidylserine exposure, respectively, in more than 300 intact and 500 fragmented mouse oocytes. The examinations were done at 24 hour intervals, during 6 days of culture, and in three different types of media.75 Under the conditions of this study, TUNEL fluorescence of MII chromosomes and annexin V staining of the oolemma were rare in newly ovulated or fragmented mouse oocytes, and when present, they were not spatially or temporally related. This suggests that these changes did not reflect regulated changes that occurred during an apoptotic process.75 However, arguments in support of the apoptotic nature of fragmentation have been bolstered by

HPE_Chapter06.qxp

7/13/2007

4:47 PM

Page 67


several other observations. Expression of a number of pro- and anti-apoptotic genes, including members of the apoptosis-related Bcl-2 protein family – Bax and Bcl-2 – has been documented in mouse embryos.76,77 Caspase activity has been found in both spontaneously fragmented eggs and in those that fragment following 24 hours of treatment with the chemotherapeutic drug doxorubicin.66 Moreover, fragmentation of eggs following drug treatment is inhibited by caspase inhibitors78 and by induction of a deficiency in caspase-2.79 One question that may be raised is whether fragmentation following long-term exposure of eggs to antineoplastic drugs that target DNA or microtubules is relevant to the developmental processes that normally lead to fragmentation. This experimental approach overlooks the tendency of such drugs to alter a variety of cellular functions and cell components.80 For example, anthracycline antibiotics, such as doxorubicin, are known to disturb intracellular calcium homeostasis and generate reactive metabolites that directly damage DNA, proteins, and cell membranes; their exact mechanism of antitumor action is still not fully understood. Thus the results of such experiments with eggs should be interpreted with caution. Fragmentation in mouse 2-cell embryos has also been linked to apoptosis. Unlike the situation with mouse eggs (and human embryos), spontaneous fragmentation in mouse embryos is rare and its experimental induction in most strains requires significant manipulation. Nonetheless it has been suggested that crossing of certain male and female mouse strains leads to different tendencies toward fragmentation,81 leading to the conclusion that fragmentation at the 2-cell stage is genetically controlled and influenced by both maternal and paternal genotypes. This conclusion was investigated further through pronuclear transfer experiments that were designed to examine the cellular basis for the parental genotype effect on fragmentation. According to the investigators, the experiments showed that fragmentation is predominantly controlled by the maternal pronucleus, and occurs as a result of ‘incomplete suppression’ (by fertilization)

of pathways leading to death.82 The role played by the cytoplasm was discounted. Interestingly, another study has also found that strain differences influence the extent of apoptosis in mouse embryos, but this was first detected at the blastocyst stage.83 Varying apoptotic indices were observed in blastocysts with similar cell numbers generated from various crosses of two different strains of mice and grown in two different culture media, or allowed to develop in vivo. Overall, the relationship between apoptosis and mouse embryo fragmentation during the cleavage stages is not well established; potential differences between spontaneous and drug-induced fragmentation have been completely overlooked. APOPTOSIS AS THE CAUSE OF FRAGMENTATION IN HUMAN EMBRYOS

As in the case of fragmented mouse oocytes, the discovery of condensed and TUNEL-labeled nuclei in arrested fragmented human embryos led to the proposal that apoptosis is the cause of early embryo fragmentation.17,84 This suggestion, however, stems from description of apoptotic nuclei in human blastocysts.85 Expression of mRNA for several apoptosisrelated proteins and localization of the proteins86 suggest that apoptosis may be activated in human eggs and embryos. The proteins include the antiapoptotic Bcl-2, family members, Bcl-2 Bcl-x, and Mcl-1 and pro-apoptotic Bax, Bik, and Bad as well as caspases.22,87–91 In one study,89 Bax and Bcl-2 mRNAs were found to be expressed throughout preimplantation development and in oocytes, but active caspases and proapoptotic Bad mRNA expression were found primarily after compaction, in morulae and blastocysts. These observations, along with the absence of phagocytic activity in early blastomeres, led the authors to conclude that apoptosis is inhibited during the early cleavage stages but may be initiated after compaction.89 A comprehensive study examined the expression profile of 11 members of the Bcl-2 family in discarded embryos with normal morphology as well as in embryos with fragmentation exceeding 50%.91 A control group consisted of embryos that 

HPE_Chapter06.qxp

7/13/2007

4:47 PM

Page 68


had been frozen at the pronuclear stage with siblings that had produced pregnancies following fresh embryo transfer. Expression of Bcl-2, Bcl-XL, Bcl-W, MCL-1, Bak, Bad, BokL, Bid, Bik, and Bcl-XS was variable, but Bax was expressed at all stages in virtually all embryos. The number of Bcl-2 genes expressed in fragmented embryos was not different from that in intact embryos; however, at the 8-cell stage fragmented embryos expressed far fewer Bcl-2 genes than the control group. The overall conclusion of the study was that the Bcl-2 family of genes is involved in both normal preimplantation development and in ‘the process of severe fragmentation.’ The authors then suggested that prevention of fragmentation in human embryos through optimization of culture media may lead to suppression of apoptosis that is necessary for the elimination of genetic damage.91 It should be kept in mind, however, that while the presence of the basic components of an apoptotic pathway suggests an ‘apoptotic capability’ on the part of human embryos, it does not prove a causal relationship between genetic abnormality and fragmentation, or between apoptosis and fragmentation. Extensive fragmentation is clearly associated with a high incidence of chromosome mosaicism,34 but it is by no means certain that mosaicism is the cause of fragmentation. Indeed, as will be discussed later in this chapter, it is possible that the two occur as a result of the same underlying abnormality. Moreover, aneuploidy appears to be strictly related to maternal age and not embryo morphology.37 This suggests that the abnormalities/deficiencies that mediate loss or gain of a single chromosome in successive divisions may be distinct from those that lead to fragmentation. The relationship between apoptosis and fragmentation is equally unclear. Van Blerkom et al.19 applied single cell alkaline gel electrophoresis (comet assay) to fragmented human embryos in which fragments occurred at the 2–4-cell stage and persisted until the 8–12-cell stage. This assay was carried out in addition to TUNEL and annexin V in order to detect DNA damage. Positive TUNEL fluorescence and ‘comets’ resulting from DNA fragments were evident only in second polar bodies, and in embryos



with cell lysis, or where DNA cleavage was experimentally induced by DNase treatment. These studies also revealed a major flaw in one previous study84 by showing that fixed embryos universally displayed annexin V labeling. Therefore, despite the perpetuation of this notion in the literature, it appears that apoptosis as a cause of fragmentation in cleavage stage human embryos is far from established. FRAGMENTATION AS A RESULT OF METABOLIC OR GENETIC DEFECTS

The distribution of mitochondria within the cytoplasm of pronuclear human eggs has been shown to be often asymmetric; in some cases, large regions of cortical cytoplasm may be devoid of mitochondria.92 Some reports suggest that this deficiency can lead to differences in the mitochondrial content of the two blastomeres that result from the first cleavage division, and, in some cases, also in blastomeres resulting from subsequent divisions.19,22 Mitochondria function as the site of ATP production by oxidative phosphorylation; they also function in regulation of Ca2⫹ homeostasis in the cell. Therefore, the developmental consequences of mitochondrial deficiencies (both numerical and functional) may be significant (reviewed in reference 20). It has been speculated that ‘global or focal differences in ATP levels’ may contribute to fragmentation phenotypes in human embryos.19 For example, formation of organelle-free fragments and blebs in certain ‘benign’ patterns of fragmentation has been likened to the initial events of oncosis.19 Oncosis has been described in some oxygen-deprived somatic cells with diminished ATP concentration, and is characterized by elaboration of plasma membrane blebs that are mostly devoid of organelles. The blebs may either be resorbed or they may swell and lead to cell lyze, depending on whether ATP levels in the cell are replenished or not.93 On the other hand, it has been argued that different types of fragmentation that are associated with decreased developmental competence may represent ‘a continuum that includes necrosis and apoptosis.’20

HPE_Chapter06.qxp

7/13/2007

4:47 PM

Page 69


Another recent proposal regarding the origin of fragmentation places emphasis on the role of the nucleus, and concerns telomere length. Telomeres are ‘distinctive structures, composed of a repetitive DNA sequence and associated proteins that cap the ends of linear chromosomes, are essential for maintaining the integrity and stability of eukaryotic genomes, and under some circumstances, can influence cellular gene expression.’94 With each cell division, telomere length decreases; thus the process of aging and cancer lead to telomere shortening.94 It has been proposed that artificial shortening of telomeres can trigger cytoplasmic fragmentation in mouse embryos.95 This proposal has formed the basis of a study in the human in which telomere length was examined in 43 in vitro matured MII oocytes obtained from 21 patients undergoing IVF (an average of two chromosome spreads per patient).96 The results showed that maximum telomere length in eggs was inversely and significantly related to cytoplasmic fragmentation in day 3 embryos, after controlling for age and basal FSH. For every kilobase decrease in telomere length, day 3 embryo fragmentation increased by 0.31%. The authors went on to conclude that telomere shortening induces apoptosis in human embryos, and that this is consistent with a theory linking telomere length to reproductive senescence in women.96 Although interesting, this study is not definitive. Cytoplasmic fragmentation on day 2 of development was not predicted by telomere length, but there is no reasonable explanation for this observation. The authors suggest that fragmentation in day 2 embryos may result from a different ‘pathobiologic process’ than that in day 3 embryos. They state that their data implicate telomere length in only 20% of the fragmentation seen in embryos, but the reason for this is not clear. Finally, in this study, telomere length was measured in in vitro matured spare eggs, with the assumption that they were representative of the embryo cohort; however, the validity of this assumption is questionable since in the same study, telomere length in sister in vitro matured eggs was found to be only ‘moderately correlated.’

FRAGMENTATION AND THE CYTOSKELETON: A NEW PERSPECTIVE

Fragmentation is reminiscent of cell division, at least in the sense that one original cell is partitioned. Normal cell division involves radical reorganization of the cytoskeleton, particularly in the microtubules and their relationship with the cortical microfilaments. In tissue culture cells, interference with actin-binding proteins that modulate filamentous actin polymerization leads to ‘ectopic contractions’ and severe membrane blebbing,97 reminiscent of fragmentation. Early experiments in the sea urchin,98 suggest that somewhat similar reorganizations might be required for fragmentation to take place. This in turn raises the question of whether fragmentation takes place at any time during the cell cycle, or is restricted to certain phases. We used a more dynamic approach to answer this question and investigate the relationship between fragmentation and other common embryo abnormalities.58 The mouse is a good experimental model for studying this question, since the timing of the second meiotic division and subsequent mitotic divisions is well defined and relatively easy to control. Artificial activation of the egg sets it on a course which begins with progression through anaphase and telophase of meiosis II (within 1 hour of activation), extrusion of a second polar body (within 2–3 hours of activation), and formation of a single (female) pronucleus (within 4–5 hours of activation). At the level of the cytoskeleton, the microtubule organizing centers (MTOCs), inactive in the metaphase-arrested egg, nucleate an extensive network of microtubules throughout the activated egg. This network remains until the onset of first mitosis, at which time the network is depolymerized and microtubules are concentrated almost exclusively in the mitotic spindle until mitosis has been completed. Drawing on experience gained during nuclear transplantation studies, we used enucleation as a means of making mouse eggs fragmentationprone.58 Following removal of the meiotic spindlechromosome complex, eggs did not fragment, so



HPE_Chapter06.qxp

7/13/2007

4:47 PM

Page 70


long as they did not become activated. Exposure of eggs to activation conditions after the spindle had been removed led to fragmentation 15 hours later, when nucleated controls entered the first mitosis. By 24 hours, some fragments and blebs in fragmented cytoplasts receded and the cytoplast assumed a more spherical shape. Enucleation of normally fertilized or artificially activated eggs after extrusion of the second polar body also led to fragmentation at about the same time, and again

coinciding with first mitosis in nucleated control embryos (Figure 6.8). Based on these observations, we concluded that fragmentation does not occur in mitotically inactive cells. Since mature eggs are arrested in the metaphase of meiosis II, egg activation is a prerequisite to fragmentation, as it is for cell cycle progression and division. Although fragmentation in the cytoplasts coincided with entry of nucleated cells into mitosis, this phenomenon represented the cytokinetic phase

mmb

fpn pbl

A

B

C

D

Figure 6.8 Investigation of cytoplasmic fragmentation in a mouse model. The status of microtubules in intact activated eggs and eggs from which the meiotic spindle–chromosome complex was removed is shown. (A) At 5.5 hours postactivation (PA), the intact egg shows extrusion of the first polar body (pb1), and development of the female pronucleus (fpn) and the meiotic midbody (mmb). (B) At 5.5 hours PA, the spindle-depleted egg shows several microtubule organizing centers (arrowheads), positioned around the center of the egg. (C) At 14 hours PA, in the intact egg, microtubules are exclusively invested in the first mitotic spindle (arrow), which rests in the center of the egg. (D) At 14 hours PA, the spindle–chromosome complex-depleted egg is in the cytokinetic phase of the cell cycle and is fragmented. Eggs have been stained for tubulin (green) and DNA (red). Images are projections reconstructed from several 3 ␮m thick optical sections. Scale bar is 20 ␮m.



HPE_Chapter06.qxp

7/13/2007

4:47 PM

Page 71


of the cell cycle in the former pushed forward by the removal of the nuclear structure, whereby the necessity for coordination of nuclear and cytoplasmic divisions (and any delay normally imposed by the former on the latter) was abolished. These results are in general agreement with those reported by Liu et al.99 who showed that oocyte activation and meiotic exit were prerequisites to fragmentation following enucleation or treatment of eggs with anticancer drugs. Our experiments went further, however. We reasoned that though fragmentation is indeed a deviant form of cytokinesis, it is not a result of apoptosis; therefore, it should be possible to induce the phenomenon during cytokinesis of other mitotic divisions and also during the second meiotic division. During the second mitotic cell cycle, blastomere fragmentation occurred only when enucleation was carried out in late interphase. Blastomeres that were enucleated during early–mid interphase of the second cell cycle remained intact during culture. The failure of fragmentation in enucleated blastomeres was reminiscent of the ‘2-cell block’ which takes effect at the late G2/M phase of the second cell cycle when maternal transcripts are depleted and the zygotic genome must be activated for normal development to continue. The fact that ‘arrest’ of cells in interphase inhibited fragmentation of enucleated cells provides further support for the notion that fragmentation represents atypical cytokinesis. To investigate the possibility of fragmentation during meiosis II, we took advantage of the properties of cytochalasin B (CCB). CCB causes microfilaments to depolymerize, which prevents cytokinesis (as it does fragmentation); activated eggs exposed to this drug do not extrude a second polar body, but instead remain single celled and form a second female pronucleus. Removal of CCB once the cell has reached interphase restores the microfilaments and subsequent division activity. In these experiments, the female pronuclei were allowed to form in the presence of CCB; both pronuclei were then removed and the cytoplasts were placed in culture free of CCB. Following approximately 9 hours of

culture – several hours before mitosis 1 – a proportion of the cytoplasts showed evidence of limited fragmentation in the area where the nuclei were situated prior to enucleation; other eggs showed division into two ‘cells’. This activity coincided with and resembled completion of meiosis II in nucleated eggs, and was thus considered to be meiotic fragmentation, albeit delayed. A second wave of fragmentation in these eggs coincided with mitosis 1 in nucleated controls, and led to complete fragmentation of the cytoplasts. This was considered to be mitotic fragmentation. These experiments therefore not only pointed to more similarities between fragmentation and cytokinesis in general, but also provided evidence that meiotic and mitotic waves of fragmentation can occur in the same eggs (Figure 6.9). It was also concluded that the common cause of fragmentation during both meiosis and mitosis was cytoskeletal disorder brought about by the absence of a nucleus or structures associated with it. As we reported and as Liu et al.99 also concluded, fragmentation primarily involved the cytoplasm/ cytoskeleton and the process was inhibited if either actin filaments or microtubules were made to depolymerize. However, in contrast to our finding that fragmentation of enucleated eggs occurred either slightly ahead of or at the same time as mitosis in nucleated eggs, Liu et al.99 found that fragmentation was delayed by 2 hours compared to cleavage in intact zygotes. They suggested that this delay might provide eggs with ‘the opportunity to assay DNA integrity for spindle assembly prior to proceeding with cleavage.’ Moreover, the higher frequency of fragmentation among enucleated eggs compared to drug-treated eggs was presented as evidence that the extent of DNA damage may determine ‘whether cells undergo cell cycle arrest for DNA repair, or apoptotic fragmentation for cell death.’ Contrary to our view of fragmentation as aberrant division due to cytoskeletal disorder, the overall conclusion drawn by Liu et al.99 was that activation of a checkpoint for DNA integrity at first mitosis leads to apoptotic fragmentation of eggs with damaged (or absent) DNA. Based on their observations, they proposed a model of fragmentation according 

HPE_Chapter06.qxp

7/13/2007

4:47 PM

Page 72


16 h PA

25 h PA

Enucleated

9 h PA

B

C

D

E

F

Control

A

Figure 6.9 Activated enucleated (A–C) and activated intact (D–F) eggs that were maintained in KSOM-CCB for 4–5 hours immediately postactivation (PA) then transferred to KSOM. Nine hours postactivation, cytoplasts showed cortical ruffling and either produced one or more polar cytoplasts or ‘cleaved’ more or less evenly (A). Beginning at 16 hours postactivation and up to 25 hours PA, these enucleated eggs showed more extensive fragmentation affecting the entire egg (B) and (C). At the same time-points, respectively, nucleated control eggs showed blebbing and formation of one or more ‘polar bodies’ (D) and entry into mitosis or cleavage (E) and (F). Scale bar is 80 ␮m.

to which the failure of damaged (or missing) chromosomes and/or centromeres to capture microtubules to form a functional spindle would lead the ‘disconnected microtubule asters’ to direct multiple furrow formation and fragmentation. While we agree that microtubules not associated with the spindle contribute to multiple furrow formation in enucleated eggs, we think that the emphasis of this model on the inability of ‘damaged chromosomes or centromeres’ to direct cytokinesis is debatable since the actual role of the chromosomes in cytokinesis is still unclear. 

In the course of our initial studies, we noted that the morphology and number of cytoplasmic MTOCs underwent a dramatic change when cytoplasts were left in culture for 24 hours. The intact eggs also showed some of these changes. In further investigations, it became clear that these changes were the underlying mechanism of fragmentation of aged eggs following late activation (Alikani and Willasden, in preparation). At present, the precise nature of the molecular changes underlying the altered state of the MTOCs during aging is uncertain. A limited molecular

HPE_Chapter06.qxp

7/13/2007

4:47 PM

Page 73


analysis of aged eggs – again from the point of view of the apoptotic origin of fragmentation – has shown a decrease in mRNA and protein levels for the antiapoptotic protein Bcl-2, while the levels of Bax (a proapoptotic protein) mRNA apparently remained unchanged.100 This and other potential ‘cytoplasmic deficiencies’ were apparently overcome (and fragmentation prevented) by fusion of recently ovulated activated eggs with the aged eggs.100 Injection of granulosa cell mitochondria has also been said to prevent the ‘apoptotic’ fragmentation of aging eggs.101 While the possibility cannot be excluded that aging related changes in cytoplasmic MTOCs described here are among several components of an apoptotic process (as they obviously can lead to cell death), a more extensive (and inclusive) molecular analysis of the aged eggs should provide further insight into the nature of this phenomenon.

RELEVANCE OF ANIMAL MODELS TO

structural junctions. Whether the impetus for this change is the activation of the genome, increased adhesion between the cells, reduced cell size or a combination of these factors remains to be determined. Nevertheless, the model presented here suggests that even though in some cases cell death is the inevitable result of fragmentation in human embryos, fragmentation as such is a manifestation of abnormal cytokinesis rather than cell death. It has been observed that fragmentation is often associated with abnormal spindle forms (Alikani, unpublished observations)(Figure 6.10).106 A model of fragmentation as aberrant cytokinesis in the presence of spindle and cytoskeletal abnormalities would provide a plausible explanation for the association of fragmentation with increased levels of chromosome mosaicism rather than age-related aneuploidy.

CONCLUSIONS AND FUTURE STUDIES

HUMAN EMBRYO FRAGMENTATION

The process of fragmentation in eggs and embryos is likely to be essentially the same, whatever the triggering cause(s). However, in the absence of direct evidence in human embryos, we can only speculate about the relevance of the observations in the mouse to fragmentation and related phenomena in human eggs and embryos – not to mention other cell types. The topography and organization of the cytoskeleton differ between species. For example, the sperm centrosome is thought to play an essential role as the microtubule organizing center in the human102–104 but not in the mouse.105 Moreover, the precise role played by the maternal MTOCs following activation in the human is not yet clear. In the human, the genome becomes activated sometime between the 4- and 8-cell stages, but fragmentation mostly occurs during the first and second divisions. Indeed, it appears that the tendency of human blastomeres to fragment is simply lost around the time of genomic activation, and particularly after compaction and the establishment of

It is becoming increasing clear that the simple interpretation of cytoplasmic fragmentation as a degenerative process is unjustified. Indeed, even those investigators who originally put forth the notion of apoptosis as the sole cause of fragmentation in human embryos,17 are now modifying their views90 to account for the great variability in fragmentation phenotypes and in the viability of fragmented embryos.21,22 From the perspective of clinical outcome, future studies are needed to establish whether implantation of extensively fragmented embryos with few cells at the cleavage stages has any long-term effects. Preliminary evidence presented here suggests that the incidence of early pregnancy loss may increase following transfer of embryos with extensive fragmentation. However, this observation needs to be confirmed in larger studies. Neonatal outcome is also of concern, and although some evidence suggests an increase in the frequency of fetal malformations and aneuploidy following transfer of embryos with ⬎25% fragmentation,33 this issue deserves to be examined more closely. 

HPE_Chapter06.qxp

7/13/2007

4:47 PM

Page 74


F

F

A

F

A⬘

B

Figure 6.10 Fragmentation (F) in cleavage stage embryos accompanied by abnormal spindle forms (arrowheads). A and A⬘ show the same embryo in two different focal planes. Scale bar is 20 ␮m.

The latter observation also highlights the importance of adopting a standardized fragmentation classification system, as the results of such studies would be otherwise difficult to interpret or compare. Such a classification system would consider both the degree and pattern of fragmentation in relation to timing of fragmentation during development in culture. With respect to experimental investigations into the origins of fragmentation, at present, the extent to which the results of experiments with mouse eggs and blastomeres58 apply to their human counterparts is still unknown. Therefore, experiments with human eggs and blastomeres will be necessary to detect and reconcile the differences that are likely to emerge. It must be kept in mind that while the starting material used in the mouse experiments could be assumed to have been normal, such an assumption cannot be made with equal confidence when human material is used. It seems reasonable to predict, as suggested by our preliminary work, that such differences will turn out to be in the detail rather than principle. Regardless of the validity of this prediction, by firmly establishing a link between fragmentation and both nuclear and cytoplasmic cell division, and the development of a relatively well defined yet versatile model, an important area for future research 

has been opened. The most obvious objective for this research is to determine factors that define and maintain cytoskeletal integrity in human eggs. This would aid in development of strategies to prevent fragmentation and related abnormalities in a significant proportion of eggs fertilized in vitro. The case for such research is compelling, since in its absence human beings rather than human cells will be the experimental subjects – as they have been hitherto in assisted human reproduction. ACKNOWLEDGMENTS

The author would like to gratefully acknowledge the following: Mr. Giles Tomkin for data analysis; Drs. Nury Steuerwald and Larry Leamy of the University of North Carolina for statistical analysis; the patients at the Institute for Reproductive Medicine and Science at Saint Barnabas Medical Center for donating their extra embryos to research, and Dr. Steen M. Willadsen as well as the editors of this volume for critical reading of the manuscript. Parts of this manuscript appear as chapters in the authors’ PhD dissertation completed under the supervision of Professor Alan Trounson at Monash University (Clayton, Victoria; Australia)

HPE_Chapter06.qxp

7/13/2007

4:47 PM

Page 75


REFERENCES 1. Tomkin G, Cohen J. Data management and interpretation– computerized database for an ART clinic: hardware and software requirements and solutions. In: Gardner D, Weissman A, Howles C, Shoham Z, eds. Textbook of Assisted Reproductive Techniques, Laboratory and Clinical Perspectives. London: Martin Dunitz, 2001: 367–80. 2. Edwards RG, Steptoe PC, Purdy JM. Fertilization and cleavage in vitro of preovulatory human oocytes. Nature 1970; 227: 1307–9. 3. Hertig AT, Rock J, Adams EC, Menkin MC. On the preimplantation stages of the human ovum: a description of four normal and four abnormal specimens ranging from the second to the fifth day of development. Contrib Embryol 1954; 35: 199–220. 4. Ortiz ME, Croxatto HB. Observations on the transport, aging, and development of ova in the human genital tract. In: Talwar GP, ed. Recent Advances in Reproduction and Regulation of Fertility. New York: Elsevier/North-Holland Biomedical Press, 1979: 307–17. 5. Buster JE, Bustillo M, Rodi IA et al. Biologic and morphologic development of donated human ova recovered by nonsurgical uterine lavage. Am J Obstet Gynecol 1985; 15: 211–17. 6. Alikani M. Cytoplasmic fragmentation in human embryos in vitro: implications and the relevance of fragment removal. In: Gardner D, Weissman A, Howles C, Shoham Z, eds. Textbook of Assisted Reproductive Techniques, Laboratory and Clinical Perspectives. London, Martin Dunitz, 2001: 169–82. 7. Hendrickx AG, and Kraemer DC. Preimplantation stages of baboon embryos. Anat Rec 1968; 162: 111–20. 8. Enders AC, Hendrickx AG, Binkerd PE. Abnormal development of blastocysts and blastomeres in the Rhesus Monkey. Biol Reprod 1982; 26: 353–66. 9. Killeen ID, Moore NW. The morphological appearance and development of sheep ova fertilized by surgical insemination. J Reprod Fertil 1971; 24: 63–70. 10. Puissant F, Van Rysselberge M, Barlow P et al. Embryo scoring as a prognostic tool in IVF treatment. Hum Reprod 1987; 2: 705–8. 11. Wakayama T, Yanagimachi R. Effect of cytokinesis inhibitors, DMSO and the timing of oocyte activation on mouse cloning using cumulus cell nuclei. Reproduction 2001; 122: 49–60. 12. Kawahara M, Mori T, Tanaka H, Shimizu H. The suppression of fragmentation by stabilization of actin filament in porcine enucleated oocytes. Theriogenology 2002; 58: 1081–95. 13. Cheong HT, Ikeda K, Martinez Diaz MA et al. Development of reconstituted pig embryos by nuclear transfer of cultured cumulus cells. Reprod Fertil Dev 2000; 12: 15–20. 14. Van Blerkom J, Antczak M, Schrader R. The developmental potential of the human oocyte is related to the dissolved oxygen content of follicular fluid: association with vascular endothelial growth factor levels and perifollicular blood flow characteristics. Hum Reprod 1997; 12: 1047–55. 15. Pickering SJ, Braude PR, Johnson MH et al. Transient cooling to room temperature can cause irreversible disruption of the meiotic spindle in the human oocyte. Fertil Steril 1990; 54: 102–8. 16. Pellestor F, Dufour MC, Arnal F, Humeau C. Direct assessment of the rate of chromosomal abnormalities in grade IV human embryos produced by in-vitro fertilization procedure. Hum Reprod 1994; 9: 293–302. 17. Jurisicova A, Varmuza S, Caspar RF. Programmed cell death and human embryo fragmentation. Mol Hum Reprod 1996; 2: 93–8. 18. Yang HW, Hwang KJ, Kwon HC et al. Detection of reactive oxygen species (ROS) and apoptosis in human fragmented embryos. Hum Reprod 1998; 13: 998–1002.

19. Van Blerkom J, Davis P, Alexander S. A microscopic and biochemical study of fragmentation phenotypes in stage-appropriate human embryos. Hum Reprod 2001; 16: 719–29. 20. Van Blerkom J. The enigma of fragmentation in early human embryos: possible causes and clinical relevance. Essential IVF; Basic Research and Clinical Applications. In: Van Blerkom J, Gregory L. eds. Norwell, Massachusetts, USA: Kluwer Academic Publishers, 2004: 377–421. 21. Alikani M, Cohen J, Tomkin G et al. Human embryo fragmentation in vitro and its implications for pregnancy and implantation. Fertil Steril 1999; 71: 836–42. 22. Antczak M, Van Blerkom J. Temporal and spatial aspects of fragmentation in early human embryos: possible effects on developmental competence and association with the differential elimination of regulatory proteins from polarized domains. Hum Reprod 1999; 14: 429–47. 23. Hoover L, Baker A, Check JH et al. Evaluation of a new embryograding system to predict pregnancy rates following in vitro fertilization. Gynecol Obstet Invest 1995; 40: 151–7. 24. Erenus M, Zouves C, Rajamahendran P et al. The effect of embryo quality on subsequent pregnancy rates after in vitro fertilization. Fertil Steril 1991; 56: 707–10. 25. Giorgetti C, Terriou P, Auquier P et al. Embryo score to predict implantation after in vitro fertilization: based on 957 single embryo transfers. Hum Reprod 1995; 10: 2427–31. 26. Ziebe S, Petersen K, Lindenberg S et al. Embryo morphology or cleavage stage: how to select the best embryos for transfer after in vitro fertilization. Hum Reprod 1997; 12: 1545–9. 27. Racowsky C, Combelles CM, Nureddin A et al. Day 3 and day 5 morphological predictors of embryo viability. Reprod Biomed Online 2003; 6: 323–31. 28. Ciray HN, Karagenc L, Ulug U et al. Use of both early cleavage and day 2 mononucleation to predict embryos with high implantation potential in intracytoplasmic sperm injection cycles. Fertil Steril 2005; 84: 1411–6. 29. Desai NN, Goldstein J, Rowland DY, Goldfarb JM. Morphological evaluation of human embryos and derivation of an embryo quality scoring system specific for day 3 embryos: a preliminary study. Hum Reprod 2000; 15: 2190–6. 30. Gerris J, De Neubourg D, Mangelschots K et al. Prevention of twin pregnancy after in vitro fertilization or intracytoplasmic sperm injection based on strict embryo criteria: a prospective randomized clinical trial. Hum Reprod 1999; 14: 2581–7. 31. Van Royen E, Mangelschots K, De Neubourg D et al. Characterization of a top quality embryo, a step towards single-embryo transfer. Hum Reprod 1999; 14: 2345–9. 32. Hardarson T, Hanson C, Sjogren, Lundin K. Human embryos with unevenly sized blastomeres have lower pregnancy and implantation rates: indications for aneuploidy and multinucleation. Hum Reprod 2001; 16: 313–8. 33. Ebner T, Yaman C, Moser M et al. Embryo fragmentation in vitro and its impact on treatment and pregnancy outcome. Fertil Steril 2001; 76: 281–5. 34. Munné S, Alikani M, Tomkin G, Grifo J, Cohen J. Embryo morphology, developmental rates, and maternal age are correlated with chromosome abnormalities. Fertil Steril 1995 ; 64 : 382–91. 35. Marquez C, Sandalinas M, Bahce M et al. Chromosome abnormalities in 1255 cleavage-stage human embryos. Reprod Biomed Online 2000; 1: 17–26. 36. Magli MC, Gianaroli L, Ferraretti AP. Chromosomal abnormalities in embryos. Mol Cell Endocrinol 2001; 183(Suppl 1): S29–34.



HPE_Chapter06.qxp

7/13/2007

4:47 PM

Page 76


37. Munné S. Chromosome abnormalities and their relationship to morphology and development of human embryos. Reprod Biomed Online 2006; 12: 234–53. 38. De Neubourg D, Gerris J, Mangelschots K et al. Single top quality embryo transfer as a model for prediction of early pregnancy outcome. Hum Reprod 2004; 19: 1476–9. 39. Lower AM, Mulcahy MT, Yovich JL. Chromosome abnormalities detected in chorionic villus biopsies of failing pregnancies in a subfertile population. Br J Obstet Gynaecol 1991; 98: 1228–33. 40. Schmidt-Sarosi C, Schwartz LB, Lublin J et al. Chromosomal analysis of early fetal losses in relation to transvaginal ultrasonographic detection of fetal heart motion after infertility. Fertil Steril 1998; 69: 274–7. 41. Philipp T, Philipp K, Reiner A et al. Embryoscopic and cytogenetic analysis of 233 missed abortions: factors involved in the pathogenesis of developmental defects of early failed pregnancies. Hum Reprod 2003; 18: 1724–32. 42. Terriou P, Sapin C, Giorgetti C et al. Embryo score is a better predictor of pregnancy than the number of transferred embryos or female age. Fertil Steril 2001; 75: 525–31. 43. Graham J, Han T, Porter R et al. Day 3 morphology is a poor predictor of blastocyst quality in extended culture. Fertil Steril 2000; 74: 495–7. 44. Balaban B, Urman B, Alatas C et al. Blastocyst-stage transfer of poorquality cleavage-stage embryos results in higher implantation rates. Fertil Steril 2001; 75: 514–8. 45. Alikani M, Calderon G, Tomkin G et al. Cleavage anomalies in early human embryos and survival after prolonged culture in vitro. Hum Reprod 2000; 15: 2634–43. 46. Hardy K, Stark J, Winston RM. Maintenance of the inner cell mass in human blastocysts from fragmented embryos. Biol Reprod 2003; 68: 1165–9. 47. Nikas G, Ao A, Winston R, Handyside AH. Compaction and surface polarity in the human embryo in-vitro. Biol Reprod 1996; 55: 32–7. 48. Stone BA, Greene J, Vargyas JM et al. Embryo fragmentation as a determinant of blastocyst development in vitro and pregnancy outcomes following embryo transfer. Am J Obstet Gynecol 2005; 192: 2014–9. 49. Hardarson T, Lofman C, Coull G et al. Internalization of cellular fragments in a human embryo: time-lapse recordings. Reprod Biomed Online 2002; 5: 36–8. 50. Hnida C, Engenheiro E, Ziebe S. Computer-controlled, multilevel, morphometric analysis of blastomere size as biomarker of fragmentation and multinuclearity in human embryos. Hum Reprod 2004; 19: 288–93. 51. Johansson M, Hardarson T, Lundin K. There is a cutoff limit in diameter between a blastomere and a small anucleate fragment. J Assist Reprod Genet 2003; 20: 309–13. 52. Alikani M, Olivennes F, Cohen J. Microsurgical correction of partially degenerate mouse embryos promotes hatching and restores their viability. Hum Reprod 1993; 8: 1723–8. 53. Rienzi L, Nagy ZP, Ubaldi F et al. Laser-assisted removal of necrotic blastomeres from cryopreserved embryos that were partially damaged. Fertil Steril 2002; 77: 1196–201. 54. Nagy ZP, Taylor T, Elliott T et al. Removal of lysed blastomeres from frozen-thawed embryos improves implantation and pregnancy rates in frozen embryo transfer cycles. Fertil Steril 2005; 84: 1606–12. 55. Alikani M. The developmental ability of isolated mouse blastomeres following exposure to heterologous cytoplasmic fragments. In: On fragmentation: Origin and consequences of abnormal cell division in



56. 57.

58.

59.

60.

61.

62.

63. 64.

65.

66. 67.

68. 69. 70. 71.

72.

73. 74. 75.

human embryos in vitro; Alikani M, PhD Thesis. Monash University, Clayton, Australia, 2006. Tarkowski AK. Experiments on the development of isolated blastomers of mouse eggs. Nature 1959; 184: 1286–7. Dozortsev D, Ermilov A, El-Mowafi DM, Diamond M. The impact of cellular fragmentation induced experimentally at different stages of mouse preimplantation development. Hum Reprod 1998; 13: 1307–11. Alikani M, Schimmel T, Willadsen SM. Cytoplasmic fragmentation in activated eggs occurs in the cytokinetic phase of the cell cycle, in lieu of normal cytokinesis, and in response to cytoskeletal disorder. Mol Hum Reprod 2005; 11: 335–44. Cohen J, Alikani M, Ferrara T et al. Rescuing abnormally developing embryos by assisted hatching. In: Mori T, Aono T, Tominaga T, Hiroi M eds. Frontiers in Endocrinology, Perspectives on assisted reproduction. Rome: Ares Serono Symposia, 1994; 4: 536–44. Keltz MD, Skorupski JC, Bradley K, Stein D. Predictors of embryo fragmentation and outcome after fragment removal in in vitro fertilization. Fertil Steril 2006; 86: 321–4. Alikani M, Willadsen SM. Human blastocysts from aggregated mononucleated cells of two or more non-viable zygote-derived embryos. Reprod Biomed Online 2002; 5: 56–8. Alikani M, Munné S. Nonviable human pre-implantation embryos as a source of stem cells for research and potential therapy. Stem Cell Rev 2005; 1: 337–44. Alikani M. Epithelial cadherin distribution in abnormal human pre-implantation embryos. Hum Reprod 2005; 20: 3369–75. Johnson MH, Maro B. A dissection of the mechanisms generating and stabilizing polarity in mouse 8- and 16-cell blastomeres: the role of cytoskeletal elements. J Embryol Exp Morphol 1985; 90: 311–34. Vestweber D, Gossler A, Boller K, Kemler R. Expression and distribution of cell adhesion molecule uvomorulin in mouse preimplantation embryos. Dev Biol 1987; 124: 451–6. Perez GI, Tao XJ, Tilly JL. Fragmentation and death (a.k.a. apoptosis) of ovulated oocytes. Mol Hum Reprod 1999; 5: 414–20. Kerr JF, Wyllie AH, Currie AR. Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. Br J Cancer 1972; 26, 239–57. Savill J, Fadok V, Henson P et al. Phagocyte recognition of cells undergoing apoptosis. Immunol Today 1993; 14: 131–6. Hussein MR. Apoptosis in the ovary: molecular mechanisms. Hum Reprod Update 2005; 11: 162–77. Leist M, Nicotera P. The shape of cell death. Biochem Biophys Res Commun 1997; 236: 1–9. LaCasse EC, Holcik M, Korneluk RG et al. Apoptosis in health, disease, and therapy: overview and methodology. Apoptosis in Health and Disease; Clinical and Therapeutic Aspects. In: Holcik M, LaCasse EC, MacKenzie AE, Korneluk RG eds. London: Cambridge University Press, 2005: 1–28. Fabbri F, Carloni S, Brigliadori G et al. Sequential events of apoptosis involving docetaxel, a microtubule-interfering agent: a cytometric study. BMC Cell Biol 2006; 7: 6. Takase K, Ishikawa M, Hoshiai H. Apoptosis in the degeneration process of unfertilized mouse ova. Tohoku J Exp Med 1995; 175: 69–76. Fujino Y, Ozaki K, Yamamasu S et al. DNA fragmentation of oocytes in aged mice. Hum Reprod 1996; 11: 1480–3. Van Blerkom J, Davis PW. DNA strand breaks and phosphatidylserine redistribution in newly ovulated and cultured mouse and human oocytes: occurrence and relationship to apoptosis. Hum Reprod 1998; 13: 1317–24.

HPE_Chapter06.qxp

7/13/2007

4:47 PM

Page 77


76. Jurisicova A, Latham KE, Casper RF et al. Expression and regulation of genes associated with cell death during murine preimplantation embryo development. Mol Reprod Dev 1998; 51: 243–53. 77. Exley GE, Tang C, McElhinny AS, Warner CM. Expression of caspase and BCL-2 apoptotic family members in mouse preimplantation embryos. Mol Reprod Dev 1998; 51: 243–53. 78. Perez GI, Knudson CM, Leykin L et al. Apoptosis-associated signaling pathways are required for chemotherapy-mediated female germ cell destruction. Nat Med 1997; 3: 1228–32. 79. Bergeron L, Perez GI, Macdonald G et al. Defects in regulation of apoptosis in caspase-2-deficient mice. Genes Dev 1998; 12: 1304–14. 80. Souida AK, Tacka KA, Galvan KA et al. Immediate effects of anticancer drugs on mitochondrial oxygen consumption. Biochem Pharmacol 2003; 66: 977–87. 81. Hawes SM, Gie Chung Y, Latham KE. Genetic and epigenetic factors affecting blastomere fragmentation in two-cell stage mouse embryos. Biol Reprod 2001; 65: 1050–6. 82. Han Z, Chung YG, Gao S, Latham KE. Maternal factors controlling blastomere fragmentation in early mouse embryos. Biol Reprod 2005; 72: 612–8. 83. Kamjoo M, Brison DR, Kimber SJ. Apoptosis in the preimplantation mouse embryo: effect of strain difference and in vitro culture. Mol Reprod Dev 2002; 61: 67–77. 84. Levy R, Benchaib M, Cordonier H et al. Annexin V labelling and terminal transferase-mediated DNA end labelling (TUNEL) assay in human arrested embryos. Mol Hum Reprod 1998; 4: 775–83. 85. Hardy K, Handyside AH, Winston RM. The human blastocyst: cell number, death and allocation during late preimplantation development in vitro. Development 1989; 107: 597–604. 86. Jurisicova A, Acton BM. Deadly decisions: the role of genes regulating programmed cell death in human preimplantation embryo development. Reproduction 2004; 128: 281–91. 87. Warner CM, Cao W, Exley GE et al. Genetic regulation of egg and embryo survival. Hum Reprod 1998; 13(Suppl 3): 178–90. 88. Liu HC, He ZY, Mele CA et al. Expression of apoptosis-related genes in human oocytes and embryos. J Assist Reprod Genet 2000; 17: 521–33. 89. Spanos S, Rice S, Karagiannis P et al. Caspase activity and expression of cell death genes during development of human preimplantation embryos. Reproduction 2002; 124: 353–63. 90. Jurisicova A, Antenos M, Varmuza S et al. Expression of apoptosisrelated genes during human preimplantation embryo development: potential roles for the Harakiri gene product and Caspase-3 in blastomere fragmentation. Mol Hum Reprod 2003; 9: 133–41. 91. Metcalfe AD, Hunter HR, Bloor DJ et al. Expression of 11 members of the BCL-2 family of apoptosis regulatory molecules during

92.

93. 94. 95.

96.

97.

98. 99.

100.

101. 102. 103. 104. 105.

106.

human preimplantation embryo development and fragmentation. Mol Reprod Dev 2004; 68: 35–50. Van Blerkom J, Davis P, Alexander S. Differential mitochondrial distribution in human pronuclear embryos leads to disproportionate inheritance between blastomeres: relationship to microtubular organization, ATP content and competence. Hum Reprod 2000; 15: 2621–33. Majno G, Joris I. Apoptosis, oncosis, and necrosis. An overview of cell death. Am J Pathol 1995; 146: 3–15. Kim Sh SH, Kaminker P, Campisi J. Telomeres, aging and cancer: in search of a happy ending. Oncogene 2002; 21: 503–11. Liu L, Blasco M, Trimarchi J, Keefe D. An essential role for functional telomeres in mouse germ cells during fertilization and early development. Dev Biol 2002; 249: 74–84. Keefe DL, Franco S, Liu L et al. Telomere length predicts embryo fragmentation after in vitro fertilization in women – toward a telomere theory of reproductive aging in women. Am J Obstet Gynecol 2005; 192: 1256–60. Somma MP, Fasulo B, Cenci G et al. Molecular dissection of cytokinesis by RNA interference in Drosophila cultured cells. Mol Biol Cell 2002; 13: 2448–60. Rappaport R. Cytokinesis in Animal Cells. Cambridge, U.K.: Cambridge University Press, 1996. Liu L, Trimarchi JR, Smith PJ et al. Checkpoint for DNA integrity at the first mitosis after oocyte activation. Mol Reprod Devel 2002; 62: 277–88. Gordo AC, Rodrigues P, Kurokawa M et al. Intracellular calcium oscillations signal apoptosis rather than activation in in vitro aged mouse eggs. Biol Reprod 2002; 66: 1828–37. Perez GI, Trbovich AM, Gosden RG, Tilly JL. Mitochondria and the death of oocytes. Nature 2000; 403: 500–1. Sathananthan AH, Kola I, Osborne J et al. Centrioles in the beginning of human development. Proc Nat Acad Sci USA 1991; 88: 4806–10. Palermo GD, Munné S, Cohen J. The human zygote inherits its mitotic potential from the male gamete. Hum Reprod 1994; 9: 1220–5. Van Blerkom J. Sperm centrosome dysfunction: a possible new class of male factor infertility in the human. Mol Hum Reprod 1996; 2: 349–54. Schatten G, Simerly C, Schatten H. Microtubule configurations during fertilization, mitosis, and early development in the mouse and the requirement for egg microtubule-mediated motility during mammalian fertilization. Proc Nat Acad Sci USA 1985; 82: 4152–6. Chatzimeletiou K, Morrison EE, Prapas N et al. Spindle abnormalities in normally developing and arrested human preimplantation embryos in vitro identified by confocal laser scanning microscopy. Hum Reprod 2005; 20: 672–82.



HPE_Chapter06.qxp

7/13/2007

4:47 PM

Page 78

HPE_Chapter07.qxp

7/13/2007

4:47 PM

Page 79

7. Analysis of blastocyst morphology David K Gardner, John Stevens, Courtney B Sheehan, and William B Schoolcraft

INTRODUCTION

The development of detailed scoring systems for the pronucleate oocyte and cleavage stage human embryo have undoubtedly increased the efficacy of human IVF.1–9 The advent of more physiological culture media, together with improvements in embryo culture systems and the laboratory, has now made it possible to culture the human embryo to the viable blastocyst stage as a matter of routine.10–13 There appear to be numerous advantages associated with extended culture and blastocyst transfer (Table 7.1). The current status of clinical blastocyst transfer has recently been analyzed,30 with the conclusion that extended embryo culture results in an increase in implantation and clinical pregnancy rates and a reduction in pregnancy loss in several subpopulations, especially patients under the age of 3731,32 and oocyte donors,33 as well as those patients with poor quality embryos,34 or who have experienced multiple IVF failures.35 Furthermore, it appears that the blastocyst stage of development more readily maintains its viability postcryopreservation than the pronucleate oocyte or cleavage stage embryo, making it the logical stage at which to freeze or vitrify embryos.28,36,37 Taken together, blastocyst culture facilitates a reduction in the number of embryos transferred, while increasing the success of cryostorage, thereby increasing the overall efficacy of an IVF cycle.29

HOW CAN VIABLE EMBRYOS BE OBTAINED IN AN IVF LABORATORY?

Human embryos can be cultured to the blastocyst stage in a variety of conditions. However, the resultant viability of these blastocysts varies enormously. Therefore, it is imperative to discuss the development of viable blastocysts. The ability to grow a viable

embryo involves more than simply the acquisition of appropriate culture media. Rather, there are many variables that can have an impact on the outcome of an IVF cycle, all of which need to be taken into account in order to optimize pregnancy rates. Figure 7.1 highlights the variables associated with an IVF cycle. Time spent analyzing the variables in Figure 7.1 will not only assist in improving procedures, but will also assist in trouble shooting should a problem arise. Figure 7.1 serves to illustrate the complex and interdependent nature of human IVF treatment. For example, the stimulation regimen not only impacts oocyte quality (hence embryo physiology and viability),38 but can also affect subsequent endometrial receptivity.20,21,39,40 It is also apparent that a patient’s etiology and genetics will have an impact on their cycle outcome. Furthermore, the health and dietary status of the patient can have a profound effect on the subsequent developmental capacity of the oocyte and embryo. The dietary status of patients attending IVF is typically not considered as a compounding variable, but growing data would indicate otherwise.41–44

Table 7.1 Potential benefits of blastocyst transfer Embryo selection; ability to identify those embryos with limited, as well as those with the highest developmental potential14,15 Synchronization of embryonic stage with the female tract; reduces cellular stress on the embryo16,17 Minimize exposure of embryos to a hyperstimulated uterine environment18–21 Reduction in uterine contractions; reduces chance of embryos being expelled22,23 Ability to undertake cleavage stage embryo biopsy without the need for cryopreservation when the biopsied blastomere has to be sent to a different locale for analysis24 Assessment of true embryo viability; assessing the embryo post genome activation25,26 High implantation rates; reduces the need to transfer multiple embryos27 Increased potential to maintain viability after cryopreservation28 Increase in overall efficiency of IVF29



HPE_Chapter07.qxp

7/13/2007

4:48 PM

Page 80


Impact on endometrial receptivity

Etiology

Stimulation Oocyte quality

Patient Genetics

Laboratory

Uterus

Embryo transfer and luteal support

Outcome

Diet Number of embryologists & training level

Number of incubators

Air quality

QC & QA

Culture System

Tissue culture ware / contact supplies Culture media

Oil overlay

Gas phase (5% or 20% O2)

Number of embryos / drop

Figure 7.1 A holistic analysis of human IVF. In this schematic, the laboratory has been broken down into its core components, only one of which is the culture system. The culture system has in turn been broken down into its components, only one of which is the culture media. Therefore, it would appear rather simplistic to assume that the results of a given laboratory or clinic can be mimicked by changing only one part of the culture system (i.e. culture media). QC, quality control; QA, quality assurance.

A major determinant of the success of a laboratory and culture system is the level of quality control and quality assurance in place. For example, one should never assume that anything coming into the laboratory that has not been pretested with a relevant bioassay (e.g. mouse embryo assay) is safe merely because a previous lot has performed satisfactorily. Only a small percentage of the contact supplies and tissue culture ware used in IVF is supplied suitably tested. Therefore, it is essential to assume that everything entering the IVF laboratory without a suitable pretest is embryo toxic until proven otherwise. In our program the 1-cell mouse embryo assay (MEA) is employed to prescreen every lot of tissue culture ware that enters the program, i.e. plastics that are approved for tissue culture. Around 25% of all such material fails the 1-cell MEA (in a simple medium lacking protein after the first 24 h).45 Therefore, if one does not perform quality control (QC) to this level, one in four of all contact supplies used 

clinically could compromise embryo development. In reality many programs cannot allocate the resources required for this level of QC; when embryo quality is compromised in the laboratory, the culture media are held responsible, when in fact the tissue culture ware is more often the culprit.29,45 Several treatises exist that deal with the details of embryo culture media composition and their effects.12,46–51 In practical terms, there are several key points to address in order to obtain viable embryo development. These include the use of microdrops (20–50 ␮l) for embryo incubation, with a prescreened paraffin oil overlay. Oils should be pretested with a 1cell MEA performed in the absence of protein.45 Whenever possible, embryos should be cultured in groups of up to five and the culture media renewed every 48 h if they contain amino acids.29,45 A carbon dioxide concentration of 5–7% should be employed in order to ensure that the media have a working pH of between 7.25 and 7.35. Furthermore, the use of a

HPE_Chapter07.qxp

7/13/2007

4:48 PM

Page 81

ANALYSIS OF BLASTOCYST MORPHOLOGY

reduced oxygen concentration of around 5% is highly recommended (rather than ambient air which contains around 20%). Oxygen at a concentration of 20% not only has a negative effect on embryo development and viability,52–56 but also has detrimental effects on embryo gene expression.57

1.

2.


Once a laboratory has validated its culture system, what is the relationship between embryo morphology at the blastocyst stage and subsequent developmental potential post-transfer? Several different morphology grading systems have been created over the years.58–60 Such systems are typically based on the degree of blastocoel expansion and amount of necrosis that can be visualized. A more comprehensive system was therefore developed that accounted for both the development of the inner cell mass and the ophectoderm.61 This approach appears to be superior to simply assessing the degree of blastocoel expansion.62 The system developed by Gardner and Schoolcraft61 is represented in Figure 7.2. Figures 7.3 and 7.4 show photomicrographs of all stages and qualities of human blastocysts. The relationship between blastocyst morphology, using the grading system of Gardner and Schoolcraft, and IVF outcome has been published previously.64 An update is presented in Table 7.2, with data from 790 IVF (non-donor) cases in which patients had two blastocysts of documented morphology transferred. Clearly, those embryos with a good score at the blastocyst stage have high implantation potential (fetal heart beat/embryo transferred of 66%). It is therefore evident that single embryo transfer should be advocated for patients who have a blastocyst with a score of ⭓ 3AA on day 5. ANALYSIS OF BLASTOCYST PHYSIOLOGY

Although the blastocyst scoring system described and highlighted in Figures 7.2–7.4 has proved useful in embryo selection,62 there is still only a limited amount of data that can be derived from morphological assessment. Therefore, the development and implementation of non-invasive tests of embryo viability

3.

4.

ICM grading A. Tightly packed, many cells B. Loosely grouped, several cells C. Very few cells

Trophectoderm grading A. Many cells forming a cohesive epithelium B. Few cells forming a loose epithelium C. Very few large cells

Figure 7.2 Scoring system for human blastocysts. Blastocysts are initially given a numerical score from 1 to 6 based upon their degree of expansion and hatching status. (1) Early blastocyst, the blastocoel occupies less than half the volume of the embryo. (2) Blastocyst, the blastocoel occupies half of the volume of the embryo or more. (3) Full blastocyst, the blastocoel completely fills the embryo, but the zona is not thinned. (4) Expanded blastocyst, the blastocoel volume is now larger than that of the early embryo and the zona is thinning. (5) Hatching blastocyst, the trophectoderm has started to herniate through the zona. (6) Hatched blastocyst, the blastocyst has completely escaped from the zona. The initial phase of the assessment can be performed on a dissection microscope. The second step in scoring the blastocysts should be performed on an inverted microscope. For blastocysts graded as 3–6 (i.e. full blastocysts onwards) the development of the inner cell mass (ICM) and trophectoderm can then be assessed. Adapted from reference 61.

have significant merit.65 Figure 7.5 highlights some of the known and possible markers that have been linked 

HPE_Chapter07.qxp

7/13/2007

4:48 PM

Page 82


A

B

C

D

E

F

G

H

I

J

K

L

Figure 7.3 Photomicrographs of human blastocysts on day 5 of development. (A)–(C) These blastocysts are scored as 1; the embryos have begun to cavitate. The cavity (blastocoel) is less than half of the embryo’s volume. (D)–(F) These blastocysts are all scored as 2; the blastocoel within the embryos is equal to or greater than half the volume of the embryo. (G)–(L) Blastocysts that have a full blastocoel cavity, and are all graded as 3, but differ in their allocation of cells to the ICM and trophectoderm. (G) Grade 3CA, the ICM is poorly developed with few cells, the trophectoderm is well formed. (H) Grade 3AB, the ICM is well developed with many cells, the trophectoderm has fewer and larger cells than expected. (I) Grade 3BA, the ICM has not formed a tightly packed group, nor does it appear to have many cells, the trophectoderm is well formed. (J) Grade 3CC, neither the ICM or trophectoderm have many cells. The right hand side of the embryo consists of just 1 cell. The blastocoel, however, is fully formed, hence the score of 3. (K) Grade 3AC, the ICM is well formed and has many cells, but the trophectoderm has not formed a competent epithelium. (L) Grade 3AA, both ICM and trophectoderm have many cells forming a beautiful blastocyst. (M)–(U) Blastocysts that have begun to expand and thin their zona, and are graded as 4, but differ in their allocation of cells to the ICM and trophectoderm. (M) Grade 4CA, the ICM is hardly evident in this embryo, while the trophectoderm is clearly formed and functional. (N) Grade 4BB, although at first glance this blastocyst looks good, a critical analysis reveals that both the ICM and trophectoderm have fewer cells than are required to score AA. (O) Grade 4BC, similar to the blastocyst in



HPE_Chapter07.qxp

7/13/2007

4:48 PM

Page 83


M

N

O

P

Q

R

S

T

U

V

w

X

photomicrograph (N), this ICM is quite well formed, although there are too few cells for it to score an A. However, by looking at the right hand side of the embryo the curvature of the trophectoderm indicates that there are very few cells. (P)–(R) Grade 4AA, this series of images is of the same blastocyst at different focal planes in order to clearly show the ICM and trophectoderm. The ICM is tightly packed with many cells, and the surrounding cells have formed a cohesive epithelium of many cells. Such an embryo should be transferred individually. (S)–(U) Grade 4CA, this series of images is of the same blastocyst at different focal planes in order to clearly show the ICM and trophectoderm. At first glance, the blastocyst in photomicrograph (S) looks of excellent quality. However, when one focuses through the embryo it can be seen that the ICM has hardly developed at all. Although such an embryo is very likely to implant and give rise to a high hCG reading, by the time of scan there is little chance of observing a fetal heart. (V) Grade 4, pulsed. The blastocyst in this photomicrograph is a 4 as evident by the thinned zona. However, the true extent of ICM and trophectoderm differentiation cannot be assessed as the blastocoel has collapsed. This is a common occurrence in blastocysts, and the cavity will re-expand in the following 1–2 hours. Such an embryo should ideally be re-scored following re-expansion. (W) Grade 5BA, the trophectoderm has begun to herniate through the surrounding zona, hence the score of 5. However, the ICM is not tightly packed and only has a few cells. (X) Grade 5AA, the trophectoderm has begun to herniate through the surrounding zona, while in contrast to the blastocyst in (W), the ICM is well formed.



HPE_Chapter07.qxp

7/13/2007

4:48 PM

Page 84


A

B

C

Figure 7.4 Photomicrographs of human blastocysts following preimplantation genetic diagnosis (PGD) or freezing. (A) This embryo has undergone PGD on day 3 of development. The hole in the zona was formed using a laser, and the resultant blastocyst (4AA) is hatching through this hole. As a result it is not technically a true grade 5. Furthermore, after hatching on day 3, the zona does not thin (B) This blastocyst was a 4AA at freezing on day 5.63 This photomicrograph was taken 2 hours after thawing. (C) This blastocyst was a 4AA at freezing on day 6. This photomicrograph was taken 3 hours after thawing. Furthermore, after hatching on day 3 the zona does not thin.

Table 7.2 Effect of blastocyst score on pregnancy outcome in 790 IVF (non-donor) cases where two blastocysts were scored and transferred

No. of embryos transferred Mean age (⫾SE) (years) Age range (years) No. of transfers Patients with ICSI (%) Blastocyst development from 2PN (%) Implantation rate (% sacs) Implantation rate (% with FHT) Clinical pregnancy rate (%) Twin (%)

Group 1 (two blastocysts ⱖ3AA)

Group 2 (one blastocyst ⱖ3AA)

Group 3 (blastocyst ⬍3AA)

2 32.8 ⫾ 0.15 21–43 560 62.3 66.7 69.6 64.0 81.8 56.6

2 33.5 ⫾ 0.3 23–44 136 72.1 53.5 59.2 57.4 78.7 39.3

2 33 3 ⫾ 0.44 19–43 94 62.8 42.1a 35.6a 33.0a 50.0a 25.5a

PN, pronuclei; FHT, fetal heart tones. aSignificant differences between groups.

to embryo viability post-transfer. This is a tremendously exciting area in modern embryology, and kits will soon be available that can rapidly quantitate some of these parameters. In conclusion, with the development of more physiological culture conditions, it is now possible to culture human embryos to the blastocyst stage. Data to date indicate that embryo transfer at this later stage of development results in a higher implantation rate than the transfer of pronucleate oocytes or cleavage stage embryos,29,69 and a reduction in pregnancy 

loss.32 Furthermore, with the advent of an alphanumeric scoring system, it is possible to identify blastocysts with very high implantation potential (⬎60%).64 The ongoing development of non-invasive means of assessing embryo physiology will augment the available morphological scoring system, leading to further increases in implantation rate. As the benefits, to both mother and child, of reducing the number of embryos transferred are well documented,70 the day of single embryo, and in particular single blastocyst transfer, has arrived.

HPE_Chapter07.qxp

7/13/2007

4:48 PM

Page 85


Uptake

Production Lactate

Glucose

Ammonium

Pyruvate

Amino acids

Amino acids

Enzymes, e.g. LDH sHLA-G

Other sugars

HOXA10 regulator

Oxygen

PAF Novel peptides and proteins µl drop of defined culture medium Figure 7.5 Non-invasive analysis of human embryo viability. Individual blastocysts are incubated in 0.5–5.0 ␮l volumes of defined medium. Serial or endpoint samples of medium can then be removed for analysis.65 An indirect measurement of metabolic pathways, i.e. glycolysis and transamination can be obtained by measuring specific nutrients in combination, such as glucose uptake and lactate production, or amino acid turnover with ammonium production.66,67 As well as testing for known molecules, surface enhanced laser desorption/ionization time-of-flight mass spectrometry can also now be used to identify novel peptides and proteins.68 LDH, lactate dehydrogenase; sHLA-G, soluble histocompatibility antigen class 1 G; HOXA10, homeobox A10; PAF, platelet activating factor.

REFERENCES 1. Cummins JM, Breen TM, Harrison KL et al. A formula for scoring human embryo growth rates in in vitro fertilization: its value in predicting pregnancy and in comparison with visual estimates of embryo quality. J In Vitro Fert Embryo Transf 1986; 3: 284–95. 2. Balakier H, Cadesky K. The frequency and developmental capability of human embryos containing multinucleated blastomeres. Hum Reprod 1997; 12: 800–4. 3. Alikani M, Cohen J, Tomkin G et al. Human embryo fragmentation in vitro and its implications for pregnancy and implantation. Fertil Steril 1999; 71: 836–42. 4. Van Royen E, Mangelschots K, De Neubourg D et al. Characterization of a top quality embryo, a step towards single-embryo transfer. Hum Reprod 1999; 14: 2345–9. 5. Fisch JD, Sher G, Adamowicz M, Keskintepe L. The graduated embryo score predicts the outcome of assisted reproductive technologies better than a single day 3 evaluation and achieves results associated with blastocyst transfer from day 3 embryo transfer. Fertil Steril 2003; 80: 1352–8. 6. Ebner T, Moser M, Sommergruber M, Tews G. Selection based on morphological assessment of oocytes and embryos at different stages of preimplantation development: a review. Hum Reprod Update 2003; 9: 251–62. 7. Scott L. Pronuclear scoring as a predictor of embryo development. Reprod Biomed Online 2003; 6: 201–14. 8. Rienzi L, Ubaldi F, Iacobelli M et al. attributes of the early embryo. Reprod BioMed Online 2005; 10: 669–81.

9. Sakkas D, Gardner DK. Noninvasive methods to assess embryo quality. Curr Opin Obstet Gynecol 2005; 17: 283–8. 10. Gardner DK, Lane M. Culture and selection of viable blastocysts: a feasible proposition for human IVF? Hum Reprod Update 1997; 3: 367–82. 11. Menezo YJ, Hamamah S, Hazout A et al. Time to switch from co-culture to sequential defined media for transfer at the blastocyst stage. Hum Reprod 1998; 13: 2043–4. 12. Gardner DK, Lane M. Embryo culture systems. In: Trounson AO, Gardner DK, eds. Handbook of In Vitro Fertilization, 2nd edn. Boca Raton: CRC Press, 2000: 205–64. 13. Pool TB. Recent advances in the production of viable human embryos in vitro. Reprod Biomed Online 2002; 4: 294–302. 14. Tesarik J. Developmental failure during the preimplantation period of human embryogenesis. In: Van Blerkom J, editor. The biological basis of early human reproductive failure. New York: Oxford University Press, 1994: 327–44. 15. Van Blerkom J. Developmental failure in human reproduction associated with chromosomal abnormalities and cytoplasmic pathologies in meiotically mature oocytes. In: Van Blerkom J, editor. The Biological Basis of Early Human Reproductive Failure. New York: Oxford University Press, 1994: 283–326. 16. Croxatto HB, Ortiz ME, Diaz S et al. Studies on the duration of egg transport by the human oviduct. II. Ovum location at various intervals following luteinizing hormone peak. Am J Obstet Gynecol 1978; 132: 629–34. 17. Gardner DK, Pool TB, Lane M. Embryo nutrition and energy metabolism and its relationship to embryo growth, differentiation, and viability. Semin Reprod Med 2000; 18: 205–18.



HPE_Chapter07.qxp

7/13/2007

4:48 PM

Page 86


18. Simon C, Cano F, Valbuena D, Remohi J, Pellicer A. Clinical evidence for a detrimental effect on uterine receptivity of high serum oestradiol concentrations in high and normal responder patients. Hum Reprod 1995; 10: 2432–7. 19. Pellicer A, Valbuena D, Cano F, Remohi J, Simon C. Lower implantation rates in high responders: evidence for an altered endocrine milieu during the preimplantation period. Fertil Steril 1996; 65: 1190–5. 20. Ertzeid G, Storeng R. The impact of ovarian stimulation on implantation and fetal development in mice. Hum Reprod 2001; 16: 221–5. 21. Kelley RL, Kind KL, Lane M et al. Recombinant human follicle stimulating hormone alters maternal ovarian hormone concentrations and the uterus, and perturbs fetal development in mice. Am J Physiol Endocrinol Metab 2006; 291: E761–70. 22. Lesny P, Killick SR, Tetlow RL, Robinson J, Maguiness SD. Uterine junctional zone contractions during assisted reproduction cycles. Hum Reprod Update 1998; 4: 440–5. 23. Fanchin R, Ayoubi JM, Righini C et al. Uterine contractility decreases at the time of blastocyst transfers. Hum Reprod 2001; 16: 1115–9. 24. Munne S, Wells D. Preimplantation genetic diagnosis. Curr Opin Obstet Gynecol 2002; 14: 239–44. 25. Braude P, Bolton V, Moore S. Human gene expression first occurs between the four- and eight-cell stages of preimplantation development. Nature 1988; 332: 459–61. 26. Taylor DM, Ray PF, Ao A, Winston RM, Handyside AH. Paternal transcripts for glucose-6-phosphate dehydrogenase and adenosine deaminase are first detectable in the human preimplantation embryo at the three- to four-cell stage. Mol Reprod Dev 1997; 48: 442–8. 27. Gardner DK, Schoolcraft WB, Wagley L et al. A prospective randomized trial of blastocyst culture and transfer in in-vitro fertilization. Hum Reprod 1998; 13: 3434–40 28. Veeck LL. Does the developmental stage at freeze impact on clinical results post-thaw? Reprod Biomed Online 2003; 6: 367–74. 29. Gardner DK, Lane M. Towards a single embryo transfer. Reprod Biomed Online 2003; 6: 470–81. 30. Gardner DK, Balaban B. Choosing between day 3 and day 5 embryo transfers. Clin Obstet Gynecol 2006; 49: 85–92. 31. Papanikolaou EG, D’haeseleer E, Verheyen G et al. Live birth rate is significantly higher after blastocyst transfer than after cleavage-stage embryo transfer when at least four embryos are available on day 3 of embryo culture. A randomized prospective study. Hum Reprod 2005; 20: 3198–203. 32. Papanikolaou EG, Camus M, Fatemi HM et al. Early pregnancy loss is significantly higher after day 3 single embryo transfer than after day 5 single blastocyst transfer in GnRH antagonist stimulated IVF cycles. Reprod Biomed Online 2006; 12: 60–5. 33. Schoolcraft WB, Gardner DK. Blastocyst culture and transfer increases the efficiency of oocyte donation. Fertil Steril 2000; 74: 482–6. 34. Balaban B, Urman B, Alatas C et al. Blastocyst-stage transfer of poorquality cleavage-stage embryos results in higher implantation rates. Fertil Steril 2001; 75: 514–8. 35. Levitas E, Lunenfeld E, Har-Vardi I et al. Blastocyst-stage embryo transfer in patients who failed to conceive in three or more day 2–3 embryo transfer cycles: a prospective, randomized study. Fertil Steril 2004; 81: 567–71. 36. Anderson AR, Weikert ML, Crain JL. Determining the most optimal stage for embryo cryopreservation. Reprod Biomed Online 2004; 8: 207–11. 37. Takahashi K, Mukaida T, Goto T, Oka C. Perinatal outcome of blastocyst transfer with vitrification using cryoloop: a 4-year follow-up study. Fertil Steril 2005; 84: 88–92.



38. Hardy K, Robinson FM, Paraschos T et al. Normal development and metabolic activity of preimplantation embryos in vitro from patients with polycystic ovaries. Hum Reprod 1995; 10: 2125–35. 39. Simon C, Garcia Velasco JJ, Valbuena D et al. Increasing uterine receptivity by decreasing estradiol levels during the preimplantation period in high responders with the use of a follicle-stimulating hormone step-down regimen. Fertil Steril 1998; 70: 234–9. 40. Van der Auwera I, Pijnenborg R, Koninckx PR. The influence of invitro culture versus stimulated and untreated oviductal environment on mouse embryo development and implantation. Hum Reprod 1999; 14: 2570–4. 41. Kwong WY, Wild AE, Roberts P, Willis AC, Fleming TP. Maternal undernutrition during the preimplantation period of rat development causes blastocyst abnormalities and programming of postnatal hypertension. Development 2000; 127: 4195–202. 42. Armstrong DG, McEvoy TG, Baxter G et al. Effect of dietary energy and protein on bovine follicular dynamics and embryo production in vitro: associations with the ovarian insulin-like growth factor system. Biol Reprod 2001; 64: 1624–32. 43. Gardner DK, Stilley K, Lane M. High protein diet inhibits inner cell mass formation and increases apoptosis in mouse blastocysts developed in vivo by increasing the levels of ammonium in the reproductive tract. Reprod Fertil Dev 2004; 16: 190. 44. Gardner DK, Hewitt EA, Linck D. Diet affects embryo imprinting and fetal development. Hum Reprod 2004; 19(Suppl 1): i27. 45. Gardner DK, Reed L, Linck D, Sheehan C, Lane M. Quality control in human in vitro fertilization. Semin Reprod Med 2005; 23: 319–24. 46. Bavister BD. Culture of preimplantation embryos: facts and artifacts. Hum Reprod Update 1995; 1: 91–148. 47. Gardner DK. Changes in requirements and utilization of nutrients during mammalian preimplantation embryo development and their significance in embryo culture. Theriogenology 1998; 49: 83–102. 48. Gardner DK, Lane M. Development of viable mammalian embryos in vitro: evolution of sequential media. In: Cibelli J, Lanza RP, Campbell KHS, West MD, editors. Principles of Cloning. San Diego: Academic Press, 2002: 187–213. 49. Summers MC, Biggers JD. Chemically defined media and the culture of mammalian preimplantation embryos: historical perspective and current issues. Hum Reprod Update 2003; 9: 557–82. 50. Gardner DK, Lane M. Culture systems for the human embryo. In: Gardner DK, Weissman A, Holwes C, Shoham Z, editors. Textbook of assisted reproductive technology: laboratory and clinical perspectives, Second edition. London: Martin Dunitz Press, 2004: 211–34. 51. Lane M, Gardner DK. Understanding cellular disruptions during early embryo development that perturb viability and fetal development. Reprod Fertill Dev 2005; 17: 371–8. 52. Quinn P, Harlow GM. The effect of oxygen on the development of preimplantation mouse embryos in vitro. J Exp Zool 1978; 206: 73–80. 53. Harlow GM, Quinn P. Foetal and placental growth in the mouse after pre-implantation development in vitro under oxygen concentrations of 5 and 20%. Aust J Biol Sci 1979; 32: 363–9. 54. Thompson JG, Simpson AC, Pugh PA, Donnelly PE, Tervit HR. Effect of oxygen concentration on in-vitro development of preimplantation sheep and cattle embryos. J Reprod Fertil 1990; 89: 573–8. 55. Batt PA, Gardner DK, Cameron AW. Oxygen concentration and protein source affect the development of preimplantation goat embryos in vitro. Reprod Fertil Dev 1991; 3: 601–7. 56. Gardner DK, Lane M. Alleviation of the ‘2-cell block’ and development to the blastocyst of CF1 mouse embryos: role of amino acids, EDTA and physical parameters. Hum Reprod 1996; 11: 2703–12.

HPE_Chapter07.qxp

7/13/2007

4:48 PM

Page 87


57. Gardner DK, Lane M. Ex-vivo early embryo development and effects on gene expression and imprinting. Reprod Fertil Dev 2005; 17: 361–70. 58. Cohen J, Simons RF, Edwards RG, Fehilly CB, Fishel SB. Pregnancies following the frozen storage of expanding human blastocysts. J In Vitro Fert Embryo Transf 1985; 2: 59–64. 59. Dokras A, Sargent IL, Barlow DH. Human blastocyst grading: an indicator of developmental potential? Hum Reprod 1993; 8: 2119–27. 60. Kovacic B, Vlaisavljevic V, Reljic M, Cizek-Sajko M. Developmental capacity of different morphological types of day 5 human morulae and blastocysts. Reprod Biomed Online 2004; 8: 687–94. 61. Gardner DK, Schoolcraft WB. In-vitro culture of human blastocysts. In: Jansen R, Mortimer D, editors. Towards Reproductive Certainty: Fertility and Genetics Beyond 1999. Carnforth: Parthenon Publishing, 1999: 378–88. 62. Balaban B, Yakin K, Urman B. Randomized comparison of two different blastocyst grading systems. Fertil Steril 2006; 85: 559–63. 63. Gardner DK, Lane M, Stevens J, Schoolcraft WB. Changing the start temperature and cooling rate in a slow-freezing protocol increases human blastocyst viability. Fertil Steril 2003; 79: 407–10.

64. Gardner DK, Lane M, Stevens J, Schlenker T, Schoolcraft WB. Blastocyst score affects implantation and pregnancy outcome: towards a single blastocyst transfer. Fertil Steril 2000; 73: 1155–58. 65. Gardner DK, Leese HJ. Assessment of embryo metabolism and viability. In: Trounson A, Gardner DK, editors. Handbook of In Vitro Fertilization, Second Ed. Boca Raton: CRC Press, Inc., 2000: 347–72. 66. Lane M, Gardner DK. Selection of viable mouse blastocysts prior to transfer using a metabolic criterion. Hum Reprod 1996; 11:1975–8. 67. Gardner DK, Lane M, Stevens J, Schoolcraft WB. Noninvasive assessment of human embryo nutrient consumption as a measure of developmental potential. Fertil Steril 2001; 76: 1175–80. 68. Katz-Jaffe MG, Gardner DK, Schoolcraft WB. Proteomic analysis of individual human embryos to identify novel biomarkers of development and viability. Fertil Steril 2006; 85: 101–7. 69. Papanikolaou EG, Camus M, Kolibianakis EM et al. In vitro fertilization with single blastocyst-stage versus single cleavage-stage embryos. N Engl J Med 2006; 354: 1139–46. 70. Adashi EY, Barri PN, Berkowitz R et al. Infertility therapy-associated multiple pregnancies (births): an ongoing epidemic. Reprod Biomed Online 2003; 7: 515–42.



HPE_Chapter07.qxp

7/13/2007

4:48 PM

Page 88

HPE_Chapter08.qxp

7/18/2007

3:08 PM

Page 89

8. Morphometric analysis of human embryos Christina Hnida and Søren Ziebe

INTRODUCTION

Morphological evaluation of embryonic developmental competence prior to transfer has for many years been one of the key challenges in assisted reproduction. Many studies have shown that morphological structures in the embryo can be used as biomarkers of embryonic quality1–4 and that embryo selection based on morphology assessment is important to improve implantation and pregnancy rates.3,5–7Most of the existing scoring systems are based on combinations of several morphological parameters such as cleavage stage, embryonic fragmentation, and blastomere uniformity.1,3,7,8 However, in recent years assessing embryo quality has generally moved from scoring the embryos using a single grade to registering individual parameters considered to be important markers of embryo quality. This means that the embryo selection procedure is based on a balanced compromise after assessment of individual parameters. This move towards assessment of individual parameters has underlined the need for more objective measurements and more precise definitions. One example is equal blastomere size. When are the blastomere unequally sized? Some consider even the slightest difference as unequal size, and others demand larger differences for a classification of unequal size. Other examples that require more precise definition include degree of fragmentation and the size cutoff limit between a small blastomere and a large fragment. Morphometric analysis of embryos is a noninvasive method of characterizing the biological properties of the embryo based on measurements of individual morphological structures such as size of the nucleus or blastomere, or features of the zona pellucida. In contrast to biometric analysis, functional parameters such as paracrine production or

nutritional uptake are not included in a morphometric analysis. A constant drive to find the best markers of embryo competence and their incorporation into clinical embryology is very important. This should be focused not only on selecting the embryos with the highest developmental competence for transfer, but also towards improving stimulation regimens and culture conditions and thus the quality of the oocytes retrieved, of the embryos, and of the results after embryo transfer. This increased demand for more precise measurement has created a requirement to utilize computers and for adding a third dimension to the assessment. A number of new techniques are now emerging to assist the laboratory in the detailed analysis of embryos without compromising quality that would occur as a result of increased handling time outside the incubator. These techniques include multilevel digital recording of embryo images and new computer systems that allow detailed analysis of these pictures.

MONITORING AND MEASURING HUMAN EMBRYOS ADVANTAGES OF MORPHOMETRIC APPROACHES

Embryo evaluation in a clinical setting is a fine balance between a detailed registration of the individual parameters and returning the embryos to the incubator as quickly as possible. Therefore, in order to avoid compromising embryo quality, methods for embryo evaluation must be non-invasive and rapidly carried out. To date, embryo selection has been based mainly on analysis of morphology at the level of the



HPE_Chapter08.qxp

7/18/2007

3:08 PM

Page 90


light microscope, evaluating classical parameters considered as biomarkers of embryonic competence such as cell number, embryonic fragmentation, and blastomere uniformity.1,3,4,9 Embryonic structures can be measured with the use of a scale mounted within an inverted microscope. Using this technique, the size of embryonic structures can be read directly while the embryo is under the microscope. Alternatively, the microscopic image can be visualized on a computer screen with a calibrated scale, so that the image can be seen and measured by more than one observer. However, it is important to realize that embryo analysis under the light microscope is based on the subjective judgment of a single operator. Furthermore, time is a limiting factor that makes it difficult to investigate embryo morphology in great detail. Thus, the lack of more precise, objective, standardized and rapid methods remains a concern with regard to defining embryo quality, as well as being able to gain new and more detailed information about embryo morphology, and its correlation with developmental competence. Many of these limitations may be at least partly overcome by implementing computer assisted embryo assessment that allows detailed analysis based on highresolution digital images of the embryos. Additionally, monitoring the developing oocyte and embryo by time lapse recordings may further provide important information regarding the kinetics of the early human embryo. DIGITAL IMAGE ANALYSIS

SINGLE IMAGE SYSTEM

Morphometric analysis can be improved by recording and visualizing digital images of embryos, which minimizes handling time and allows more detailed analysis. For this purpose a digital camera must be mounted on the microscope, with appropriate software to support image recording. Some systems allow the computer to control image recording and storing as well as image analysis. Most of these systems allow morphometric analysis based on a single image of the embryo. In some applications,



selected morphological structures can be outlined and subsequently analyzed semiautomatically. Outlining the relevant structures allows values describing the size (diameter and area) of closed structures such as zona pellucida, oocytes, polar bodies, blastomeres, or nuclear structures to be retrieved. New parameters can be calculated automatically based on these values, including the size of enclosed spaces such as the perivitelline space. Furthermore, the number of nucleolar precursor bodies and nucleoli can be registered. However, analyzing only a single image is associated with some fundamental problems. Important information about the embryo may remain undetected, as morphological structures that appear outside the focal plane may not be correctly analyzed. This may present a problem particularly when analyzing the number of blastomeres and the degree of fragmentation. In addition, the number of nuclei in blastomeres outside the focal plane cannot be registered, and multinucleated embryos may remain undetected. Therefore, an optimal system requires the capacity to retrieve information from the whole embryo, in several focal planes. MULTILEVEL SYSTEM

The FertiMorph from IH-Medical, Denmark, is an example of a multilevel system. This system is equipped with a computer controlled, motorized stepper mounted on the microscope which will automatically focus through the embryo, producing a sequence of digital images. This enables an automatic and rapid recording in user-defined steps (Figure 8.1). The subsequent semiautomatic morphometric analysis is based on this sequence of images. All the images of one sequence can be viewed in detail, so that an image where the different morphological structures are in focus can be selected. Morphometric analysis of the embryo is thus based on information from several images. The morphological structures are outlined and subsequently analyzed in a similar manner to that described for the single image system (Figure 8.2). In addition to retrieving data on size and volume for all outlined structures, the system

HPE_Chapter08.qxp

7/18/2007

3:08 PM

Page 91

MORPHOMETRIC ANALYSIS OF HUMAN EMBRYOS

Figure 8.1 The multilevel system is equipped with a computer controlled, motorized stepper mounted on the microscope which automatically focuses through the embryo, producing a sequence of digital images of the embryo.

also provides information on their spatial positions inside the embryo. TIME-LAPSE MONITORING

It is important to realize that a general problem when analyzing embryo morphology and calculating morphometric information at a given time point is that it reflects a static situation. Such embryo assessment does not consider the kinetics and morphodynamics of the early embryo. Therefore, time-lapse

cinematography set-ups that combine culture systems and microscopes have been developed.10,11 Furthermore, such monitoring of development from the oocyte to the early embryo will help to elucidate the correct timing and morphology associated with several developmental key processes such as syngamy, the first cell cleavages, compaction, and blastocyst formation. The morphodynamics of embryonic structures such as the nucleolar precursor bodies, fragments, or blastomere size can also be monitored by time-lapse photography.10,11,41,42



HPE_Chapter08.qxp

7/18/2007

3:08 PM

Page 92


Figure 8.2 Embryonic structures such as blastomeres or nuclei are outlined in the particular images where the structure is in focus.

OOCYTE SIZE

MORPHOMETRIC ANALYSIS OF THE OOCYTE AND EARLY EMBRYO

Several studies have applied a variety of morphometric principles to the analysis of oocytes and embryos. However, very few studies have implemented information from the whole embryo. In the following section we will try to present the current status of all types of morphometric analysis on the parameters considered as important biomarkers of oocyte and embryo quality. 

Studies have shown that the diameter of the normal egg including the zona pellucida is approximately 150–160 ␮m, with a zona pellucida thickness of about 17–20 ␮m.12–14 The diameter of an unfertilized metaphase II oocyte, measured a short time after oocyte aspiration, is approximately 115 ␮m.12–14 An oocyte of abnormally small or large size may be an indicator of compromised biological and developmental competence of the egg. A study by Wolf et al14 showed abnormally low fertilization rates for oocytes

HPE_Chapter08.qxp

7/18/2007

3:08 PM

Page 93


with a diameter less than 108 ␮m. Abnormally large oocytes occur throughout the reproductive life of most women with a frequency between 0.2 and 1.03%.15 The development of giant oocytes during IVF treatment may be linked to an augmented response to gonadotropin therapy.15 Furthermore, most giant oocytes have been shown to be chromosomal abnormal.15,45 Based on measurements of oocyte diameter using a millimeter scale under the microscope, giant oocytes have been shown to be approximately 30% larger in diameter than normal oocytes, ranging from approximately 150 ␮m excluding the zona pellucida to 200 ␮m including the zona pellucida. ZYGOTE SIZE

Whether the size of the zygote differs from that of the oocyte remains uncertain. Wolf et al14 found a significant increase in oocyte diameter from day 0 to day 1. In contrast, Goyanes et al13 found that the volume of the zygote was reduced 10% compared with the oocyte. Based on the literature, the size of a morphologically normal human zygote is approximately 112 ␮m in diameter without the zona pellucida and approximately 150 ␮m including the zona pellucida. BLASTOMERE SIZE DURING CLEAVAGE STAGES

Previous studies have shown that total cytoplasmic volume remains constant from the zygote stage and during early embryonic development.13,16,17 In theory, this means that the volume of a blastomere at any given time will be 50% of that in the previous cell generation. This is, however, influenced by factors such as fragmentation, multinucleation, and synchrony and symmetry of cell cleavage (Figure 8.3). Cleavage stage as well as blastomere uniformity is involved in most of the existing embryo scoring systems.1,3,7,8 However, traditional scoring of blastomere uniformity is intuitive and subjective. Very little is known about the actual sizes and variations in blastomeres at different cleavage stages during early development, and precise and detailed knowledge

about normal blastomere size is necessary in order to identify deviations from optimal development. Additionally, minimum sizes for biologically competent blastomeres must be defined in order to distinguish between blastomeres and large fragments. Hnida et al18 used multilevel morphometric analysis to show that the average blastomere volume is approximately 0.3 ⫻ 106 ␮m3 in the 2-cell embryo and 0.15 ⫻ 106 ␮m3 in the 4-cell embryo. The corresponding blastomere diameters were 80 ␮m and 65 ␮m, respectively. This halving in cell volume during cleavage has been demonstrated right up to the 16-cell stage,13,16,17 and supports the general impression that total cytoplasmic volume remains constant during early embryonic development. However, despite the fact that minor asynchrony in the cleavage process is thought to be normal and commonly observed,16 the extent to which a cell cleavage can be asynchronous and yet normal is still unclear (Figure 8.3). Large variations in blastomere size have been demonstrated to be significantly associated with increased levels of chromosomal abnormalities in the embryos.19 THE IMPACT OF FRAGMENTATION ON BLASTOMERE SIZE

The correlation between increasing degree of fragmentation, decreasing embryo quality, and clinical outcome are well documented.1,3–8,20,44 Fragments originate from blastomeres, and they may therefore extract considerable amounts of cytoplasm; two studies have suggested that this may result in blastomeres with a quantitative deficiency of important cytoplasmic contents such as cell organelles, mRNA, or proteins.21,22 The question, therefore is whether the fragments themselves compromise embryo quality, or whether it is due to the reduction in blastomere size is associated with a high degree of fragmentation. Blastomeres in highly fragmented embryos may be too small to be biologically competent. Additionally, two studies10,11 have demonstrated that some fragments in some of the early embryos may disappear at later stages by resorption or lysis. However, fragmentation was still apparent during late cleavage stages in other embryos.10 These findings underline 

HPE_Chapter08.qxp

7/18/2007

3:08 PM

Page 94


1

1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

2

2

1

2

3 3 3 3 3 3 3 3

Theory

3 3 3 3 3 2 3

Light asynchrony

3 3 3 3 3 3 3 3

Asymmetry

2 1

2

3 3 Severe asynchrony

1

2 1

1 2

3 1

3 3

3

Figure 8.3 Examples of variation in blastomere size and cell generations.

the general assumption that fragments may have different etiologies and effects on embryo competence. Traditional analysis of fragmentation using light microscopy is highly subjective, producing only rough estimates16 and often with huge interobserver variations.16,19 A recent paper40 indicated that the use of digital images in embryo assessment may reduce this variation. The lack of objectivity and standardized methods of assessing fragmentation makes it difficult to analyze correlations between fragmentation and embryo quality in detail. However, as fragments by definition are anucleate structures of 

blastomeric origin, the degree of embryonic fragmentation should be reflected proportionally in the size of the blastomeres. This was demonstrated in a morphometric analysis by Hnida et al18 using the FertiMorph computer system for multilevel analysis based on recorded embryo sequences of a large consecutive cohort of 2-, 3-, or 4-cell embryos. The study showed that the mean blastomere volume decreased significantly with increasing degree of fragmentation (Figure 8.4). This decrease was correlated in a negative linear fashion for all analyzed cleavage stages. Two-cell embryos assessed as having more

HPE_Chapter08.qxp

7/18/2007

3:08 PM

Page 95

Blastomere volume (µm3 × 106)


0.4

2-cells 3-cells 4-cells

0.3 0.2 0.1 0

0

1–10 11–20 21–50 % Fragmentation

>50

Figure 8.4 Blastomere volume is significantly correlated to the degree of fragmentation at the early cleavage stages.

than 50% fragmentation showed a 67% reduction in mean blastomere volume compared to 2-cell embryos with no fragmentation. The corresponding reduction was 44% and 43% for the 3- and 4-cell embryos, respectively. The results from this study demonstrate the correlation between fragmentation and blastomere size.

Furthermore, the embryos were assessed for fragmentation by traditional evaluation. The outer border of the cell was outlined for all zygotes and for all embryos, and the outer border of all morphological structures considered to be blastomeres was also outlined. The mean reduction in cytoplasmic volume was found to increase significantly in a linear manner with increasing degree of fragmentation, as assessed by traditional evaluation. Comparing these findings to traditional fragmentation analysis showed that 27% of the embryos were allocated to a fragmentation category that did not correspond to the cytoplasmic reduction analysis (Figure 8.5). Overall, the data from this study suggest that the reduction in cytoplasmic volume from the zygote to the cleaved embryo stage might be used to quantify the degree of fragmentation in individual early embryos. This is in line with the finding by Goyanes et al13 showing that the total blastomeric volume is significantly decreased in fragmented embryos compared with the volume of their zygotes. However, correct differentiation between blastomeres and fragments is a key parameter when evaluating the degree

TOTAL CYTOPLASMIC VOLUME AS A NEW TECHNIQUE TO EVALUATE FRAGMENTATION

30 20 10

Cytoplasmic reduction (%)

As the total volume of cytoplasm remains constant during early embryonic development,13,16,17,23 it was suggested that the total cytoplasmic volume of the early embryo should equal the volume of the corresponding zygote. Thus, the degree of fragmentation equals the total reduction in cytoplasm calculated as the cytoplasmic volume of the zygote minus the combined volume of the blastomeres in the embryo. This method also corrects for inter-embryonic size variation, as the total volume of the blastomeres in each particular embryo is correlated to its own zygotic volume. Assessment of fragmentation based on such embryo-specific reduction in cytoplasmic volume would provide a more objective and standardized method. However, precise assessment of the zygote and blastomere sizes is a minimum requirement for successful application of this method Hnida et al23 used multilevel digital imaging system to take sequential images of all included zygotes, and of all their resulting day 2 embryos.

0 –10 –20 –30 –40 –50 –60 –70

0

1–10 11–20 % Fragmentation

21–50

>50

Figure 8.5 The mean reduction in total cytoplasmic volume is significantly correlated to degree of fragmentation, as assessed by traditional evaluation.



HPE_Chapter08.qxp

7/18/2007

3:08 PM

Page 96


of fragmentation based on blastomere sizes or reduction in cytoplasm. CELL SIZE CUT-OFF LIMITS

Minimum sizes for biologically competent blastomeres must be defined in order to distinguish between small blastomeres and large fragments. Based on the presence of DNA, Johansson et al22 suggested a cut-off size limit between blastomeres and fragments of 45 ␮m in diameter in day 2 embryos and 40 ␮m in day 3 embryos. However, this study did not differentiate between different cleavage stages observed on day 2 or day 3, respectively. Additionally, Hnida et al24 analyzed separate blastomeres and found that none of the analyzed 4-cell blastomeres smaller than 50 ␮m in diameter contained DNA, whereas 96% of the blastomeres with a diameter larger than 50 ␮m contained DNA. However, as the diameter of the blastomeres in the intact 4-cell embryos was approximately 3% smaller compared with the separate blastomeres, Hnida et al suggest a cut-off diameter of approximately 45–50 ␮m between blastomeres and fragments in 4-cell stage embryos.24 EMBRYOS WITH BLASTOMERES OF UNEVEN SIZE

The impact of unequal-sized blastomeres in an embryo has long been discussed, as this might be part of normal embryo development, particularly in dividing embryos (Figure 8.3). There is also no general definition regarding how large the difference must be in order for blastomeres to be classified as of unequal size. However, in recent years a number of studies have demonstrated that the developmental potential of embryos with blastomeres of unequal sizes is compromised. Studies have demonstrated6,30 that implantation and pregnancy rates were both significantly lowered after transfer of embryos with unevenly sized blastomeres. Two studies19,30 found a highly significant correlation between embryos with blastomeres having more than 25% difference in blastomere size and chromosomal abnormality. Further, the findings by Hardarson et al30 indicate that embryos with unevenly sized blastomeres have increased rates of multinucleation and that there is 

a link between blastomere size and multinucleation. This is supported by the findings of Hnida et al.18,24 One possible explanation for the impaired quality of these embryos could be that the presence of unequally sized blastomeres indicates that they have divided in an asynchronous or asymmetrical pattern, or that one or more of the cells have ceased to divide (Figure 8.3). However, the fact remains that we are currently at the stage of having only limited knowledge about the order of magnitude of difference between blastomere size that is required in order to compromise the embryo’s developmental potential. MULTINUCLEATION

Transfer of embryos with multinucleated blastomeres has been shown to be associated with decreased implantation, pregnancy, and birth rates.25–27 Furthermore, multinucleated embryos have increased rates of chromosomal abnormalities.28–30 Thus, multinucleated embryos should be excluded from transfer28,29 and assessment of nuclear status should be included in embryo scoring systems.25,26 Van Royen et al27 found multinucleation in 34% of a cohort of embryos from patients undergoing IVF or ICSI treatment. Hnida et al18 showed that the volume of multinucleated blastomeres was significantly larger than their mononucleated sibling blastomeres (22% in 2-cell and 30% in 4-cell embryos). These findings support other studies indicating that an intraembryonic variation in blastomere diameters of more than approximately 25% is associated with increased rates of multinucleation and chromosomal abnormalities.19,30 However, these studies did not measure the precise blastomere sizes. It has previously been suggested that multinucleated blastomeres can originate from an uncoupling of processes that control karyokinesis and cytokinesis, resulting in duplication of the nucleus without subsequent cell cleavage.17,31 The consequence of this would be a multinucleated blastomere without the size reduction from cell cleavage, thus retaining the size of the previous cell generation. These speculations are supported by the findings of Hnida et al18,24 demonstrating that a multinucleated 4-cell blastomere was approximately the same size as a non-multinucleated 2-cell blastomere (Figure 8.6).

HPE_Chapter08.qxp

7/18/2007

3:08 PM

Page 97


Blastomere volume (µm3 × 106)

In normal cleaving cells, there is a very close interaction and timing of the processes that control karyokinesis and cytokinesis.32,33 However, other mechanisms may also be involved in the formation of multinucleated blastomeres, including errors in chromosome migration at mitosis, incorrect packaging of chromosomes by the nuclear membrane after mitosis, or fragmentation of the nuclei.17,22,31 Additionally, mononucleated blastomeres originating from multinucleated embryos are on average smaller in size than the blastomeres from mononucleated embryos (Figure 8.7), and the frequency of anucleate blastomeres (by definition large fragments) is higher in multinucleated embryos.18,24 This indicates that asymmetric cell cleavage may also be associated with the occurrence of multinucleation.17,18,24 In conclusion, detection of nuclear status, especially in embryos that are otherwise of good morphology is of great importance in order to improve clinical outcome. However, assessment of multinucleation cannot be evaluated on the basis of blastomere size alone. Detection of embryos with unevenly sized blastomeres should be used in combination with visual verification of nuclear structures. Hnida et al24 analyzed the nuclear status in a cohort of embryos and showed that significantly more embryos were correctly categorized using the

0.4 Multinucleated blastomeres Mononucleated blastomeres

0.3 0.2 0.1 0

2-cells

3-cells

4-cells

Figure 8.6 It has previously been suggested that multinucleation may originate from an uncoupling of processes that control karyokineses and cytokinesis, resulting in duplication of the nucleus without subsequent cell cleavage. These speculations are supported by the findings that a multinucleated 4-cell blastomere is approximately the same size as a non-multinucleated 2-cell blastomere.

Figure 8.7 The volume of multinucleated blastomeres is significantly larger than their mononucleated sibling blastomeres. Further, mononucleated blastomeres from multinucleated embryos are smaller in size than the blastomeres from mononucleated embryos.

multilevel digital imaging system compared with traditional evaluation. SIZE OF NUCLEI

Despite the fact that nuclear : cell volume ratio is known to control the timing of events in early embryonic development in other species,34,35 very little is known about nuclear sizes in human embryos. Using morphometric multilevel measurements to assess the size of the nuclei in good quality mononucleated 2-cell embryos showed a diameter of 22.1 ␮m and a volume of 0.006 ⫻ 106 ␮m3. This decreased to a diameter of 18.7 ␮m and a volume of 0.003 ⫻ 106 ␮m3 in 4-cell embryos. These findings suggest a consistent nuclear : cell volume ratio of approximately 0.2 at least up to the 4-cell stage in human embryos.24 KINETICS OF EARLY EMBRYONIC DEVELOPMENT

Embryo assessment traditionally consists of scoring individual features such as cell number and fragmentation. However, it is important to bear in 

HPE_Chapter08.qxp

7/18/2007

3:08 PM

Page 98


mind that embryo development is a dynamic process and that the kinetics involved can yield additional information about embryo competence. A number of studies have demonstrated that the timing of cell cleavage is a significant indicator of embryonic competence. The embryo needs not only to develop to the 4-cell stage, but also it needs to do so at the correct time. Cleavage that occurs too rapidly or too slowly is an indication of impaired competence. Likewise, the onset of mitoses and the appearance/disappearance of the pronuclei after fertilization need to take place during a narrow time interval (22–25 hours) in high quality embryos, as suggested by Fancsovits et al.36 It has also been suggested that the interval between pronuclear breakdown and the first cleavage division should be relatively constant, about 3 hours.37,38 Other studies have demonstrated that the occurrence of early cleavage may be a good prognostic factor. However, the specific timing of early cleavage seems to be related to the method of fertilization, suggesting that different kinetics are involved in the processes of ICSI vs regular IVF.39

CONCLUSION

In the past, embryo evaluation has been based mainly on subjective evaluation of morphological parameters considered to be important markers of quality. However, a number of drawbacks are associated with this type of analysis. One example is the differentiation between large fragments and blastomeres, and another is imprecise estimation of the degree of fragmentation. The introduction of computer-based morphometric analysis has allowed us to enter a new level of embryo evaluation. These techniques open an array of possibilities for standardization and more precise measurements, including total cytoplasmic reduction as a new means of describing fragmentation, and detection of multinucleation based on blastomere size. In the final analysis, the combination of kinetics and morphometrics that include detailed information retrieved over several days is a new and fascinating aspect. The ‘3-dimensionality’ of multilevel 

analysis also presents many exciting possibilities. Although this technique is still young and needs further development, some very promising possibilities are already available today. REFERENCES 1. Puissant F, Van Rysselberge M, Barlow P et al. Embryo scoring as a prognostic tool in IVF treatment. Hum Reprod 1987; 2: 705–8. 2. Schulman A, Ben-Num I, Gethler Y et al. Relationship between embryo morphology and implantation rate after in vitro fertilization treatment in conception cycles. Fertil Steril 1993; 60: 123–6. 3. Giorgetti C, Terriou P, Auquier P et al. Embryo score to predict implantation after in-vitro fertilization: based on 957 single embryo transfers. Hum Reprod 1995; 10: 2427–31. 4. Van Royen E, Mangelschots K, De Neubourg D et al. Characterization of top quality embryo, a step towards single-embryo transfer. Hum Reprod 1999; 14: 2345–9. 5 Hill GA, Freeman M, Bastias MC et al. The influence of oocyte maturity and embryo quality on pregnancy rate in a program for in vitro fertilization-embryo transfer. Fertil Steril 1989; 52: 801–6. 6. Ziebe S, Petersen K, Lindenberg S et al. Embryo morphology or cleavage stage: how to select the best embryo for transfer after in vitro fertilization. Hum Reprod 1997; 12: 1545–9. 7. Erenus M, Zouves C, Rajamahendran P et al. The effect of embryo quality on subsequent pregnancy rates after in vitro fertilization. Fertil Steril 1991; 56: 707–10. 8. Steer CV, Mills CL, Tan SL, Campbell S and Edwards RG. The cumulative embryo score: a predictive embryo scoring technique to select the optimal number of embryos to transfer in an in-vitro fertilization and transfer programme. Hum Reprod 1992; 7: 117–19. 9. Van Royen E, Mangelschots K, De Noubourg D et al. Calculating the implantation potential of day 3 embryos in woman younger than 38 years of age: a new model. Hum Reprod 2001; 16: 326–32. 10. Van Blerkom J, Davis P and Alexander S. A microscopic and biochemical study of fragmentation phenotypes in stage-appropriate human embryos. Hum Reprod 2001; 16: 719–29. 11. Hardarson T, Lofman C, Coull G et al. Internalization of cellular fragments in a human embryo: time-lapse recordings. Reprod Biomed Online 2002; 5: 36–8. 12. Tsuji K, Sowa M and Nakano R. Relationship between human oocyte maturation and different follicular sizes. Biol Reprod 1985; 32: 413–17. 13. Goyanes VJ, Ron-Corzo A, Costas E and Maneiro E. Morphometric categorization of the human oocyte and early conceptus. Hum Reprod 1990; 5: 613–18. 14. Wolf JP, Bulwa S, Rodrigues D and Jouannet P. Human oocyte cytometry and fertilisation rate after subzonal insemination. Zygote 1995; 3: 101–9. 15. Balakier H and Cadesky K. The frequency and developmental capability of human embryos containing multinucleated blastomeres. Hum Reprod 1997; 12: 800–4. 16. Roux C, Joanne C, Agnani G et al. Morphometric parameters of living human in-vitro fertilization embryos; importance of asynchronous division process. Hum Reprod 1995; 10: 1201–7. 17. Hardy K, Winston RML and Handyside AH. Binucleate blastomeres in preimplantation human embryos in vitro: failure of cytokinesis during early cleavage. J Rep Fertil 1993; 98: 549–58. 18. Hnida C, Engenheiro E, Ziebe S. Computer controlled multi-level morphometric analysis of blastomere size as biomarker of fragmentation and multinuclearity in human embryos. Hum Reprod 2004; 19: 288–93.

HPE_Chapter08.qxp

7/18/2007

3:08 PM

Page 99


19. Ziebe S, Lundin K, Loft A for the CEMAS II and III Study Group et al. FISH analysis for chromosomes 13, 16, 18, 21, 22, X and Y in all blastomeres of IVF pre-embryos from 144 randomly selected donated human oocytes and impact on pre-embryo morphology. Hum Reprod 2003; 18: 2575–81. 20. Ebner T, Yaman C, Moser M et al. Embryo fragmentation in vitro and its impact on treatment pregnancy outcome. Fertil Steril 2001; 76: 281–5. 21. Alikani M, Cohen J, Tomkin G et al. Human embryo fragmentation in vitro and its implications for pregnancy and implantation. Fertil Steril 1999; 71: 836–42. 22. Johansson M, Hardarson T, Lundin K. There is a cutoff limit in diameter between a blastomere and a small anucleate fragment. J Assist Reprod Genet 2003; 20: 309–13. 23. Hnida C, and Ziebe S. Total cytoplasmic volume as biomarker of fragmentation in human embryos. J Assist Reprod Genet 2004; 20: 335–40. 24. Hnida C, Agerholm I and Ziebe S. Traditional detection versus computer-controlled multilevel analysis of nuclear structures from donated embryos. Hum Reprod 2005; 20: 665–71. 25. Jackson KV, Ginsburg ES, Hornstein MD, Rein MS and Clarke RN. Multinucleation in normal fertilized embryos is associated with an accelerated ovulation induction response and lower implantation rates in in vitro fertilization-embryo transfer cycles. Fertil Steril 1998; 70: 60–6. 26. Pelinck MJ, De Vos M, Dekens M et al. Embryos cultured in vitro with multinucleated blastomeres have poor implantation potential in human in-vitro fertilization and intracytoplasmic sperm injection. Hum Reprod 1998; 13: 960–3. 27. Van Royen E, Mangelschots K, Vercruyssen M et al. Multinucleation in cleavage stage embryos. Hum Reprod 2003; 18: 1062–9. 28. Kligman I, Benadiva C, Alikani M, and Munné S. The presence of multinucleated blastomeres in human embryos is correlated with chromosomal abnormalities. Hum Reprod 1996; 11: 1492–8. 29. Balakier H. and Cadesky K. The frequency and developmental capability of human embryos containing multinucleated blastomeres. Hum Reprod 1997; 12: 800–4. 30. Hardarson T, Hanson C, Sjögren A, and Lundin K. Human embryos with unevenly sized blastomeres have lower pregnancy and implantation rates: indications for aneuploidy and multinucleation. Hum Reprod 2001; 16: 313–18. 31. Pickering SJ, Taylor A, Johnson MH, and Braude PR. An analysis of multinucleated blastomere formation in human embryos. Hum Reprod 1995; 10: 1912–22.

32. Burke B. and Ellenberg J. Remodelling the walls of the nucleus. Nat Rev 2002; 3: 487–97. 33. Straight AF. and Field CM. Microtubules, membranes and cytokinesis. Curr Biol 2000; 10: 760–70. 34. Masui M, and Kominami T. Change in the adhesive properties of blastomeres during early cleavage stages in sea urchin embryo. Dev Growth Differ 2001; 43: 43–53. 35. Masui M, Yoneda M, and Kominami T. Nucleus : cell volume ratio directs the timing of increase in blastomere adhesiveness in starfish embryos. Dev Growth Differ 2001; 43: 295. 36. Fancsovits P, Toth L, Takacs Z.F et al. Early pronuclear breakdown is a good indicator of embryo quality and viability. Fertil Steril 2005; 84: 881–7. 37. Van Wissen B, Wolf JP, Bomsel-Helmreich O, Frydman R, and Jouannet P. Timing of pronuclear development and first cleavages in human embryos after subzonal insemination: influence of sperm phenotype. Hum Reprod 1995; 10: 642–848. 38. Capmany G, Taylor A, Braude PR, and Bolton VN. The timing of pronuclear formation, DNA synthesis and cleavage in the human 1cell embryo. Mol Hum Reprod 1996; 2: 299–306. 39. Van Montfoort APA, Dumoulin JCM, Kester ADM, and Evers JLH Early cleavage is a valuable addition to existing embryo selection parameters: a study using single embryo transfers. Hum Reprod 2004; 9: 2103–8. 40. Arce J-C, Ziebe S, Lundin K et al. Interobserver agreement and intraobserver reproducibility of embryo quality assessments. Hum Reprod 2006; 21: 2141–8. 41. Lehtonen E, et al. Changes in cell dimensions and intercellular contacts during cleavage-stage cell cycles in mouse embryonic cells. J Embryol Exp Morphol 1980; 58: 231–9. 42. Massip A. and Mulnard J. Time-lapse cinematographic analysis of hatching of normal and frozen-thawed cow blastocysts. J Reprod Fertil 1980; 58: 475–8. 43. Aiken CEM, Swoboda PPL, Skepper JN. and Johnson MH. The direct measurement of embryonic volume and nucleo-cytoplasmic ratio during mouse pre-implantation development. Reproduction 2004; 128: 527–35. 44. Alikani M, Cohen J, Tomkin G et al. Human embryo fragmentation in vitro and its implications for pregnancy and implantation. Fertil Steril 1999; 71: 836–42 45. Munne S, Alikani M. and Cohen J. Monospermic polyploidy and atypical embryo morphology. Hum Reprod 1994; 9: 506–10.



HPE_Chapter08.qxp

7/18/2007

3:08 PM

Page 100

HPE_Chapter09.qxp

7/13/2007

4:49 PM

Page 101

9. Development rate, cumulative scoring, and embryonic viability Christine C Skiadas and Catherine Racowsky

INTRODUCTION

Since the inception of clinical in vitro fertilization (IVF), there has been a drive to optimize pregnancy rates. Although this was initially achieved by transferring greater numbers of embryos, transfers of multiple embryos resulted in the negative side-effect of increasing the incidence of high-order multiple gestations.1–4 In order to reduce high-order multiple gestations and at the same time maintain pregnancy rates, there has been a progressive move toward decreasing the number of day 3 embryos transferred,5 as well as performing day 5 transfers of only one or two blastocysts.6–8 However, blastocyst transfer may not be ideal in all cases, and may compromise a successful outcome that would otherwise be achieved following a day 3 transfer.8–10 Therefore, one of the most important challenges in IVF is the ability to determine which embryos are associated with the greatest developmental potential, in order to select optimally only one, or at most two, of these embryos for transfer. Ideal methods for embryo selection include: ease of assessment, standardization among embryologists, minimal harm to the embryo, and a high correlation with pregnancy rates; therefore morphological assessment remains the first-line approach, although noninvasive biomarker methods are currently under development (Chapters 12 and 20). Since human embryonic development follows a specifically timed, coordinated sequence of events, developmental rate (assessed by certain milestones being reached at particular points in time) and morphological characteristics (defined at specified intervals after the day of insemination) provide the two main measures of embryonic development. Although morphological selection of embryos represents the current

standard of care, even the most rigorous selection paradigms have limitations, including the inability to detect genetic disorders or predict pregnancy with 100% accuracy. Other embryonic markers of development and metabolic assessment are currently being explored and these may, in the future, be used alone or in combination with morphological evaluations for improved embryo selection. This chapter reviews the normal timeline and sequence of embryonic development from fertilization through progression to the blastocyst stage. We also review the key features of morphological criteria for embryo selection, cumulative grading systems and their association with implantation rates and viability. Finally, we consider issues surrounding the optimum day for embryo transfer.

DEVELOPMENTAL RATE: NORMAL TIMELINE OF EVENTS

Preimplantation development follows a programmed timeline during which an organized series of critical events take place (Figure 9.1). In vivo, fertilization and early cleavage occur in the fallopian tube, with the embryo traversing the uterotubal junction at the morula stage. The procedure of clinical IVF has allowed this timeline of events to be observed in detail and researched. In terms of defining embryo morphology, one of the first time points that has been evaluated is that of the zygote or pronuclear embryo,12–15 approximately 16–18 hours after fertilization. Key features of the zygote stage are the development of the two pronuclei (one from the oocyte and one from the sperm), each containing multiple nuclear precursor bodies. The two pronuclei (PNs) migrate towards each other and their 

HPE_Chapter09.qxp

7/13/2007

4:49 PM

Page 102


Figure 9.1 In vivo embryo maturation. Natural timeline of embryonic development in vivo. Early cleavage stages occur in the fallopian tube and the embryo enters the uterus once it has reached the blastocyst stage. Reprinted from Figure 2-24, Moore and Persaud, The Developing Human, 6 edn. Philadelphia: WB Sanders, 1998: 44, Copyright 1998, with permission from Elsevier.11

multiple nucleoli align at the pronuclear interface in preparation for syngamy.16,17 Specific features of the PNs have been used in zygote scoring systems, which are discussed in Chapters 3 and 4. Following syngamy, the newly formed zygote undergoes first cleavage between 20 and 27 hours after insemination,18–20 and second cleavage to form the 4-cell stage at approximately 48 hours. The embryo reaches the 8-cell stage by approximately 

72 hours. Early cycles of cell division are thought to be regulated by the maternal genome, with the embryonic genome becoming activated to dictate further cell divisions at approximately day 3, between the 4- and 8-cell stage.21 Following the 8-cell stage, the cells become increasingly polarized and the embryo develops cell–cell adhesions and gap junctions during the process of compaction. This is expected to occur on day 4,22 with progression to

HPE_Chapter09.qxp

7/13/2007

4:49 PM

Page 103

DEVELOPMENT RATE, CUMULATIVE SCORING, AND EMBRYONIC VIABILITY

blastocyst development early on day 5 and completion of blastulation by late day 5.23 Figure 9.2 shows representative images of embryos at these stages.

EMBRYONIC MORPHOLOGY AND SCORING SYSTEMS

The optimal timing and method of morphological assessment has been debated amongst embryologists, and has been affected by laws governing embryo transfer and cryopreservation. ZYGOTE STAGE

Zygote scoring systems allow embryos of optimal prognosis to be identified immediately after fertilization, which has been particularly useful in countries where embryo selection at cleavage stages is restricted. In Germany, the German embryo protection law

A

B

C

D

Figure 9.2 Representative embryo images. (A) Pronuclear stage embryo. In this photograph, the two pronuclei are aligned in the middle of the zygote with their nucleoli aligned at the pronuclear interface. (B) 2-Cell embryo. (C) 4-Cell embryo. (D) 8-Cell embryo. In both the 4-cell and 8-cell embryo, blastomeres appear symmetrical and display no fragmentation.

dictates that the number of embryos to be transferred must be selected at the PN stage, with the remainder being cryopreserved. However, despite the positive results and practical benefits of PN scoring, there remains debate over whether such evaluation is superior to that of standard morphological assessment. Indeed, a recent study24 investigated whether PN scoring was superior to standard day 2 or day 3 morphological assessment; the results failed to detect a difference in pregnancy rates based on the scoring system. Although this study was underpowered to conclude a negative result, it suggests that the ideal method of scoring has yet to be determined. Zygote scoring systems, and their physiological basis, are covered in detail in Chapters 3 and 4. TWO CELL EMBRYOS

Although very few studies address the morphological features of the 2-cell embryo, the time to first cell division has been extensively studied as a predictor of improved pregnancy outcomes. Embryos that undergo ‘early cleavage’ have been postulated to have a greater degree of developmental competence than embryos that do not undergo early cleavage.25 In 1997, Shoukir et al designated those embryos that had reached the 2-cell stage by 25 hours postinsemination as having undergone ‘early cleavage,’ and this occurred in 19% of cycles (see Figure 9.3). Cycles with early cleavage were associated with significantly higher pregnancy rates.26 As the timing of fertilization is often unknown with conventional insemination, a follow-up study using only ICSI cycles again confirmed that those embryos undergoing early cleavage were associated with a significantly higher pregnancy rate, supporting the idea that early cleavage is related to developmental competence and not to the timing of fertilization.27 Tsai et al reproduced these findings in a retrospective study3 and Sakkas et al confirmed their original findings with a prospective study performed in 2001, where the presence of an increased number of early cleaving embryos was again associated with increased implantation rate.28 This study assessed early cleavage on alternate weeks to determine if selecting embryos on this basis had an impact on implantation rates. 

HPE_Chapter09.qxp

7/13/2007

4:49 PM

Page 104


Transfer Insemination

25 h after insemination

Early cleavage

43–47 h after insemination

4-cells

2-cells

No distinction between embryo stages No early cleavage

2-cells

4-cells

Figure 9.3 Timing of assessment of early cleavage. The timeline for assessing early cleavage is crucial, as there is only a certain window where true early cleavage can be seen. This earlier time point provides differentiation between embryos that may have similar morphology once they reach the 4-cell stage. Reproduced from Shoukir Y et al. Early cleavage of in-vitro fertilized embryos to the 2-cell stage: a novel indicator of embryo quality and viability. Hum Reprod 1997; 12(7): 1531–6. © European Society of Human Reproduction and Embryology, with permission from Oxford University Press/Human Reproduction.26

Of note, there was no difference in the total number of embryos between groups. In the group where early cleavage was checked, the pregnancy rate was 48%, compared with 31% in the group where this was not checked.28 The fact that checking for early cleavage did improve the pregnancy rate suggests that this may be a feature of improved developmental competence. In a retrospective analysis, Ciray et al showed an association between early cleavage, a higher day 3 embryo quality score and increased implantation rates, demonstrating that there is a link between early cleavage and improved embryo morphology at the 8-cell stage.29 Therefore, if early cleavage is another surrogate marker for improved day 3 morphology and implantation, the question arises as to whether it is necessary to evaluate the embryo at both stages. A further discussion of multiday embryo assessment is undertaken later in this chapter. FOUR-CELL EMBRYOS (DAY 2)

The embryo normally reaches the 4-cell stage on day 2 after insemination, and many of the embryo transfers were carried out on day 2 in the earliest reports of IVF. The features of cell number, degree 

of fragmentation, and equal size of blastomeres have been frequently evaluated at this stage to determine viability potential. Cummins et al and Puissant et al were two of the earliest authors to describe such scoring systems.1,2 Ideal 4-cell embryos are those with equal sized blastomeres and minimal fragmentation. Ebner et al also confirmed that day 2 embryos displaying little fragmentation showed an improved clinical pregnancy rate.30 Even in these early scoring systems, developmental rate was recognized as an additional marker of improved implantation. Cummins et al reported an ideal embryo developmental rate (EDR) associated with implantation;2 Puissant et al and Steer et al both used embryo cell number as an approximate marker for achieving a certain developmental state, with additional points awarded if the embryo had reached certain milestones.1,31 Table 9.1 summarizes studies that focus on day 2 embryo scoring.1,2, 31–33 Further discussion of the individual morphological features (cell number, fragmentation, symmetry, and compaction) is discussed in the section below on 8-cell embryos. The nuclear status of the blastomeres and the presence of mononucleation provide an additional means of assessment in the 4-cell embryo (for additional

EQ Scoring system (Grade 1–4) assessed regularity or symmetry of blastomeres, presence of fragments, and quality of cytoplasm. Ideal being grade 4 embryos: regular blastomeres, no fragments and clear cytoplasm (no granularity)

Did not choose embryos to transfer based on scoring system Both optimal EDR and EQ were significantly associated with pregnancy rates

Day 2 features

Day of transfer

NA, not applicable.

Outcome: pregnancy rates

‘Good embryos’ scored 5 or 6 and the number of ‘good embryos’ transferred was significantly associated with pregnancy rates

Pregnancy rates rose as CES rose to 42. Above this level, there was no further increase in pregnancy rates, but only an increase in multiple gestation rates

Day 2

CES was calculated by multiplying grade of embryo ⫻ cell number and then summated the scores of all embryos transferred Grade 4: equal sized blastomeres Grade 3: uneven blastomeres with ⬍10% fragmentation Grade 2: 10–50% fragmentation Grade 1: ⬎50% fragmentation

Day 2

The Cumulative Embryo Score (CES) Retrospective analysis

Steer et al, 199231

The odds ratio of AMS was 1.63 (0.990–2.7). Probability of pregnancy increased by 63% when the AMS increased by one point. However, this odds ratio crosses one. The AMS was included in the model, as it improved the goodness of fit of the model

Day 2

Number of blastomeres, symmetry of blastomeres, percentage of extracellular fragmentation. Symmetrical blastomeres ⫽ 2 points. Embryo morphology score ⫽ symmetry score/ (1 ⫹ % fragmentation). Average morphology score ⫽ embryo morphology score/# transferred embryos

NA

Average Morphology Score (AMS) Retrospective multivariate logistic regression Day 2

Roseboom et al, 199532

Pregnancy rate increased with each additional point (increase in pregnancy rate approx 4% per point)

Day 2

A point scale was established, which assigned a point to each feature: 1) Achieved cleavage state 2) No fragmentation, or ⬍20% fragmentation 3) No irregular blastomeres 4) 4-cell stage

NA

Retrospective analysis of single embryo transfers Day 2

Giorgetti et al, 199533

4:49 PM

Grade 4: embryos with clear, regular blastomeres with no fragmentation or a maximum of 5 small anucleate fragments Grade 3: embryos with unequal blastomeres, few or no fragments Grade 2: fragments ⬍1/3 of embryo surface Grade 1: fragments over ⬎1/3 of embryo surface Day 2

An additional two points were added if the embryo had reached the 4-cell stage by 48 hours

Day 2

Retrospective analysis

Puissant et al, 19871

7/13/2007

Assessment of developmental timing

Multiple assessments of growth rate. Grading performed on day of transfer EDR: based on growth rates of high EQ embryos. Calculated a ratio of time observed/time expected to predict optimal rate of development

Embryo Quality (EQ) Embryo Development Rating (EDR) Retrospective analysis

Timing of embryo assessment

Type of study

Title of study

Cummins et al, 19862

Table 9.1 Compilation of scoring systems assessing day 2 morphological features

HPE_Chapter09.qxp Page 105




HPE_Chapter09.qxp

7/13/2007

4:49 PM

Page 106


review, see Chapter 5). Embryos with multinucleated blastomeres have been associated with a higher rate of chromosomal abnormalities34 and lower implantation rates, when compared with embryos with mononucleated blastomeres.35 Saldeen et al added the number of mononucleated blastomeres to traditional morphological features of cell number, fragmentation, and equivalence in the size of blastomeres in order to select embryos. The ideal embryo for transfer contained four mononucleated blastomeres, and embryos with multinucleated blastomeres were excluded from transfer.36 Interestingly, although standard morphological features were used to guide embryo grading, mononucleation of all four blastomeres was the only morphological sign that was associated with implantation in this study.36 A shift to transferring embryos on day 3 led to a further opportunity for evaluation, and this led to a refinement in criteria for assessing embryo morphology, while maintaining the original features of day 2 embryo assessment.

symmetry. Each of these features has been shown to correlate with implantation and pregnancy rates after day 3 transfer.37–39 More recent attention has also been given to early compaction as a further predictor of developmental competence.40,41 We briefly review some of the studies that have analyzed the individual elements of morphology as they relate to implantation potential. CELL NUMBER

Several studies have shown that a proportional relationship exists between the number of blastomeres on day 3 (up to 8 cells) and implantation rates following day 3 transfer.1,31 When specifically evaluating the day 3 embryo, Carillo et al demonstrated that embryos with ⱖ8 cells on day 3 resulted in significantly higher pregnancy rates when compared with embryos with ⬍8 cells,37 and Racowsky et al demonstrated that those embryos with exactly 8 cells on day 3 had the highest implantation rates (see Figures 9.4).39 Some authors have also used progression to the blastocyst stage as a surrogate marker of developmental competence, and these studies have also revealed a relationship between the number of cells

EIGHT-CELL EMBRYOS (DAY 3)

Evaluation of day 3 embryos typically focuses on cell number, degree of fragmentation, and blastomere

40

40

% Viable

30

30

c d

20

40 30

a

20

20

10

a

0 n = 512 8

0

1–9

10–25

>25

None

Some

Severe

Cell number

% Fragmentation

Asymmetry

Figure 9.4 Association of individual day 3 features (cell number, fragmentation, and asymmetry) with implantation rates. Embryo viability according to morphological parameters: cell numbers (left panel), degree of fragmentation (middle panel), and asymmetry (right panel). Total numbers (N) of embryos analyzed in each group are indicated. Different letters denote significant differences: (left panel) a vs b, p ⬍0.0001; b vs c, p ⬍0.0001; c vs d, p ⫽ 0.016; (middle panel) a vs b, p ⬍0.0001; b vs c, p ⬍0.0001; (right panel) a vs b, p ⫽ 0.042; b vs c, p ⬍0.0001. Reprinted from figure 2, Racowsky et al. Day 3 and Day 5 Morphological predictors of embryo viability. Reprod Biomed Online 2003; 6(3): 323–331. With permission from Reproductive Healthcare Ltd.



HPE_Chapter09.qxp

7/13/2007

4:49 PM

Page 107


on day 3 and blastocyst conversion,23 suggesting that an optimal number of blastomeres on day 3 is a key developmental feature. Alikani et al demonstrated that embryos with 7–9 cells on day 3 have a significantly higher blastocyst conversion rate, compared with day 3 embryos with ⬍7 cells or ⬎9 cells.42 The fact that embryos with exactly 8 cells have both higher implantation rates and blastocyst conversion rates points to an interaction between morphological assessment and developmental timing. The idea that there is an ideal rate of development for embryonic growth has been postulated since the early studies of embryo assessment2 and as we learn more about developmental progression, embryo grading may focus more on overall growth and multiday assessment rather than the static evaluation of day 3 blastomere number.

percentage fragmentation.39 For additional review, see Rienzi et al, 2005.44 COMPACTION

Blastomere fragmentation in embryos has been extensively described, but the exact significance and cause of fragmentation remains a topic of debate. Several types and degrees of fragmentation have been identified, with varying impact on implantation and pregnancy rates, as described in detail in Chapter 6.

Few studies have focused on the degree of embryo compaction, or the utility of assessing early compaction (i.e. the presence of compaction on day 3). Tao et al devised a grading system for the day 4 embryo, when compaction is typically initiated,22 and concluded that the degree of compaction on day 4 was associated with implantation potential.45 Desai et al also used compaction grading as part of a combined embryo grading score on day 3. In this scoring system, the pregnancy rate was found to increase with the transfer of a compacting embryo, but this observation did not reach statistical significance.40 We have recently shown that early compaction is related to implantation potential, but the degree of fragmentation has an influence on this effect. In embryos (ⱖ8 cells) displaying ⬍10% fragmentation, early compaction is associated with a significantly higher implantation rate, whereas, in embryos with ⱖ10% fragmentation, early compaction is negatively associated with implantation potential.41

SYMMETRY

COMBINING DAY 3 CHARACTERISTICS

Blastomere symmetry has been included in the majority of standard systems that assess both day 2 and day 3 morphology (see Table 9.1), but it has infrequently been evaluated as an independent marker of implantation potential. Generally, asymmetry has been thought to be less important than either cell number or fragmentation; however, embryos with marked cellular asymmetry have demonstrated substantially reduced implantation rates. Hardarson reported that embryos displaying uneven cleavage had lower implantation rates than embryos with even cleavage (23.9% vs 36.4%).43 Racowsky et al also described that the implantation rates significantly decreased as the degree of asymmetry increased (no asymmetry 22.4%, some asymmetry 13.3%, and severe 1.4%). The detrimental effects of asymmetry were seen even after controlling for cell number and

Most embryologists have tried to weigh these individual factors in a combined manner in order to determine the ‘best’ embryo for transfer. Several studies have evaluated the implantation potential of these ‘top quality’ embryos. Volpes et al determined that an increased number of embryos of ‘good quality’ (defined as 8-cell stage with ⬍20% fragmentation) were associated with increased pregnancy rates. In addition, embryos transferred from cohorts having at least one good quality embryo resulted in significantly greater pregnancy rates than those arising from cohorts with zero good quality embryos.38 Van Royen et al, also attempted to identify ‘top quality embryos,’ defined as: absence of multinucleated blastomeres, presence of 4 or 5 blastomeres on day 2 and 7 or more cells on day 3, with ⱕ20% fragmentation. In this study, transfer of two top quality

FRAGMENTATION



HPE_Chapter09.qxp

7/13/2007

4:49 PM

Page 108


embryos was significantly more likely to result in a twin pregnancy, and the presence of a top quality embryo was associated with a higher implantation rate.46 Clearly, there are interactions between these features of morphology. Racowsky et al investigated the relative weight of cell number, fragmentation, and degree of asymmetry, proposing that the optimal embryo for day 3 transfer is an 8-cell embryo with ⬍10% fragmentation and no asymmetry; these embryos were associated with the highest percentage viability.39 Desai et al expanded the number of features involved in day 3 morphological grading and developed a day 3 embryo quality score (D3EQ) that assessed: cell number, presence of equal sized blastomeres, blastomere expansion (blastomeres touching the zona with minimal perivitelline space), cellular cytoplasm clear of vacuoles, presence of cytoplasmic pitting, signs of compaction, and the pattern of fragmentation.40 Using the D3EQ scores of the two highest scoring embryos transferred, the scoring system had a sensitivity of 83% for predicting pregnant versus non-pregnant cycles.

CRITICISMS OF DAY 3 MORPHOLOGICAL GRADING

Despite the improved implantation rates seen with certain features of day 3 embryo selection, there have been criticisms of morphological grading. Some authors have argued that day 3 embryo quality is a poor predictor for progression to the blastocyst stage or for prediction of blastocyst quality.47,48 Furthermore, experienced embryologists were only able to pick the embryos that would develop into the two best blastocysts in 23% of cycles.48 Finally, normal morphology does not guarantee euploidy or a normal pregnancy. Gianaroli et al biopsied day 3 blastomeres to assess chromosomes X, Y, 13, 18, and 21 using multicolored FISH (fluorescence in-situ hybridization) in a study population of 75 embryos. Of 61 embryos (81%) with normal morphology, 50 were successfully biopsied, and 24 were diagnosed as chromosomally abnormal.49 Although blastomere



biopsy is a potential tool for embryo selection, it also risks potential harm to the embryo and has been associated with misinformation due to the errors inherent in single-cell FISH. For review of the interaction between embryo morphology and chromosomal abnormalities see Munne et al50 and Chapter 18.

BLASTOCYST DEVELOPMENT

Improvements in culture media and the subsequent ability to culture embryos to the blastocyst stage has made blastocyst transfer possible, and this has been proposed as a means of selecting embryos with improved developmental competence and implantation potential. Blastocyst transfer also mirrors the natural timeline of development in vivo, potentially improving synchrony between uterine receptivity and embryonic development. Blastocyst development is influenced by patient factors, including sperm quality23,51 and maternal age52,53 as well as factors associated with the earlier stages of embryo development. The number of oocytes collected, number of oocytes inseminated, number of two pronucleate zygotes, and number of embryos cleaving to at least the 8-cell stage by day 3 in culture also influence blastocyst formation.23 BLASTOCYST MORPHOLOGY

Assessment of blastocyst morphology focuses on characteristic features: extent of cavitation, the number of cells in the trophectoderm and inner cell mass (ICM), and the shape of the ICM (see Figure 9.5). Please refer to Chapter 7 for a detailed discussion of blastocyst morphology. The blastocyst stage of development is usually reached by day 5, although some slow developing blastocysts do not reach this stage until day 6 or 7.54 Hardarson et al evaluated the differences in blastocyst morphology and chromosomal abnormalities in blastocysts derived from surplus embryos versus those derived from good quality embryos,

HPE_Chapter09.qxp

7/13/2007

4:49 PM

Page 109


A

B

C

Figure 9.5 Representative blastocyst images.These photographs depict the typical development of the blastocyst from early (A), expanding (B) to fully expanded (C).

using FISH to identify chromosomes 13, 16, 18, 21, 22, X, and Y.55 Grade A blastocysts were defined as expanded or expanding with a distinct trophectoderm and eccentrically located ICM. Grade B blastocysts were defined as those blastocysts that were poorly expanded and/or with less defined trophectoderm and ICM cells, but without signs of degenerative foci. Grade C blastocysts exhibited a number of degenerative foci in the ICM and trophectoderm and a poorly developed blastocyst cavity. Good morphology blastocysts had a significantly higher cell number and a significantly higher number of chromosomally normal cells, compared with poor quality blastocysts.55 As described in Chapter 7, Gardner used a similar scoring system, but evaluated the ICM and the trophectoderm independently. There is also a relationship between the size and shape of the ICM, as well as the arrangement of ICM cells and implantation potential. Richter et al demonstrated that blastocysts with an ICM size ⬎4500 ␮m2 had significantly higher implantation rates than those with a size ⱕ4500 ␮m2 (measured as length ⫻ width). These authors also identified an optimal shape of the ICM, using what they phrased as ‘the roundness index.’ This value was calculated by dividing length by the width of the ICM. The optimal shape (associated with the highest implantation rates) was a roundness index of 1.04–1.20.

When they combined these factors together, the implantation rates were highest for blastocysts that demonstrated both optimal size and shape.56

CUMULATIVE SCORING: MULTIDAY ASSESSMENTS

In an effort to improve embryo selection, several investigators have combined key morphological criteria for predicting the developmental competence of cleavage stage embryos. Given the multitude of scoring systems available, many of which have focused on individual time points in embryo development, several authors have turned to multi-day scoring as a means of enhancing embryo selection (see Tables 9.2 and 9.3 for review of multiday scoring systems). Early investigators of embryo scoring systems also foresaw the utility of combining aspects of both developmental rate and morphological assessment: Cummins, Puissant and Steer all included some assessment of developmental rate in their respective scoring systems.1,2,31 In 1998, Scott and Smith devised the corrected embryo score, which was based on a combination of pronuclear morphology and assessment of early cleavage. A retrospective analysis found that the corrected embryo score was associated with improved



 Corrected Embryo Score Retrospective analysis Two assessments: 1) Assessment of pronuclear morphology 2) Early cleavage

Alignment of pronuclei, nucleolar, and cytoplasmic features Pronuclear alignment: Score 5 Nucleolar scoring: Score 5: nucleoli aligned in a row at the PN junction. Score 4: nucleoli beginning to align. Score 3: nucleoli scattered Cytoplasmic scoring: Score 5: clear area around the pronuclei with a darkened ring or halo in the middle. Score 3: Embryos with a pitted or darkened cytoplasm. Max. 15 points.

If the embryo had progressed to nuclear membrane breakdown or had cleaved to the 2-cell stage, it received an additional 10 points

Title Type of study Timing of embryo assessment

Assessment of pronuclear (PN) morphology (16–18 hours)

Assessment of early cleavage (24–27 hours)

Additional points were awarded if cleavage was present. Symmetrical cleavage and ⬍20% fragmentation at the first cell division were also given additional weight

Three assessments: 1) Pronuclear morphology 2) Assessment of day 2 and day 3 embryo morphology 3) Growth rate 1) Position of pronuclei 2) Position and type of nucleoli 3) Cytoplasmic morphology Equally divided among three factors (rated 1–5, with 5 being the best score). Max. score 15


De Placido et al, 200258

Zygotes scoring according to Scott et al, 200061 Z-1: equal number and size of pronuclei, equal numbers of nucleoli aligned at pronuclear junction Z-2: equal number and size of nuceloli, but scattered Z-3: equal numbers of nucleoli with one pronucleus displaying alignment and the other with scattered nucleoli, OR unequal numbers or size of nucleoli Z-4 pronuclei not aligned or of grossly different size

Two assessments: 1) Combined Z-score 2) Day 3 morphology assessment


Lan et al, 200359

Embryos were inspected 26 hours after ICSI (all cycles ICSI in this study). 2-cell embryos were regarded as ‘early cleaved’

NA

Three assessments: 1) Early cleavage 2) Day 2 mononucleation 3) Day 3 features


Ciray et al, 200560

4:49 PM

Three assessments: 1) Pronuclear scoring 2) Early cleavage 3) Weighted assessment of day 3 morphology Alignment of pronuclei, nucleolar, and cytoplasmic features. Alignment of the nucleoli were given additional weight (See Table 9.4, definition of GES)

The Graduated Embryo Score Retrospective analysis

Fisch et al, 200157

7/13/2007

Scott and Smith, 199814

Table 9.2 Retrospective studies using multiday embryo scoring



NA

NA

25 per embryo. CS was calculated by adding the scores for each embryo and then dividing by no. embryos transferred

Day 1

Day 2 features (44–48 hours)


Maximum embryo score

Day of transfer

Day 3 and day 5

Day 2 or day 3. Embryos were selected for transfer based on 1) Number of blastomeres 2) Embryo morphologic score 3) Zygote score

Day 3, 4, or 5

Grade I: 8 cells, blastomeres of equal size and no cytoplasmic fragments Grade II: 8 cells, blastomeres of equal size and ⬍20% fragments Grade III: 8 cells with uneven blastomere sizes and no cytoplasmic fragments Grade IV: 4 or 8 cells with ⬎20% fragmentation Grade V: few blastomeres of any size or major fragmentation Combined

NA

(Continued)

Embryo score was calculated by multiplying blastomere number by the given point value. Two or three embryos with the highest scores were transferred. Embryos were transferred on the basis of day 3 scores Day 3

Grade 1: (4 points) ⫽ even and clear blastomeres with no fragments Grade 2: (3 points) uneven and irregular blastomeres, moderately clear cytoplasm and fragmentation ⬍20% Grade 3: (2 points) 20–50% fragmentation Grade 4: (1 point) fragmentation ⬎50% Embryos with the highest scores were transferred

A 4-cell embryo with a single nucleus in all cells was given the designation of ‘mononucleated’

4:49 PM

Weighted score ⫽ (embryo score ⫻ number of blastomeres) ⫻ zygote score Embryos were divided into three groups on the basis of score

Additional points awarded for advanced growth (optimal being ⬎4 blastomeres on day 2 and ⬎7 blastomeres on day 3) Embryos were scored on three parameters (rated 1–5, with 5 being the best) 1) Blastomere volume and synchrony of cleavage 2) Observation of a single nucleus within individual blastomeres 3) Extent of fragmentation Max. score of 15

7/13/2007

100

Cell number and grade were assessed. Additional weight was given to embryos containing 7–9 cells with Grade I morphology Grade I: symmetrical blastomeres and absent fragmentation Grade II: slightly uneven blastomeres and ⬍20% fragmentation Grade III: uneven blastomeres and ⬎20% fragmentation

NA





 GES 70–100 had an implantation rate of 39% compared with 24% of GES 0–65

The pregnancy rates were significantly correlated with GES. If the best embryo transferred had GES ⬎70, pregnancy rates were 59% compared with 34% if the best scoring embryo was ⬍65

When CS ⱖ15, the implantation rate was 28%, compared with 2% when the CS ⬍15

When CS ⬎15, the pregnancy rate was 71%, compared with 8% when the CS ⬍15. In addition, patients with CS ⬎15 had a multiple pregnancy rate of 29%

Outcome: implantation rates


The weighted day 3 score had the highest correlation with pregnancy rates. Using the weighted day 3 scores, the top group of embryos had a 71% pregnancy rate, compared with 45% pregnancy rate in group 2 and 12.5% pregnancy rate in group 3

NA

De Placido et al, 200258

The implantation rate was the highest in the group of embryos that had both early cleavage and mononucleation. (45% versus 15% compared with embryos without early cleavage or mononucleation) Pregnancy rates were highest in the cohort of embryos displaying at early cleavage and mononucleation (69% versus 33% compared with embryos without early cleavage or mononucleation)

NA

Ciray et al, 200560

4:49 PM

Pregnancy rates increased as the number of top quality embryos available for transfer increased (p ⫽ 0.008). The multiple pregnancy rate also increased significantly as well

Implantation rates increased as the number of top quality embryos for transfer increased (p ⬍0.001)

Embryos derived from Z-1 zygotes and grade I day 3 morphology were defined as ‘top quality’ These top quality embryos had a 92% day 5 ‘survival’ rate

Lan et al, 200359

7/13/2007

NA, not applicable; CS, cytoplasmic scoring; GES, graduated embryo score.

Blastocyst conversion was correlated with GES. Embryos with GES ⬎70 had 44% blastocyst conversion compared with 9% conversion in embryos scoring 0–65

NA

Fisch et al, 200157

Outcome: blastocyst conversion

Scott and Smith, 199814

Table 9.2 (Continued)



Zygote scoring: Z-1: equal number and size of pronuclei, equal numbers of nucleoli aligned at pronuclear junction Z-2: equal number and size of nuceloli, but scattered Z-3: equal numbers of nucleoli with one pronucleus displaying alignment and the other with scattered nucleoli OR unequal numbers or size of nucleoli Z-4 pronuclei not aligned or of grossly different size NA

Assessment of pronuclear (PN) morphology (16–18 hours)


Cell number and grade were assessed. Additional weight was given to embryos containing 7–9 cells with ⬍20% fragmentation. (See Table 9.4)

30 points were awarded if early cleavage was present. Cells with ⬍20% fragmentation at the first cell division were also given additional points

Three assessments: 1) Pronuclear scoring and 2) Early cleavage and 3) Weighted assessment of day 3 morphology Assessment of nucleolar alignment. If nucleoli were aligned along the pronuclear axis, 20 points were awarded

Cell number and size of blastomeres, presence of fragments were assessed. Grade A: ⬎4 blastomeres on day 2, ⬎6 on day 3 with ⬍10% fragmentation

NA

Pronuclear assessment occurred 20 hours following ICSI. 1) Position and size of pronuclei 2) Distribution of nucleoli (polarized or non-polarized)

Prospective randomized trial Group 1: selection (for transfer) based only on PN morphology Group 2: selection based only on day 3 morphology Group 3: selection based on PN ⫹ day 3 morphology Two assessments: 1) Pronuclear morphology 2) Day 3 morphology

Nagy et al, 200463

(Continued)

The exact criteria for ideal day 3 embryo morphology is not discussed.

Group 2: Three subgroups Subgroup A: zygotes that cleaved into 2 cells at 26 h Subgroup B: zygotes where PN breakdown had occurred but cleavage had not occurred Subgroup C: PN still intact

Group 1: Three subgroups: Subgroup A: nucleoli large or medium sized and nucleolar alignment seen Subgroup B: Nucleoli large or medium without alignment Subgroup C: Nucleoli small or pinpoint without any nucleolar alignment

Prospective randomized trial Group 1: evaluation of pronuclear morphology combined with day 3 morphology and progression Group 2: evaluation of early cleavage combined with day 3 morphology and progression Two assessments: 1) Pronuclear morphology or 2) Early cleavage combined with 3) Day 3 morphology

Chen et al, 200664

4:49 PM

Grade 1: 8 cells, ⬍10% fragmentation, no multinucleation. Grade 2: 8-cell, 10–20% fragmentation or lacking cell–cell contact, no

Two assessments: 1) Assessment of pronuclear morphology and 2) Day 3 morphology

Timing of embryo assessment

The Graduated Embryo Score Prospective cohort: Comparison of GES versus traditional day 3 morphology

Fisch et al, 200362

7/13/2007

Assessment of early cleavage (24–27 hours)

Z-Score Prospective cohort: Comparison of selection with Z-score ⫹ day 3 morphology versus selection based on day 3 morphology alone

Title Type of study

Scott et al, 200061

Table 9.3 Prospective studies using multiday embryo scoring







NA, not applicable; GES, graduated embryo score.


Implantation rates were higher for patients randomized to have combined pronuclear and day 3 scoring, compared with either parameter alone (21%, compared with 12% and 15% respectively)

Pregnancy rates were higher for patients randomized to have combined pronuclear and day 3 scoring, compared with either parameter alone (39%, compared with 23% and 26% respectively)

Using GES to select embryos for transfer was associated with a higher implantation rate than traditional morphology alone (36% implantation versus 27%)

With ⱖ1 embryo with GES ⱖ70, pregnancy rates were 62%. With ⱖ1 grade A embryo, pregnancy rate was 50%. Transferring ⱖ1 embryo with GES ⱖ70 did not improve pregnancy rates, but increased rates of multiples

Day 3 NA

Grade B: 10–30% fragmentation Grade C: 30–50% fragmentation

Nagy et al, 200463

No difference in implantation seen between group 1 and group 2 or between corresponding subgroups. Among both group 1 and group 2 patients, embryos in subgroup A showed significantly higher pregnancy rates

No difference in implantation seen between group 1 and group 2 or between corresponding subgroups. Among both group 1 and group 2 patients, embryos in subgroups A showed significantly higher implantation rates

Day 3 NA

Chen et al, 200664

4:49 PM

Outcome: implantation rates

Day 3 and day 5

Fisch et al, 200362

7/13/2007

Day of transfer Outcome: blastocyst conversion

multinucleation Grade 3: 6–7 cells or 8 cells with 20% fragmentation or uneven blastomeres, no multinucleation Grade 4: ⬎8 cells or 4–6 cells or 8-cells with ⬎20% fragmentation or uneven blastomeres or multinucleation Grade 5: ⬍4 cells or grossly fragmented or with ⬎50% multinucleated blastomeres Day 3 or day 5 Zygotes displaying equality between the nuclei (Z1 or Z2) had a significantly higher blastocyst conversion rate than those zygotes without nuclear equality (49.5% versus 28%) Using Z-score ⫹ day 3 morphology resulted in significantly higher implantation rates than day 3 morphology alone. (Day 3 transfers 31% versus 19%, day 5 transfers 52% versus 39%) Using Z-score ⫹ day 3 morphology resulted in significantly higher pregnancy rates than day 3 morphology alone. (Day 3 transfers 57% versus 33%, day 5 transfers 73% versus 58%)

Scott et al, 200061

Table 9.3 (Continued)



HPE_Chapter09.qxp

7/13/2007

4:49 PM

Page 115


implantation and pregnancy rates.14 In 2000, the same group designed a prospective cohort study in which they compared the utility of selecting embryos based on combining Z-score (modified from their original work) with day 3 embryo morphology, versus day 3 morphology alone. These results indicated that the combination of pronuclear and day 3 assessment resulted in significantly higher implantation and pregnancy rates.61 Using similar principles of multiday assessment, Fisch et al devised the graduated embryo score (GES), see Table 9.4.62 The GES is comprised of three total embryo assessments (two interval evaluations of early developmental milestones, combined with a weighted assessment of conventional morphological characteristics on day 3). The first assessment was carried out at 16–18 hours and evaluated cytoplasmic halo and vacuoles, pronuclear size and juxtaposition, nucleolar alignment and polar body apposition and fragmentation. The embryos were then evaluated at 25–27 hours postinsemination for dissolution of the pronuclear membrane, blastomere cleavage, symmetry, and degree of fragmentation. A final evaluation was done at 64–67 hours postinsemination for blastomere morphology and number,

and then a total GES was calculated and correlated with blastocyst development and implantation rate.57 In their initial retrospective analysis, a GES ⬎70 was significantly associated with higher pregnancy rates when compared to scores 0–65 (59% versus 34%, p ⬍0.0015) and this also helped to predict blastocyst conversion rate.57 A follow-up study compared the outcomes of embryos selected using the GES versus those chosen for transfer based on morphological features alone. IVF outcomes using the GES to select embryos for transfer resulted in a statistically significant increase in ongoing pregnancy rate at 12 weeks compared with embryos graded by day 3 morphological assessment alone (62% vs 50%).62 Several additional studies using combined assessment of pronuclear morphology and day 3 embryo assessment have followed the publication of the combined Z-score assessment and the graduated embryo score. Nagy et al. randomized patients to one of three groups for embryo selection (selection based on PN morphology alone, day 3 morphology alone, or combined assessment). In this randomized study, those patients assigned for combined assessment had significantly higher implantation and pregnancy rates.63 Lan et al., also used the approach of combined

Table 9.4 Graduated embryo score. Reprinted from Fisch et al. The graduated embryo score predicts the outcome of assisted reproductive technologies better than a single day 3 evaluation and achieves results associated with blastocyst transfer from day three embryo transfer. Fertil Steril 2003; 80: 1352–8, ©2003 with permission from The American Society for Reproductive Medicine.62 Evaluation 1 2

3 Total score

Hours after insemination 16–18 25–27

64–67

Developmental milestone Nucleoli aligned along pronuclear axis Cleavage regular and symmetrical Fragmentationa Absent ⬍20% ⬎20% Cell number and gradeb 7CI, 8CI, 8CII, 9CI, 7CII, 9CII, 10CI, 11CI, compacting I

Score 20 30 30 25 0 20 10 100

aIf

the embryo was not cleaved at 25–27 hours, grading of fragmentation should occur at the 64–67 hour evaluation, if the embryo reached the 7-cell stage and had ⬍20% fragmentation. bGrade I ⫽ symmetrical blastomeres and absent fragmentation. Grade II ⫽ slightly uneven blastomeres and ⬍20% fragmentation. Grade III ⫽ uneven blastomeres and ⬎20% fragmentation. Grade A embryos are 7 or more cells with ⬍20% fragmentation.



HPE_Chapter09.qxp

7/13/2007

4:49 PM

Page 116


pronuclear and day 3 features in their retrospective study, which defined embryos derived from Z-1 zygotes (based on Scott et al61) that became grade 1 embryos as ‘top quality.’ These ‘top quality’ embryos had a 92% day 5 survival rate and were associated with significantly greater implantation and pregnancy rates.59 De Placido et al also used a combined pronuclear and day 3 assessment, but included a weighting based on embryonic growth rates/cell number. The weighted day 3 embryo score had the highest correlation with pregnancy rates, and top scoring embryos achieved a pregnancy rate of 71%.58 Several multiday assessments have used the combination of pronuclear morphology and day 3 morphology, and some have included early cleavage as part of the weighted score. A recent prospective trial randomized embryos to two groups according to the manner in which they were scored.64 Group 1 embryos were given a score based on a combination of pronuclear morphology and day 3 morphology; pronuclear morphology was scored as follows: subgroup A, nucleoli large or medium sized and nucleolar alignment seen; subgroup B, nucleoli large or medium with no alignment; and subgroup C, nucleoli small or pinpoint with no nucleolar alignment. Group 2 embryos were assessed at 26 h for early cleavage combined with day 3 morphology.64 Embryos at early cleavage were segregated into those that had cleaved to 2 cells (subgroup A), those in which pronuclear breakdown had occurred but cleavage had not occurred (subgroup B), and those still exhibiting two distinct PNs (subgroup C). These authors found no difference in implantation rates between group 1 and group 2, although they did find that subgroup A (in both groups) had a higher implantation and pregnancy rate than subgroup C. However, when the corresponding subgroups were compared, there was no difference in pregnancy rates.64 This study raises questions as to whether or not it is necessary (or even beneficial) to evaluate embryos for both pronuclear morphology and for early cleavage. In addition, these authors found that pronuclear morphology and early cleavage were interrelated. For example, those embryos with subgroup A pronuclear morphology were more likely to undergo early cleavage, and likewise the embryos that underwent 

early cleavage were more likely to arise from subgroup A. Therefore, evaluation of only one of these markers of developmental competence may achieve results similar to those following evaluation of both markers. Along these lines, another combined scoring system has been recently reported, which used the relatively novel combination of early cleavage, day 2 mononucleation and day 3 morphology. In this paper, the implantation and pregnancy rates were highest in the group displaying both early cleavage and mononucleation.60 Available evidence suggests that multiday assessment provides a more accurate picture of developmental progression as opposed to a single static observation. However, the ultimate combination of morphological features required for optimum evaluation of developmental competence has yet to be resolved. When the critical time points have been identified, we can then determine the ideal timing for embryo transfer.

TIMING OF EMBRYO TRANSFER: OPTIMIZING PREGNANCY RATES

There has been much debate surrounding the ideal timing of embryo transfer in human IVF. Compared with animal models, the human uterus is far more forgiving in terms of receptivity to embryo transfer, and this has allowed pregnancies to be successfully established not only with transfers from day 1 to day 6, but also, remarkably on day 0 with eggs and sperm being introduced into the uterus.65 However, as implantation rates and pregnancy rates continue to improve, the optimal embryo transfer window for each patient needs to be determined. Until recently, the vast majority of embryo transfers were carried out on day 2; since culture conditions did not permit extended culture, it was felt that the embryos should be transferred to the maternal environment as quickly as possible. In the late 1990s, when improved culture media became available, several groups began to question this timing, particularly since the embryo typically enters the uterus late on day 4 in vivo. Extending embryo culture may

HPE_Chapter09.qxp

7/13/2007

4:49 PM

Page 117


also serve to improve selection of embryos with greater degrees of developmental competence. In 1995, Dawson et al retrospectively analyzed day 2 versus day 3 transfers, and although there was no difference in implantation or pregnancy rates, they found that day 3 transfers were associated with a significant difference in development to the fetal heart stage, with fewer miscarriages.66 These authors also concluded that the additional day in culture allowed them to select embryos that had passed an additional test of development, in that they had survived the stage of embryonic gene activation without arresting at the 4- to 8-cell stage. Carillo et al also compared day 2 and day 3 transfers retrospectively, and found that delaying the transfer by 24 hours did result in an improved implantation rate and ongoing pregnancy rate.37 However, prospective studies have not substantiated these findings. Laverge et al designed a prospective randomized study to compare day 2 and day 3 transfers in terms of implantation and pregnancy rates. They found no differences in implantation or pregnancy rates, but did observe a significant decrease in embryo quality when embryos were kept in culture until day 3.67 These results were consistent with two recent additional prospective studies that also found no difference in pregnancy rates between day 2 and day 3 transfers.68,69 Finally, a meta-analysis carried out in 2004 that reviewed the timing of embryo transfer found a slightly improved clinical pregnancy rate for day 2 versus day 3 transfer, but no difference in live birth rate.70 Based on the principle that additional observation in culture would allow embryos that did not arrest to be identified, several authors have suggested that extended culture to the blastocyst stage may further improve implantation rates, thus allowing transfer of fewer embryos and decreasing multiple gestations. Indeed, blastocyst transfer became a viable option following development of sequential media systems,71 although there is currently debate regarding the need to use a sequential two-step system, rather than a single one-step system.72,73 A retrospective pilot study by Gardner et al showed that blastocyst transfer resulted in a significantly higher implantation rate, when compared with

day 3 transfer.74 A follow-up prospective, randomized study confirmed increased implantation rates from day 5 transfers, but showed equivalent clinical pregnancy rates with day 3 transfer.71 These observations have led some authors to postulate that blastocyst transfer may provide a strategy to optimize pregnancy rates while decreasing multiple gestations. However, blastocyst transfer may not be ideal in all cases, and may compromise a successful outcome that might otherwise have been achieved following a day 3 transfer.8 This suggests that the uterus may provide a superior environment compared with that of the in vitro system, and that day 3 embryo transfer may have ‘rescued’ embryos that otherwise may not have developed to blastocysts in culture.42 A Cochrane review of blastocyst transfer versus day 3 embryo transfer identified 16 trials for metaanalysis and concluded that there was no significant difference in live birth rate, pregnancy rate, incidence of multiple gestations, or miscarriage rates. In addition, they found that those patients who underwent a day 3 transfer were more likely to have additional embryos available for cryopreservation, and patients undergoing a day 5 transfer were more likely to have no embryos available for transfer.75 In contrast, a recent prospective randomized trial comparing single blastocyst transfer versus single cleavage stage transfer concluded that the delivery rate from fresh transfers was significantly higher after single blastocyst transfer.76 However, fewer embryos were frozen on day 5 (2.2 vs 4.2, p ⫽ 0.001) and the livebirth rate was higher following transfer of thawed day 3 embryos (15% vs 11%). Given these findings, and depending on whether pregnancy rates are calculated from solely fresh transfers versus fresh and frozen transfers (i.e. the cumulative pregnancy rate), debate remains as to whether single transfer at the blastocyst or the cleavage stage is the optimal procedure. Furthermore, many programs have difficulty with successful blastocyst freezing and therefore have reverted to performing day 3 transfer. The optimum day for embryo transfer has yet to fully be elucidated, and is also influenced by the success of cryopreservation on specific days, within an individual laboratory. It is clear that a variety of patient and laboratory factors play a role in determining the 

HPE_Chapter09.qxp

7/13/2007

4:49 PM

Page 118


ideal timing for each individual patient. Moreover, while the majority of studies use pregnancy rates after fresh transfer as the index of success, cumulative pregnancy rates derived from both fresh and frozen transfers for each stimulated cycle provide a more accurate reflection of cycle efficacy and success. A recent study proposing single blastocyst transfer reported high pregnancy rates of cryopreserved blastocysts in either a subsequent natural cycle or hormonally prepared endometrium, suggesting that perhaps the transfer of only one blastocyst at first attempt may lead to a higher cumulative pregnancy rate, by allowing transfer of cryopreserved blastocysts to a more favorable uterine environment.77

NUMBER OF EMBRYOS TO TRANSFER

Single embryo transfer provides the potential opportunity to change the landscape of IVF. Currently, the incidence of multiple gestations remains high. In 2003, the percentage of live births that had multiple infants resulting from ART in the United States ranged from 17.4% to 38.4%, depending on maternal age.78 Although there have been efforts to decrease high-order multiple gestations (triplets and higherorder multiples), until recently, both patients and providers have accepted the consequence of twins, in return for an acceptable pregnancy rate. This implied agreement between patients and doctors has recently come under scrutiny, as twin pregnancies are associated with increased maternal and fetal complications compared with singleton gestations. In the early days of IVF, a high number of embryos was typically transferred so as to achieve a reasonable pregnancy rate.1 Indeed, in early studies, the total number of embryos transferred was highly correlated with pregnancy.2–4,79 In the late 1990s, investigators began questioning the practice of transferring three or more embryos in routine IVF.5,7 The logical next step in the attempt to reduce multiple gestations is to provide appropriate patients with the option of elective single embryo transfer. However, selecting the appropriate patient population and the 

best embryo from any given cohort remains the clinical challenge. Van Royen et al defined a ‘top quality embryo’ by identifying the embryonic features associated with ongoing twin pregnancies following double embryo transfer. They found that the presence of top quality embryos was highly associated with both ongoing pregnancies and multiple gestations, and that twin pregnancies after double embryo transfer only occurred in patients ⬍38 years of age.46 Hellberg et al used similar methods (analysis of twin pregnancies after a double transfer) to identify the following characteristics: maternal age ⱕ38 years, infertility cause other than male factor (male factor was associated with a decreased risk of twinning), two high quality embryos, ability to freeze at least one embryo, and ⱖ5 fertilized oocytes, and concluded that all of the above were associated with increased twinning.80 Building on the definition of a top quality embryo, De Neubourg et al81 used Van Royen’s definition of a top quality embryo (see above) to select embryos for single embryo transfer on day 3. Using these selection criteria, they were able to demonstrate a 50% pregnancy rate with transfers to women ⬍38 years of age. Thurin et al82 randomized women ⬍36 years of age with at least two good quality embryos to undergo single or double embryo transfer and then analyzed cycles with 0% or 100% implantation to identify variables associated with success. Multivariate analysis revealed that first IVF cycle, IVF as method of fertilization (as opposed to ICSI), 4-cell embryos (day 2), and number of IU of FSH per oocyte retrieval were predictive of implantation.83 Recently, randomized trials have shown that single embryo transfer shows increasing promise. A randomized multicenter trial of single embryo transfer (with subsequent transfer of frozen embryos in the event of no live birth) versus double embryo transfer (in women ⬍36 years of age with at least two good quality embryos) demonstrated roughly equivalent cumulative pregnancy rates (39% versus 43%) with a substantial reduction in multiple pregnancy rates (0.8% compared with 33%).82 Single blastocyst transfer followed by cryostored blastocyst transfer has also showed equivalent pregnancy rates with

HPE_Chapter09.qxp

7/13/2007

4:49 PM

Page 119


decreased rates of multiple gestations. In addition, there were five perinatal deaths in the double embryo transfer group, all involving twin pregnancies.84

CONCLUSION

Although IVF has made remarkable strides in the optimization of culture media, stimulation regimens, and, ultimately, pregnancy rates, the challenge to reduce the number of twin pregnancies remains. Clearly, single embryo transfer is the answer. However, in order for this practice to become acceptable to both doctors and patients, both the patient and embryo must be appropriately selected. To accomplish optimal embryo selection, investigators have evaluated several features and time points for embryo morphologic assessment including pronuclear morphology, early cleavage, cell number, the extent of fragmentation, the degree of asymmetry, the presence of mononucleation on day 2, and standard morphological features on day 3, including early compaction. Compared with exclusive day 3 morphological assessment, evidence is accumulating that multiday scoring systems provide improved predictability as to which embryos are likely to implant.61–63 Nevertheless, it remains to be determined whether evaluation must be performed on each day of culture, to what extent repeated observations may compromise embryo development, and which combination of features provides greatest selection sensitivity. In addition, the incremental improvements in implantation potential must be weighed against the time and labor of the embryology staff; indeed, multiday assessments may not be possible in every IVF program. As a randomized trial recently demonstrated, it may not be necessary to perform both pronuclear and early cleavage scoring.64 Going forward, we must demand more rigorous studies, comprised of prospective, randomized trials, to evaluate not only the scoring systems themselves, but also the differential weighting of all the different parameters and time points. The ultimate result of such trials will hopefully improve standardization of embryo scoring systems, and provide improved efficacy and safety for our patients.

ACKNOWLEDGMENTS

We thank the entire embryology team at Brigham and Women’s Hospital for their expertise in embryo grading. We would particularly like to thank Kerry Kelleher for photographing the embryo images in this chapter and Paulette Nippet for her assistance in acquiring copyright permissions. REFERENCES 1. Puissant F, Van Rysselberge M, Barlow P et al. Embryo scoring as a prognostic tool in IVF treatment. Hum Reprod 1987; 2: 705–8. 2. Cummins J, Breen T, Harrison K. A formula for scoring human embryo growth rates in in vitro fertilization: its value in predicting pregnancy and in comparison with visual estimates of embryo quality. J In Vitro Fert Embryo Transf 1986; 3: 284–95. 3. Tsai Y, Chung M, Sung Y et al. Clinical value of early cleavage embryo. Int J Gynaecol Obstet 2002; 76: 293–7. 4. Ziebe S, Petersen K, Lindenberg S et al. Embryo morphology of cleavage stage: how to select the best embryos for transfer after in-vitro fertilization. Hum Reprod 1997; 12: 1545–9. 5. Templeton A, Morris J. Reducing the risk of multiple births by transfer of two embryos after in vitro fertilization. N Engl J Med 1998; 339: 573–7. 6. Gardner D, Lane M, Stevens J et al. Blastocyst score affects implantation and pregnancy outcome: towards a single blastocyst transfer. Fertil Steril 2000; 73: 1155–8. 7. Milki A, Fisch J, Behr B. Two-blastocyst transfer has similar pregnancy rates and a decreased multiple gestation rate compared with threeblastocyst transfer. Fertil Steril 1999; 72: 225–8. 8. Racowsky C, Jackson K, Cekleniak N et al. The number of eight-cell embryos is a key determinant for selecting day 3 or day 5 transfer. Fertil Steril 2000; 73: 558–64. 9. Kolibianakis E, Zikopoulos K, Verpoest W et al. Should we advise patients undergoing IVF to start a cycle leading to a day 3 or a day 5 transfer. Hum Reprod 2004; 19: 2550–4. 10. Papanikolaou E, D’haeseleer E, Verheyen G et al. Live birth rate is significantly higher after blastocyst transfer than after cleavage-stage embryo transfer when at least four embryos are available on day 3 of embryo culture. A randomized prospective study. Hum Reprod 2005; 20: 3198–203. 11. Moore KL, Persaud TVN. The Developing Human: Clinically oriented embrydogy, 6th edn. Philadelphia: WB Saunders, 1998: 44. 12. Ludwig M, Schopper B, Al-Hasani S et al. Clinical use of a pronuclear stage score following intracytoplasmic sperm injection: impact of pregnancy rates under the conditions of the German embryo protection law. Hum Reprod 2000; 15: 325–9. 13. Zollner U, Zollner K-P, Hartl G et al. The use of a detailed zygote score after IVF/ICSI to obtain good quality blastocysts: the German experience. Hum Reprod 2002; 17: 1327–33. 14. Scott L, Smith S. The successful use of pronuclear embryo transfers the day following oocyte retrieval. Hum Reprod 1998; 13: 1003–13. 15. Tesarik J, Greco E. The probability of abnormal preimplantation development can be predicted by a single static observation on pronuclear stage morphology. Hum Reprod 1999; 14: 1318–23. 16. Wright G, Wiker S, Elsner C et al. Observations on the morphology of pronuclei and nucleoli in human zygotes and implications for cryopreservation. Hum Reprod 1990; 5: 109–15.



HPE_Chapter09.qxp

7/13/2007

4:49 PM

Page 120


17. Payne D, Flaherty S, Barry M et al. Preliminary observations on polar body extrusion and pronuclear formation in human oocytes using time-lapse video cinematography. Hum Reprod 1997; 12: 532–41. 18. Balakier H, MacLusky N, Casper R. Characterization of the first cell cycle in human zygotes: implications for cryopreservation. Fertil Steril 1993; 59: 359–65. 19. Capmany G, Taylor A, Braude P et al. The timing of pronuclear formation, DNA synthesis and cleavage in the human 1-cell embryo. Mol Hum Reprod 1996; 2: 299–306. 20. Trounson A, Mohr L, Wood C et al. Effect of delayed insemination on in-vitro fertilization, culture and transfer of human embryos. J Reprod Fertil 1982; 64: 285–94. 21. Braude P, Bolton V, Moore S. Human gene expression first occurs between the four and eight cell stages of preimplantation development. Nature 1988; 332: 459–61. 22. Nikas G, Ao A, Winston R et al. Compaction and surface polarity in the human embryo in-vitro. Biol Reprod 1996; 55: 32–7. 23. Jones G, Trounson A, Lolatgis N et al. Factors affecting the success of human blastocyst development and pregnancy following in vitro fertilization and embryo transfer. Fertil Steril 1998; 70: 1022–9. 24. Payne J, Raburn D, Couchman G et al. Relationship between preembryo pronuclear morpholoy (zygote score) and standard day 2 or 3 embryo morphology with regard to assisted reproductive technique outcomes. Fertil Steril 2005; 84: 900–9. 25. Edwards R, Fishel S, Cohen J. Factors influencing the success of in vitro fertilization for alleviating human infertility. J In Vitro Fert Embryo Transf 1984; 1: 3–23. 26. Shoukir Y, Campana A, Farley T et al. Early cleavage of in-vitro fertilized embryos to the 2-cell stage: a novel indicator of embryo quality and viability. Hum Reprod 1997; 12: 1531–6. 27. Sakkas D, Shoukir Y, Chardonnens D et al. Early cleavage of human embryos to the two-cell stage after intracytoplasmic sperm injection as an indicator of embryo viability. Hum Reprod 1998; 13: 182–7. 28. Sakkas D, Percival G, D’Arcy Y et al. Assessment of early cleaving in vitro fertilized human embryos at the 2-cell stage before transfer improves embryo selection. Fertil Steril 2001; 76: 1150–6. 29. Ciray H, Karagenc L, Ulug U et al. Early cleavage morphology affects the quality and implantation potential of day 3 embryos. Fertil Steril 2006; 85: 358–65. 30. Ebner T, Yaman C, Moser M et al. Embryo fragmentation in vitro and its impact on treatment and pregnancy outcome. Fertil Steril 2001; 76: 281–5. 31. Steer C, Mills C, Tan S et al. The cumulative embryo score: a predictive embryo scoring technique to select the optimal number of embryos to transfer in an in-vitro fertilization and embryo transfer program. Hum Reprod 1992; 7: 117–9. 32. Roseboom T, Vermeiden J, Schoute E et al. The probability of pregnancy after embryo transfer is affected by the age of the patient, cause of infertility, number of embryos transferred and the average morphology score, as revealed by multiple logistic regression analysis. Hum Reprod 1995; 10: 3035–41. 33. Giorgetti C, Terriou P, Auquier P et al. Embryo score to predict implantation after in-vitro fertilization: based on 957 single embryo transfers. Hum Reprod 1995; 10: 2427–31. 34. Kligman I, Benadiva C, Alikani M et al. The presence of multinucleated blastomeres in human embryos is correlated with chromosomal abnormalities. Hum Reprod 1996; 11: 1492–8. 35. Jackson K, Ginsburg E, Hornstein M et al. Multinucleation in normally fertilized embryos is associated with an accelerated ovulation induction response and lower implantation and pregnancy rates in in vitro fertilization-transfer cycles. Fertil Steril 1998; 70: 60–6.



36. Saldeen P, Sundstrom P. Nuclear status of four-cell preembryos predicts implantation potential in in vitro fertilization treatment cycles. Fertil Steril 2005; 84: 584–9. 37. Carillo A, Lane B, Pridham D et al. Improved clinical outcomes for in vitro fertilization with delay of embryo transfer from 48 to 72 hours after oocyte retrieval: use of glucose- and phosphate-free media. Fertil Steril 1998; 69: 329–34. 38. Volpes A, Sammartano F, Coffaro F et al. Number of good quality embryos on day 3 is predictive for both pregnancy and implantation rates in in vitro fertilization/intracytoplasmic sperm injection cycles. Fertil Steril 2004; 82: 1330–6. 39. Racowsky C, Combelles C, Nureddin A et al. Day 3 and day 5 morphological predictors of embryo viability. Reprod Biomed Online 2003; 6: 323–31. 40. Desai N, Goldstein J, Rowland D et al. Morphological evaluation of human embryos and derivation of an embryo quality scoring system specific for day 3 embryos: a preliminary study. Hum Reprod 2000; 15: 2190–6. 41. Skiadas C, Jackson K, Racowsky C. Early compaction on day 3 may be associated with increased implantation potential. Fertil Steril 2006; 86: 1386–91. 42. Alikani M, Calderon G, Tomkin G et al. Cleavage anomalies in early human embryos and survival after prolonged culture in-vitro. Hum Reprod 2000; 15: 2634–43. 43. Hardarson T, Hanson C, Sjogren A et al. Human embryos with unevenly sized blastomeres have lower pregnancy and implantation rates: indications for aneuploidy and multinucleation. Hum Reprod 2001; 16: 313–18. 44. Rienzi L, Ubaldi F, Iacobelli M et al. Significance of morphological attributes of the early embryo. Reprod Biomed Online 2005; 10: 669–81. 45. Tao J, Tamis R, Fink K et al. The neglected morula/compact stage embryo transfer. Hum Reprod 2002; 17: 1513–18. 46. Van Royen E, Mangelschots K, De Neubourg D et al. Characterization of a top quality embryo, a step towards single embryo transfer. Hum Reprod 1999; 14: 2345–9. 47. Graham J, Han T, Porter R et al. Day 3 morphology is a poor predictor of blastocyst quality in extended culture. Fertil Steril 2000; 74: 495–7. 48. Milki A, Hinckley M, Gebhardt J et al. Accuracy of day 3 criteria for selecting best embryos. Fertil Steril 2002; 77: 1191–5. 49. Gianaroli L, Magli M, Ferraretti A et al. Preimplantation genetic diagnosis increases the implantation rate in human in vitro fertilization by avoiding the transfer of chromosomally abnormal embryos. Fertil Steril 1997; 68: 1128–31. 50. Munne S. Chromosome abnormalities and their relationship to morphology and development of human embryos. Reprod Biomed Online 2006; 12: 234–53. 51. Janny L, Menezo Y. Evidence for a strong paternal effect on human preimplantation embryo development and blastocyst formation. Mol Reprod Dev 1994; 38: 36–42. 52. Schoolcraft W, Gardner D, Lane M et al. Blastocyst culture and transfer: analysis of results and parameters affecting outcome in two invitro fertilization programs. Fertil Steril 1999; 72: 604–9. 53. Langley M, Marek D, Gardner D et al. Extended embryo culture in human assisted reproduction treatments. Hum Reprod 2001; 16: 902–8. 54. Dokras A, Sargent I, Barlow D. Human blastocyst grading: an indicator of developmental potential? Hum Reprod 1993; 8: 2119–27. 55. Hardarson T, Caisander G, Sjogren A et al. A morphological and chromosomal study of blastocysts developing from morphologically suboptimal human pre-embryos compared with control blastocysts. Hum Reprod 2003; 18: 399–407.

HPE_Chapter09.qxp

7/13/2007

4:49 PM

Page 121


56. Richter K, Harris D, Daneshmand S et al. Quantitative grading of a human blastocyst: optimal inner cell mass size and shape. Fertil Steril 2001; 76: 1157–67. 57. Fisch J, Rodriguez H, Ross R et al. The graduated embryo score (GES) predicts blastocyst formation and pregnancy rate from cleavage-stage embryos. Hum Reprod 2001; 16: 1970–5. 58. De Placido G, Wilding M, Strina I et al. High outcome predictability after IVF using a combined score for zygote and embryo morphology and growth rate. Hum Reprod 2002; 17: 2402–9. 59. Lan K, Huang F-J, Lin Y-C et al. The predictive value of using a combined Z-score and day 3 embryo morphology score in the assessment of embryo survival on day 5. Hum Reprod 2003; 18: 1299–306. 60. Ciray H, Karagenc L, Ulug U et al. Use of both early cleavage and day 2 mononucleation to predict embryos with high implantation potential in intracytoplasmic sperm injection cycles. Fertil Steril 2005; 84: 1411–16. 61. Scott L, Alvero R, Leondires M et al. The morphology of human pronuclear embryos is positively related to blastocyst development and preimplantation. Hum Reprod 2000; 15: 2394–403. 62. Fisch J, Sher G, Adamowicz M et al. The graduated embryo score predicts the outcome of assisted reproductive technologies better than a single day 3 evaluation and achieves results associated with blastocyst transfer from day 3 embryo transfer. Fertil Steril 2003; 80: 1352–8. 63. Nagy Z, Dozortsev D, Diamond M et al. Pronuclear morphology evaluation with subsequent evaluation of embryo morphology significantly increases implantation rates. Fertil Steril 2004; 80: 67–74. 64. Chen C, Kattera S. Comparison of pronuclear zygote morphology and early cleavage status of zygotes as additional criteria in the selection of day 3 embryos: a randomized study. Fertil Steril 2006; 85: 347–52. 65. Craft I, McLeod F, Green S et al. Human pregnancy following oocyte and sperm transfer to the uterus. Lancet 1982; 1: 1031–3. 66. Dawson K, Conaghan J, Ostera G et al. Delaying transfer to the third day post-insemination, to select non-arrested embryos, increases development to the fetal heart stage. Hum Reprod 1995; 10: 177–82. 67. Laverge H, De Sutter P, Van de Elst J et al. A prospective, randomized study comparing day 2 and day 3 embryo transfer in human IVF. Hum Reprod 2001; 16: 476–80. 68. de los Santos M, Mercader A, Galan A et al. Implantation rates after two, three or five days of embryo culture. Placenta 2003; 24: S13–S19. 69. Pantos K, Makrakis E, Stavrou D et al. Comparison of embryo transfer on day 2, day 3, and day 6: a prospective randomized study. Fertil Steril 2004; 81: 454. 70. Oatway C, Gunby J, Daya S. Day three versus day two embryo transfer following in vitro fertilization or intracytoplasmic sperm injection. The Cochrane Database of Systematic Reviews 2004: Art. No. CD004378.

71. Gardner D, Schoolcraft W, Wagley L et al. A prospective randomized trial of blastocyst culture and transfer in in vitro fertilization. Hum Reprod 1998; 13: 3434–40. 72. Biggers J, Racowsky C. The development of fertilized human ova to the blastocyst stage in KSOMAA medium: is a two-step protocol necessary? Reprod Biomed Online 2002; 5: 133–40. 73. Biggers J, McGinnis L, Lawitts J. One-step versus two-step culture of mouse pre-implantation embryos: is there a difference? Hum Reprod 2005; 20: 3376–84. 74. Gardner D, Vella P, Lane M et al. Culture and transfer of human blastocyst increases implantation rates and reduces the need for multiple embryo transfers. Fertil Steril 1998; 69: 84–8. 75. Blake D, Proctor M, Johnson N et al. Cleavage stage versus blastocyst stage embryo transfer in assisted conception. Cochrane Database Syst Rev 2005; (2): CD002118. 76. Papanikolaou E, Camus M, Kolibianakis E et al. In vitro fertilization with single blastocyst-stage versus single cleavage-stage embryos. N Engl J Med 2006; 354: 1139–46. 77. Criniti A, Thyer A, Chow G et al. Elective single blastocyst transfer reduces twin rates without compromising pregnancy rates. Fertil Steril 2005; 84: 1613–19. 78. Centers for Disease Control and Prevention. 2003 Assisted reproductive technology success rates. National summary and fertility clinic reports: US Department of Health and Human Services. Centers for Disease Control and Prevention, 2005. http://www.cdc.gov/ART/ Art2003/index.htm. 79. Hill G, Freeman M, Bastias M. The influence of oocyte maturity and embryo quality on pregnancy rate in a program for in vitro fertilization-embryo transfer. Fertil Steril 1989; 52: 801–6. 80. Hellberg D, Blennborn M, Nilsson S. Defining women who are prone to have twins in in vitro fertilization – a necessary step towards single embryo transfer. J Assist Reprod Genet 2005; 22: 199–206. 81. De Neubourg D, Gerris J, Mangelschots K et al. Single top quality embryo tansfer as a model for prediction of early pregnancy outcome. Hum Reprod 2004; 19: 1476–9. 82. Thurin A, Hausken J, Hillensjo T et al. Elective single-embryo transfer versus double embryo transfer in in vitro fertilization. N Engl J Med 2004; 351: 2392–402. 83. Thurin A, Hardarson T, Hausken J et al. Predictors of ongoing implantation in IVF in a good prognosis group of patients. Hum Reprod 2005; 20: 1876–80. 84. Henman M, Catt J, Wood T et al. Elective transfer of single fresh blastocysts and later transfer of cryostored blastocysts reduces the twin pregnancy rate and can improve the in vitro fertilization live birth rate in younger women. Fertil Steril 2005; 84: 1620–7.



HPE_Chapter09.qxp

7/13/2007

4:49 PM

Page 122

HPE_Chapter10.qxp

7/13/2007

4:50 PM

Page 123

10. Human embryo cryopreservation and its effects on embryo morphology James J Stachecki and Klaus Wiemer

CRYOPRESERVATION AND ITS EFFECTS ON EMBRYO MORPHOLOGY

Cryopreservation may have a variety of effects on embryo morphology, ranging from subtle damage on intracellular organelles, cytoplasm, or processes that can negatively affect normal cell development, to overtly blatant effects such as the lysis of one or more blastomeres. Many underlying factors can influence the outcome of cryopreservation, e.g. the quality of the oocyte or embryo itself will often determine the effects of the process on subsequent morphology. The type of freezing protocol used (slow-cooling or vitrification) and the concentration and type of cryoprotectant(s) will also have an effect on subsequent morphology. The goal of this chapter is to offer an insight into the question: how does cryopreservation affect subsequent embryo morphology? We address this question in three major sections: background, types of cryopreservation damage, and causes of damage.

BACKGROUND

Cells have been successfully cryopreserved for over 100 years. Glycerol was discovered to have protective effects on sperm cells in 1949, and since that time both gametes and embryos have been successfully frozen from a number of mammalian species including mice, sheep, cows, pigs, horses, hamsters, rats, cats, and humans. The foundation of modern cryobiology was established throughout the 1940s, 1950s, and 1960s by Lovelock, Meryman, Polge, Smith, Levitt, Luyet, Mazur, and others.1–4 Further advances in animal oocyte and embryo cryopreservation were achieved during the 1970s by Whittingham, Willadsen, Leibo, Mazur, and Wilmut,5–7 and these prepared the way for human oocyte and embryo storage

in the 1980s. Protocols for human embryo storage were refined during the subsequent years up to 2000, so that they have now become a routine procedure in IVF clinics throughout the world. Cryopreservation exposes embryos and oocytes to numerous types of stress, and the fact that these cells can survive and go on to form a viable fetus after freezing and thawing is truly remarkable. However, freeze–thawing protocols are not perfect, and many cells do not survive and/or go on to develop normally. CONSEQUENCES OF CRYOPRESERVATIONINDUCED DAMAGE

Damage can occur at any time throughout the cryopreservation process, and may be manifested in different ways and at different times. Extreme types of damage such as intracellular ice formation (IIF) and cell fracture will lead to immediate cell lysis and death, and these are easily documented through routine microscopic observation of morphology. Damage can also occur on a cellular structural/functional level, involving intracellular organelles, and this is more difficult to diagnose.8 There is very little ultrastructural data documenting morphological damage in cryopreserved oocytes and embryos. However, differences in appearance between fresh and cryopreserved embryos are sometimes mentioned in the discussion section of manuscripts that focus on various freezing techniques. From these studies we can gather at least some information about the effect of cryopreservation on embryo morphology. However, reported differences in morphology postcryopreservation may not be universal, and are likely to be due to the specific protocol and overall set of circumstances used in the study. Cryopreservation procedures may not inevitably result in specific observable morphological aberrations, but the embryo might still be 

HPE_Chapter10.qxp

7/13/2007

4:50 PM

Page 124


affected. In other words, the fact that one investigator using propanediol finds 80% embryo mortality, does not necessarily mean that propanediol is a toxic or inferior cryoprotectant. Likewise, if the same investigator shows 95% embryo survival with dimethylsulfoxide (DMSO), this does not mean that DMSO is the best or the safest cryoprotectant to use. Ultimately, regardless of morphological appearance, the most important feature is that an embryo should remain viable after cryopreservation.

cryoprotectant concentration is high enough so that the remaining intracellular water will vitrify, preventing IIF. (6) Thawing and rehydration: during thawing, the dehydrated cells are exposed to hypotonic conditions and rehydrate as the cryoprotectant is removed.12

CRYOPRESERVATION THEORY

INTRACELLULAR COMPONENTS

A basic understanding of the cryopreservation procedure is important in order to better comprehend common morphological characteristics observed following cryopreservation. The first theoretical basis for cryopreservation of cells was proposed by Mazur3 and later applied to mouse embryos5,9,10 and other species including cows and sheep.7,11 Conventional cryopreservation methods consist of several steps:

The effects of cryopreservation are not always evident; embryos may be adversely affected by cryopreservation at the intracellular level, with the potential of altering the function of the intracellular organelles and cytoplasm. Protein structure and function, as well as metabolism can also be affected. It is likely that embryos require a period of ‘recovery’ following cryopreservation, before they are able to continue normal intracellular function. Therefore, those embryos that are able to compensate for lost or decreased function are the ones most likely subsequently to implant and develop further. We have noted that human embryos following thawing often begin to develop initially slightly slower than their fresh in vitro counterparts. Following extended culture, frozen embryos may resume normal rates of development prior to transfer.

(1) Pre-equilibration: embryos are exposed to a simple salt solution containing a permeable cryoprotectant (1,2-propanediol, DMSO, glycerol, ethylene glycol, etc.) and usually a low concentration of non-permeable cryoprotectant (sucrose). (2) Cooling: after a brief time of exposure to allow uptake of cryoprotectant and initial dehydration, the cells are cooled rapidly to a temperature slightly below the melting point of the solution (usually around ⫺7⬚C). (3) Seeding: at this point the container with the cells is super-cooled in a process known as ‘seeding’ so that ice forms in the extracellular solution. (4) Slow cooling: upon ice formation and further cooling at a slow rate (usually ⬍1⬚C/min to below ⫺30⬚C), the osmolarity of the extracellular solution increases as water freezes to ice, causing the cells to dehydrate with the increasing tonicity. (5) Plunging/vitrification: dehydration continues during slow-cooling until the cells are plunged into liquid nitrogen, usually at a temperature below ⫺30⬚C. At this point the intracellular 

TYPES OF DAMAGE

EMBRYO ORIGIN

The origin of the embryo itself may have a profound impact on the survival of embryos following cryopreservation. In cattle, we have noted that in vivo oocytes have a large perivitelline space; cleaved embryos have uniform blastomeres, and compaction at the morula stage is not irregular. In most cases, expanded blastocysts contain distinct inner cell masses that are not dark in appearance. In contrast, in vitro 1-cell embryos have a very small perivitelline space, blastomeres are irregular in early cleavagestage embryos, and morulae tend to exhibit irregular patterns of compaction. Resulting blastocysts retain a dark appearance and a slightly more irregular shape.

HPE_Chapter10.qxp

7/13/2007

4:50 PM

Page 125

HUMAN EMBRYO CRYOPRESERVATION AND ITS EFFECTS ON EMBRYO MORPHOLOGY

Evidence suggests that in vitro developed bovine embryos may be more sensitive (to cryopreservation) than their in vivo counterparts.13,14 A reduced implantation rate for in vitro produced blastocysts following cryopreservation was noted. Reduced survival and pregnancy rates after transfer of in vitro produced bovine embryos when compared with in vivo embryos has also been reported by others.15,16 The reason for this is unclear, other than the fact that in vivo produced embryos are inherently different to in vitro embryos. Some of the inherent differences may be due to suboptimal maturation and culture systems, inability to adapt cryopreservation methodologies to suit the kinetics of in vitro embryos, and inabilities to manipulate energy substrates or biochemical pathways following thawing. INTRACELLULAR LIPIDS

It is well known that embryos from certain species including goat, pig, and cow are substantially more sensitive to injury from a reduction in temperature than are mouse or human embryos. This chill sensitivity or direct chilling injury is defined as an irreversible damage from exposure to low temperatures. Martino et al17 describe the effect of embryo stage, cooling temperature, and cooling duration on subsequent fertilization and development in vitro of bovine oocytes. Leibo and Loskutoff18 noted that in vitro and in vivo cattle embryos have different buoyant densities. In vivo embryos will sink in 2.35 M sucrose solutions, while their in vitro counterparts will float in solutions containing more than 1.6 M sucrose. This may be due to altered ratios of lipids to proteins found in cell membranes. In vitro produced bovine embryos have a higher lipid content, making them less buoyant, and this lower buoyancy renders the embryo more sensitive to chilling and freezing when compared with its in vivo counterpart. Horvath and Seidel19 pointed out that membranes with higher cholesterol concentrations are more fluid at lower temperatures, and are thus more chillresistant. They demonstrated this effect by loading cumulus–oocyte complexes with cholesterol-loaded methyl-beta-cyclodextrin; these cholesterol-loaded oocytes showed a higher survival after vitrification

than did the non-loaded controls. However, no data on further development and birth were given. Arav et al20 also showed that there is a difference in lipid phase transition, and the temperature at which this phase transition took place was directly related to the effect of temperature on the plasma membrane. Phase transition is simply the progression from one phase or physical state to another. For example from liquid to solid or from liquid to gas. During cooling, ‘liquid’ water will progress to ‘solid’ ice at a warmer temperature than would oil or lipids, or the cryoprotectants propanediol and DMSO. Differences in lipid phase transition were observed between in vivo and in vitro matured oocytes. This suggests that alteration of membrane composition affects chill sensitivity, and subsequently survival following freezing. Nagashima et al21 showed that chilling sensitivity of porcine embryos was directly related to their lipid content. Embryos that were partially or fully delipidated survived cooling better than control non-delipidated embryos, and delipidated non-cooled embryos did not differ in their development compared with control embryos. In a more recent manuscript, Beeb et al22 noticed that early stage porcine blastocysts could survive vitrification and produce piglets following centrifugation of lipids within the embryos prior to cooling. This was the first report of piglets from frozen early-stage blastocysts with lipid removal. It may be possible that the variation found in in vitro human embryos is in part due to the amount and distribution of lipids found in the cell membranes. In accordance with this hypothesis, Ghetler et al23 showed that human zygotes had a higher resistance to chilling injury compared with oocytes of different stages of maturation. Since many species that have been studied to date exhibit variable, and more specifically, higher lipid content in the cell membranes of in vitro derived embryos compared to their in vivo counterparts, it is reasonable to suppose that human in vitro derived embryos may also exhibit this variation. INTRA- AND INTERSPECIES DIFFERENCES

The effects of cryopreservation on intracellular organelles may not be limited to in vitro produced embryos. 

HPE_Chapter10.qxp

7/13/2007

4:50 PM

Page 126


A recent study suggests that high quality cattle embryos from different breeds appeared morphologically similar to their non-frozen counterparts.24 The type of freezing protocol used did have a slight effect on cellular appearance within the embryo itself. At the intracellular level, embryos from two types of cattle exhibited minor cytoplasmic injuries as well as vacuolization of the nucleus. These types of findings are similar to those found in cells that have been subjected to hyperosmotic stress. The authors also noted that embryos from Bos tarus had higher amounts of intracellular lipids than embryos from Bos indicus, and notably embryos from Bos tarus had a slightly better morphological appearance than their Bos indicus counterparts. Likewise, Dinnyes et al25 showed that embryos from different strains of mice treated with the same cryopreservation protocol responded differently, with variable survival and development rates. The in vivo development of cryopreserved embryos was also influenced by genotype and cryopreservation method (slow-cooling vs vitrification). Although genotype did not affect the ability of embryos to develop (all the embryos could develop), the differences in overall embryo development after cryopreservation were related to the genotype and not the sensitivity to cryo injury. In other words, differences in the developmental ability of embryos of different genotypes became apparent only after cryopreservation. Furthermore, variability often noted within human embryos from different patients may be due to the genetic background of the patient population. The ‘freezablity’ of human embryos from different genetic backgrounds has never been evaluated. CORTICAL GRANULES

Cortical granule release occurs naturally during egg activation and/or fertilization. The release of cortical granules leads to zona hardening, as a natural block to polyspermy. Zona hardening due to both cortical granule release and culture in vitro, may impair the embryo’s ability to hatch during the time of implantation. Cryopreservation may also contribute to zona hardening over and above what would occur naturally, and this may affect blastocyst hatching.26 

Likewise, oocyte freezing may cause premature hardening of the zona, an effect possibly related to premature cortical granule release as observed in several species.27–32 Vincent et al27 showed that exposure to DMSO caused a reduction in cortical granules in mouse oocytes, and Ghetler et al32 similarly found that exposure to propanediol causes cortical granule release from human oocytes, observing a significant reduction of cortical granules in electron micrographs of frozen–thawed oocytes. However, Wood et al29 discovered that under certain conditions, cortical granule release could be avoided during cryopreservation, and that other alterations in the zona pellucida were to blame for reduced fertilization rates of cryopreserved mouse oocytes. Whatever changes occur to modify the zona following cryopreservation, this can be overcome by choosing either artificial zona opening or ICSI as the method of choice for insemination of cryopreserved oocytes.33–37 MEIOTIC SPINDLE

Detailed morphological analysis of the nucleus and/or spindle (oocytes) requires the use of a PolScope, immunohistochemistry, FISH, or other specialized technique. Spindle re-formation and function play an important role in the further development of the oocyte following cryopreservation, and numerous studies on the effect of freezing on spindle morphology have been reported (for reviews see references 38 and 39). Improper chromosome segregation could lead to aneuploidy and genetic errors, which may result in embryonic and fetal abnormalities. With the use of numerous different cryopreservation protocols, a wide variety of outcomes have been reported regarding spindle morphology, as well as actin and microfilament disruption postcryopreservation. These range from almost 0% to nearly 100% re-formation and normal spindle morphology.8,40 This range of results makes it difficult to interpret the true situation with regards to potential spindle damage after cryopreservation. In a recent report by Stachecki et al38 a more global view of spindle disassembly and reassembly was presented by investigating the effects of cryopreservation on spindle morphology in three evolutionarily distinct mammalian species: mouse,

HPE_Chapter10.qxp

7/13/2007

4:50 PM

Page 127


bovine, and human. Their observations suggest that other results may be misleading, due to the fact that the majority of the investigators report only 35–65% oocyte survival after thawing. Therefore, if the freezing technique were sufficiently stressful to kill half of the eggs, it is likely that the remaining oocytes might not have normal function, and this may be manifested in poor or improper spindle re-formation. Stachecki et al38 further demonstrated that their cryopreservation technique yields significantly higher survival rates with oocytes from all three species after freeze–thawing, in the range of 80–100%. When returned to a physiological temperature (37–39⬚C) for 30–90 minutes, nearly all of the oocytes were able to reassemble their spindle, and the majority (ⱖ70%) had a normal barrel-shaped spindle with chromosomes aligned. Results for frozen–thawed oocytes were statistically similar to non-frozen control oocytes. It is important to bear in mind that the appearance of the spindle does not necessarily correlate with its function: a spindle that does not have a perfect appearance may nonetheless have normal function, and vice versa. Therefore, spindle function needs to be further investigated by analyzing second polar bodies and/or the genetic makeup of embryos that result from frozen eggs. Interestingly, there has been very little attention to spindle function associated with embryo freezing. Although most blastomeres are in interphase and a spindle is not present during cryopreservation, a spindle still has to form in order for the cell to progress through mitosis, and freeze– thawing could have detrimental effects on this process. EMBRYO ULTRASTRUCTURE

Cryopreservation can have effects on the zona pellucida and cell membrane, in addition to intracellular components. For example, the zona can be damaged by cracking, splitting, elongation, and distortion, caused by different stresses during freezing and/or thawing. However, the zona is not always damaged during cryopreservation, or it may be damaged to differing degrees, some of which may not be microscopically apparent. Anyone who has dropped an ice cube into a glass of warm water, soda, or similar solution has seen and

heard it crack. This is caused by the rapid thermal expansion that occurs within the ice cube when it is removed from the freezer, around ⫺20⬚C, and placed into a substantially warmer environment, such as a 23⬚C liquid, a temperature difference of only 43⬚C. This amount of heat exchange is enough to fracture an ice cube, and yet when a straw or vial is removed from liquid nitrogen at ⫺196⬚C and placed in room temperature air (or water) the temperature difference is over 200⬚C. This dramatic change can and does, at least in some cases, cause fracture damage to the contents of the straw or vial, most often reported as zona cracking. Rall and Meyer41 specifically studied the mechanism behind such injury. They compared thawing rates and surmized that their observations were consistent with the view that zona damage is directly associated with thermal-induced fracturing of the cryoprotectant suspension during rapid changes of temperature that occur during thawing. Stachecki et al42,43 specifically studied thawing rates of mouse oocytes and observed similar evidence of zona cracking from thawing too rapidly. They also reported significant changes in survival and development with different thawing regimens. Moreira da Silva and Metelo44 recently described the effects of cryopreservation methods on the physical properties of the zonae pellucidae of in vitro produced bovine embryos, in order to explain the loss of embryo developmental capacity following freezing and thawing. These authors noted that pore size in the zona was correlated with viability following cryopreservation. When bovine embryos were frozen either by slow methods or by vitrification, pore size and subsequent viability was most affected by vitrification. Pore size was smallest (0.27 ␮m) after vitrification, compared with 0.34 ␮m for slow cooled and 0.48 ␮m for the control embryos, respectively. Of interest, the survival rate of embryos following vitrification was lower; indicating that pore size within the zona could be related to viability. Results indicate alterations in the zona might be caused by the steps of cryopreservation process itself which may be responsible for irreversible damage on subsequent development of bovine embryos. In many cases, the effects of cryopreservation are not overtly evident. The embryo may appear to 

HPE_Chapter10.qxp

7/13/2007

4:50 PM

Page 128


be microscopically normal following the thawing process, when in fact minute changes have occurred. It is possible that freeze/thawing can induce changes that negatively affect cell membranes and organelles. In the case of frozen–thawed blastocysts, distinctive cell types within the embryos themselves may be affected differently by this process. Detailed effects of these processes on embryos can be assessed by ultrastructure studies, and these have noted that in vitro produced bovine embryos have a higher proportion of degraded trophectoderm and dead inner cell mass (ICM) cells following thawing, when slow freezing methods were used.45 The methodologies used to freeze these embryos greatly affected the proportion of embryos with dead cells within the ICM. Embryos frozen after introducing the cryoprotectant in one step had a higher proportion of necrotic cells within the ICM than did blastocysts frozen after introducing the cryoprotectant in three steps. These authors also noted that the basic salt solutions used for cryopreservation influenced the proportion of necrotic cells within the ICM, and that necrotic cells were extruded from the trophectoderm junction when the embryos were cultured prior to fixation. Morulae that were frozen, thawed, and cultured prior to fixation often produced blastocysts that were delayed in development, with reduced cell numbers within the trophectoderm as well as the ICM. The ultrastructural effects of cryopreservation on equine embryos have also been studied.46 Blastocysts were non-surgically recovered from mares, and some were frozen using a medium containing 10% glycerol with a slow freezing protocol. Embryos were thawed and fixed for subsequent evaluation. Wilson et al46 observed that embryos that were exposed to glycerol only (without freezing) had changes in lipid droplets within the ICM and changes in the appearance of mitochondria. Embryos that were cryopreserved prior to fixation also had structural changes that were associated with the cryoprotectant used, as well as the freezing procedure itself. Specifically, the mitochondria associated with the ICM and trophoblast were affected, and the greatest damage was associated with the cells within the ICM. These data suggest that glycerol did not sufficiently permeate the embryo prior to freezing. In other words, proper dehydration 

may have not occurred, allowing ice crystals to form. This would be particularly true for cells within the ICM, since the reduced permeation of glycerol through the embryo would be most evident within the ICM. This might be due to the presence of junctional processes between the trophoblast cells that could reduce the permeation of cryoprotectants into the ICM. For the cryoprotectant to come in contact with the ICM, it must first pass through the trophoblast cells that might limit permeation because of the junctional processes or some other physical (cell or membrane) barrier. Ultrastructural analyses of in vitro cultured human blastocysts also revealed structural damage.47 This study revealed that tight junctions could not be detected following cryopreservation and thawing of expanded blastocysts, although cells remained in close apposition; however, tight junctions were detected in the cryopreserved specimens. The collapse of the cavity during the freeze–thaw process may have caused the disruption of the apically located zona occludentes. More recently, Escriba et al48 detected no ultrastructural changes following cryopreservation of human blastocysts. These authors noted that epithelial junctions formed by apical tight junctions and basal desmosomes remained intact and polarized. Cells within the ICM retained their shape and intercellular junctions. These data differ from those previously reported by Wiemer et al47 using a different freezing protocol. Escriba et al48 used vitrification techniques, whereas the embryos frozen by Wiemer et al47 were frozen using a slow freezing technique. It seems that vitrification may be the best method for freezing human embryos. More recently Wiemer et al (unpublished data) used an optimized vitrification method known as S3-vitrification, with an improvement in results. With donated or spare material, survival rates of blastocysts have been in excess of 90%. In addition, we have achieved a pregnancy rate in excess of 50% in a small group of patients. Nottola et al (personal communication) analyzed fresh and frozen thawed human oocytes using light microscopy and transmission electron microscopy (TEM). Oocytes were frozen with a conventional slow cooling protocol using 0.1 mol/l or 0.3 mol/l

HPE_Chapter10.qxp

7/13/2007

4:50 PM

Page 129


sucrose, and oocytes had fairly good preservation and normal organization of the cytoplasm following freezing. There was a reduced amount and density of cortical granules in some of the frozen cells. They also found slight to moderate vacuolization, specifically in the group with 0.3 mol/l sucrose. The spindles in the oocytes were morphologically normal after thawing, and a normal complement of mitochondrial aggregates was found in both fresh and frozen oocytes. It is difficult to determine all of the effects of freezing, as only the oocytes that survived were analyzed; however, the authors suggest that the use of high sucrose concentrations in freezing protocols should be approached with care until more information is known about its effects on embryo development. EMBRYO QUALITY

The initial quality of embryos prior to cryopreservation is also a determining factor in their resistance to the freezing process. In a recent study, cattle embryos of differing quality were cryopreserved, thawed, and evaluated at the ultrastructural level.49 Apoptosis was assessed using terminal deoxynucleotidyl transferase mediated dUTP nick end labeling (TUNEL) for embryos of varying quality prior to cryopreservation, and both electron microscopy and TUNEL analysis showed that the number of lyzed cells and apoptotic cells increased following cryopreservation. The degree of apoptosis was directly related to the initial morphology of the embryo prior to freezing: embryos of good morphology had significantly lower degrees of apoptosis than embryos of poor morphology. In higher quality blastocysts, apoptotic cells were more prevalent within the ICM, whereas in blastocysts of lesser quality, apoptotic cells were randomly distributed. The data suggest that embryos of higher quality (based on their appearance) have better resistance to the damage that is often associated with cryopreservation. Of interest, seasonal effects were also noted in this study. Embryos collected during the seasonally stressful time, when natural forage is less abundant, had a higher proportion of lyzed cells following cryopreservation than embryos produced during more ideal conditions.

As previously mentioned, embryos that are derived from IVF might be more susceptible to effects induced by cryopreservation. Similar to data found with cattle embryos, the quality of embryos prior to cryopreservation has a significant effect on the morphology following thawing. Both cattle and human embryos are affected by the osmotic effects associated with the introduction and removal of cryoprotectants, and tight and gap junctions were disrupted in human embryos.47 In cattle, there was evidence of vacuolization of the nucleus,24 apoptotic cells were present within the inner cell mass of high quality blastocysts, or throughout the cells of lower quality cattle embryos.49 In some cases, IVF centers prefer to freeze cleavage stage embryos at the 8-cell stage, with a rationale that their extended culture conditions may not be optimal for blastocyst culture, or their experience with blastocyst cryopreservation is less than optimal. Freezing embryos at the 8-cell stage (day 3) allows high quality embryos to be frozen immediately following the embryo transfer procedure. The advantage of this practice is that no further culture is required. This practice works well in order to reduce the work load in the IVF laboratory as well as laboratories that have poor blastocyst development rates. Freezing 8-cell embryos often maximizes the chance of pregnancy from a single stimulated cycle, after frozen embryo transfer. In general, only embryos of the highest quality are frozen, due to the stresses involved with cryopreservation, which can further reduce an embryo’s chance of full development. Additionally, embryos of suboptimal morphological quality sometimes, but not always, have poorer survival rates and a significantly higher proportion of lyzed blastomeres. The transfer of embryos with lyzed/degenerate blastomeres is associated with lower pregnancy and implantation rates.50 The same authors noted that the birth rate was three times higher after the transfer of fully intact embryos, when compared with the transfer of damaged embryos. No attempt to remove any damaged cells was carried out in this study. The reason for blastomere lysis in high quality embryos undergoing cryopreservation is unclear. It is possible that these cells lyze due to the presence of intracellular ice caused by incomplete dehydration 

HPE_Chapter10.qxp

7/13/2007

4:50 PM

Page 130


of the cell, reduced permeability, membrane weakness, or inappropriate thawing conditions. The potentially deleterious effect of contraction and reexpansion of the cells during dehydration and rehydration processes may also be a possible cause. Furthermore, the mechanism whereby embryos become compromised by the presence of lyzed cells is not clear, but it is possible that the lyzed and/or damaged cell might create a locally ‘toxic’ condition that could affect subsequent development. Cohen et al51,52 developed the assisted hatching technique in order to bypass potential implantation failure associated with impaired embryo hatching, and this also allows access to necrotic cells or fragments, which can be removed via the breach created in the zona pellucida. Assisted hatching has been most advantageous for patients that have a history of producing suboptimal embryos or those with thickened zonae,53–55 and it has also been applied to thawed cleavage stage embryos. Following assisted hatching and removal of necrotic cells, pregnancy rates increased from 17% to 45.7% compared with embryos that did not have necrotic cells removed.56 Other studies have also noted that the removal of lyzed cells following assisted hatching improved implantation and pregnancy rates.57,58 Based upon personal experience as well as the current literature, it seems that removal of lyzed blastomeres in freeze–thawed embryos after assisted hatching may be advantageous, both in alleviating impaired hatching due to potential zona hardening, and in order to eliminate potentially toxic effects of necrotic products. CELLULAR FUNCTION

The effects of cryopreservation on cellular metabolism are difficult to assess morphologically, but this important aspect must also be considered. Disruption of mitochondrial membranes leads to loss of protons and reduces the oxidative potential for ATP production. Rieger et al59,60 noted that cryopreservation of horse and cattle embryos caused an increase in glutamine production, possibly due to a disruption in mitochondrial ATP production and therefore a flux in the Krebs cycle. Gardner et al61 also noted that the 

process of freezing and thawing on IVF-produced bovine blastocysts had a significant effect on nutrient uptake and utilization. The freeze–thaw process had a negative impact on the rate of glucose and pyruvate uptake as well as lactate production, and the viable embryos did not recover the metabolic activity that was recorded prior to freezing. Gardner et al61 noted that damage to the mitochondria may have resulted in changes in oxidative phosphorylation, thus increasing oxygen consumption. The review of intracellular effects of cryopreservation by Smith and Silva8 mentions that cryopreservation may alter the nuclear envelope, causing downstream disruption in replication and/or transcription. Van Blerkom62 showed that although germinal vesicle (GV)-stage mouse oocytes survived vitrification and were capable of resuming meiosis and undergoing normal chromosomal and cytoplasmic maturation to metaphase II, profound alterations in the structure and organization of the cytoplasm, nucleus, nucleolus, and chromatin occurred during the dehydration stage. The majority of cytoplasmic and nuclear perturbations returned to normal postthawing, but the potential for adverse development after fertilization remains. Enzymatic regulation and protein structure/function before and after cryopreservation have not yet been assessed.

CAUSES OF DAMAGE

It is not always easy to determine the exact causes of damage that can occur throughout the cryopreservation process, since the dysmorphism(s) induced can have several origins. The problems that can occur during cryopreservation differ in association with slow cooling and with vitrification. Problems associated with slow cooling that have been described include IIF and osmotic effects, whereas chemical toxicity is a major obstacle with current vitrification techniques. However, this interpretation may not be entirely correct, because it does not encompass sodium-loading, other ion effects, and the possibility of problems as yet undiscovered. Damage can occur at any step in the cryopreservation process, and the variety of aberrations that

HPE_Chapter10.qxp

7/13/2007

4:50 PM

Page 131


can occur is dependent upon the protocol used for cryopreservation. Some protocols and media result in less damage, and close examination of each step in the procedure can give an indication of the common types of damage that can occur. Two types of cryoprotectants are commonly used to preserve human oocytes and embryos: permeable and impermeable. Permeable cryoprotectants include propanediol, glycerol, DMSO, and ethylene glycol. Sucrose and trehalose are impermeable. All of these cryoprotectants cause the oocyte/embryo to undergo changes in osmolarity. The subsequent contraction and expansion known collectively as solution effects might be partially responsible for some of the changes commonly found in embryos following cryopreservation. During pre-equilibration, the cell is exposed to a hyperosmotic environment that allows some dehydration, along with uptake of permeable cryoprotectant(s). If the dehydration is too severe, membrane damage can occur, sometimes resulting in blebbing of the membrane. The cryoprotectant can be toxic to the cell if the concentration is too high. The toxicity of cryoprotectants is also related to the temperature at which they are used: the higher the temperature, the greater the toxicity.63 Cryoprotectant toxicity may result in immediate cell lysis, or lysis after thawing. The pre-equilibration process is usually performed in a series of steps, in order to reduce the stress of dehydration and facilitate cryoprotectant uptake. During initial cooling, temperature shock may occur and damage the cell, resulting in lysis or degeneration following thawing. Temperature shock is most likely to occur in species that are ‘chill-sensitive’, including bovine and porcine,17,21 whereas mouse and human embryos seem better able to tolerate cooling from incubator temperatures to zero or below. The process of seeding can also cause cell damage. Seeding is usually done by touching the side of a cryovial or straw with liquid nitrogen-cooled forceps, and if the cooled area is too close to the cells themselves, they may freeze and then die upon thawing. Seeding is usually carried out at a position away from the location of the cells, but damage can still occur if the seeding temperature is not appropriate.64,65

For example, if seeding is carried out at a temperature of less than ⫺10⬚C, the ice crystal may form and grow too rapidly, causing heterogeneous ice formation intracellularly. The cell may also become deformed in supercooled areas of liquid between growing ice crystals. If the seeding temperature is too high, above ⫺4.5⬚C, the ice crystal may melt and never progress, with very rapid extracellular ice formation when the temperature falls below ⫺15⬚C, resulting in possible cell death upon thawing. During slow cooling, the remaining solutes become more concentrated as water freezes, and this exerts hypertonic pressure on the cell, resulting in its further dehydration and osmotic stress. At this time, the concentration of cryoprotectant increases to potentially lethal levels. If the cell is not sufficiently dehydrated during the slow cooling process, and the intracellular concentration of cryoprotectant is not sufficient to intercalate with the residual water inside the cell, intracellular ice may form and kill the cell. If the cell is excessively dehydrated, it may be incapable of rehydrating sufficiently to resume normal function after thawing.66 Numerous types of injury can occur during thawing. The process of taking a cell from a resting temperature of ⫺196⬚C and warming it to 0⬚C or higher over the period of a minute or less, is an extreme temperature change. For example, a straw is usually held in room temperature air for a period of time before further warming in a water bath (usually 30⬚C). Thawing that is too rapid can result in large and/ or small fractures in the zona pellucida or the cell itself, as described previously. These fractures are caused by a non-uniform change in the volume of the medium during rapid phase changes that occur during thawing. 41,67 If the cell membrane fractures it will lyze immediately, but even if the cell membrane does not fracture, intracellular components may be fractured or damaged leading to the eventual demise of the cell. During re-warming, ice can form once again if the vitrified solution warms at a rate that permits the process of devitrification. This occurs when the temperature reaches a point where the molecular mobility of water increases so that water molecules can move and rearrange themselves from a disorderly amorphous vitrified position to an orderly 

HPE_Chapter10.qxp

7/13/2007

4:50 PM

Page 132


crystalline position. This happens well below the melting point, and is therefore a potentially lethal problem.68,69 IIF at this stage can lead to cell lysis and/or damage to organelles or other intracellular components. Damage of this type may not be apparent from simple observation of morphology. During the final step of thawing, the previously frozen cell will be very dehydrated and must undergo rehydration and removal of cryoprotectant(s) in order to continue development. Because water permeates more rapidly than does cryoprotectant, the cell may swell and lyze in the process of trying to remove the cryoprotectant. Sucrose is usually used in the step-out process in order to reduce the osmotic effects.70,71 Damage at this stage or during the next several hours of culture may be manifested as cell expansion and rupture, lysis, or a darkening of the cytoplasm and cell death. Sometimes the cell appears to have survived thawing, but after culture fails to develop, lyzes, or degenerates. Kasai et al71 discussed embryo dysmorphism following cryopreservation; they tried to mimic common types of cellular damage by subjecting mouse blastocysts to a range of treatments, including the use of excess cryoprotectant to induce toxic injury, inducing IIF, exposure to hyper- and hypotonic solutions, and rapid thawing to induce zona/cytoplasm fractures. By specifically trying to induce damage they were able to relate the morphological appearance of the cell to the type of cryodamage, and clearly demonstrated a relationship between different types of damage and the method of cryopreservation. Stachecki et al43 showed that it is relatively easy to freeze mouse eggs using a standard slow cooling protocol with a high break-point (the temperature at which the cells are plunged into liquid nitrogen or solidified) of ⫺20⬚C, compared with the more conventional ⫺30⬚C break-point. It is often that we learn more by experiments that supposedly ‘fail’ than when experiments ‘work’ and all of the cells survive, and through this experiment that attempted to kill oocytes under stressful experimental conditions, it was found that mouse eggs could be exposed to stresses previously thought impossible, and still form viable pups.



CONCLUSIONS

Although embryo cryopreservation has been used for over 25 years, there are still many aspects that we do not understand about the process. The majority of studies report only morphological observations of survival and subsequent development rates. These reports have allowed modifications to protocols that have improved the success of cryopreservation, but there is still room for improvement. A few investigations have delved into the subcellular alterations that can occur during freezing and thawing, and these more detailed studies have given us a more complete understanding of the effects of freezing on embryos and oocytes. Although a vast array of cellular and subcellular components and processes can be affected by cryopreservation, the resilience of the embryo in resisting and/or adapting to these changes is remarkable. As we have alluded to above, the nucleus, cytoplasm, internal and external membranes, etc. are all subject to alteration during the process of freezing. Some of these alterations, such as lyzed blastomeres, are manifested upon thawing, and others, including genetic abnormalities, are apparent only following subsequent development over time. Therefore, early as well as late effects must be considered when optimizing a cryopreservation protocol. As mentioned throughout this chapter, there are many similarities between human embryos and cattle embryos. The ability to study cattle embryos or those of other species may allow us to understand more fully the effects that cryopreservation has on human embryos. Indeed, most of the current vitrification protocols have been used for years with cattle embryos and have only recently been tested on human embryos. Bovine and mouse embryos are readily available, and they therefore provide good models for testing cryopreservation protocols. The data obtained from these models can serve as a general baseline of what to expect when trying to adapt protocols to humans or other mammalian species. However, more in depth analyses, specifically with human embryos and eggs, is necessary in order to optimize freezing and thawing protocols.

HPE_Chapter10.qxp

7/13/2007

4:50 PM

Page 133


REFERENCES 1. Luyet B, Gehenio P. Life and death at low temperatures. In: Life and death at low temperatures. Luyet B, Gehenio P eds. Biodynamica. Normandy, MO; 1940: 61. 2. Lovelock J. The haemolysis of human red blood cells by freezing and thawing. Biochimice et Biophysica 1953; 10: 414–26. 3. Mazur P. Causes of injury in frozen and thawed cells. Fed Elsever Proc 1965; 24(Suppl 15): 175–82. 4. Meryman H. The exceeding of a minimum tolerable cell volume in hypertonic suspension as a case of freezing injury. In: Wolstenholme G, O’Connor M, eds. The Frozen Cell. London: J & A Churchill, 1970: 51–67. 5. Whittingham DG, Leibo SP, Mazur P. Survival of mouse embryos frozen to ⫺196 degrees and ⫺269 degrees C. Science 1972; 178(59): 411–4. 6. Wilmut I. The low temperature preservation of mammalian embryos. J Reprod Fertil 1972; 31(3): 513–4. 7. Willadsen SM. Factors affecting the survival of sheep embryos during freezing and thawing. In: The freezing of mammalian embryos. Amsterdam: North-Holland Publishing Co. 1977; 175–201. 8. Smith GD, Silva ESCA. Developmental consequences of cryopreservation of mammalian oocytes and embryos. Reprod Biomed Online 2004; 9(2): 171–8. 9. Whittingham DG, Anderson E. Ultrastructural studies of frozenthawed 8-cell mouse embryos. J Reprod Fertil 1976; 48(1): 137–40. 10. Mazur P. Freezing of living cells: mechanisms and implications. Am J Physiol 1984; 247(3 Pt 1): C125–42. 11. Willadsen S, Polge C, Rowson LE. The viability of deep-frozen cow embryos. J Reprod Fertil 1978; 52(2): 391–3. 12. Mazur P. Slow-freezing injury in mammalian cells. Ciba Found Symp 1977; (52): 19–48. 13. Hasler JF, Hurtgen PJ, Jin ZQ, Stokes JE. Survival of IVF-derived bovine embryos frozen in glycerol or ethylene glycol. Theriogenology 1997; 48(4): 563–79. 14. Hasler JF. Factors affecting frozen and fresh embryo transfer pregnancy rates in cattle. Theriogenology 2001; 56(9): 1401–15. 15. Kajihara Y, Kometani N, Shitanaka Y et al. Pregnancy rates and births after direct transfers of frozen-thawed bovine IVF embryos. Theriogenology 1992; 37: 233. 16. Wurth YA, Reinders JMC, Rall WF, Kruip TAM. Developmental potential of in vitro produced bovine embryos following cryopreservation and single-embryo transfer. Theriogenology 1994; 42: 1275–84. 17. Martino A, Pollard JW, Leibo SP. Effect of chilling bovine oocytes on their developmental competence. Mol Reprod Dev 1996; 45(4): 503–12. 18. Leibo SP, Loskutoff NM. Cryobiology of in vitro derived bovine embryos. Theriogenology 1993; 39: 81–94. 19. Horvath G, Seidel GE, Jr. Vitrification of bovine oocytes after treatment with cholesterol-loaded methyl-beta-cyclodextrin. Theriogenology 2006; 66(4): 1026–33. 20. Arav A, Zeron Y, Leslie SB et al. Phase transition temperature and chilling sensitivity of bovine oocytes. Cryobiology 1996; 33(6): 589–99. 21. Nagashima H, Kashiwazaki N, Ashman RJ et al. Removal of cytoplasmic lipid enhances the tolerance of porcine embryos to chilling. Biol Reprod 1994; 51(4): 618–22. 22. Beeb LF, Cameron RD, Blackshaw AW, Higgins A, Nottle MB. Piglets born from centrifuged and vitrified early and peri-hatching blastocysts. Theriogenology 2002; 57(9): 2155–65.

23. Ghetler Y, Yavin S, Shalgi R, Arav A. The effect of chilling on membrane lipid phase transition in human oocytes and zygotes. Hum Reprod 2005; 20(12): 3385–9. 24. Visintin JA, Martins JF, Bevilacqua EM et al. Cryopreservation of Bos taurus vs Bos indicus embryos: are they really different? Theriogenology 2002; 57(1): 345–59. 25. Dinnyes A, Wallace GA, Rall WF. Effect of genotype on the efficiency of mouse embryo cryopreservation by vitrification or slow freezing methods. Mol Reprod Dev 1995; 40(4): 429–35. 26. De Vos A, Van Steirteghem A. Zona hardening, zona drilling and assisted hatching: new achievements in assisted reproduction. Cells Tissues Organs 2000; 166(2): 220–7. 27. Vincent C, Pickering SJ, Johnson MH. The hardening effect of dimethylsulphoxide on the mouse zona pellucida requires the presence of an oocyte and is associated with a reduction in the number of cortical granules present. J Reprod Fertil 1990; 89(1): 253–9. 28. Vincent C, Turner K, Pickering SJ, Johnson MH. Zona pellucida modifications in the mouse in the absence of oocyte activation. Mol Reprod Dev 1991; 28(4): 394–404. 29. Wood MJ, Whittingham DG, Lee SH. Fertilization failure of frozen mouse oocytes is not due to premature cortical granule release. Biol Reprod 1992; 46(6): 1187–95. 30. George MA, Johnson MH, Vincent C. Use of fetal bovine serum to protect against zona hardening during preparation of mouse oocytes for cryopreservation. Hum Reprod 1992; 7(3): 408–12. 31. Fuku E, Xia L, Downey BR. Ultrastructural changes in bovine oocytes cryopreserved by vitrification. Cryobiology 1995; 32(2): 139–56. 32. Ghetler Y, Skutelsky E, Ben Nun I et al. Human oocyte cryopreservation and the fate of cortical granules. Fertil Steril 2006; 86(1): 210–16. 33. Kazem R, Thompson LA, Srikantharajah A et al. Cryopreservation of human oocytes and fertilization by two techniques: in-vitro fertilization and intracytoplasmic sperm injection. Hum Reprod 1995; 10(10): 2650–4. 34. Porcu E, Fabbri R, Seracchioli R et al. Birth of a healthy female after intracytoplasmic sperm injection of cryopreserved human oocytes. Fertil Steril 1997; 68(4): 724–6. 35. Stachecki JJ, Cohen J, Willadsen SM. Detrimental effects of sodium during oocyte cryopreservation. Biol Reprod 1998; 59: 395–400. 36. Fabbri R, Porcu E, Marsella T et al. Human oocyte cryopreservation: new perspectives regarding oocyte survival. Hum Reprod 2001; 16(3): 411–6. 37. Stachecki JJ, Cohen J. An overview of oocyte cryopreservation. Reprod Biomed Online 2004; 9(2): 152–63. 38. Stachecki JJ, Munne S, Cohen J. Spindle organization after cryopreservation of mouse, human, and bovine oocytes. Reprod Biomed Online 2004; 8(6): 664–72. 39. Coticchio G, Bonu MA, Bianchi V, Flamigni C, Borini A. Criteria to assess human oocyte quality after cryopreservation. Reprod Biomed Online 2005; 11(4): 421–7. 40. Vincent C, Johnson MH. Cooling, cryoprotectants, and the cytoskeleton of the mammalian oocyte. Oxford Rev Reprod Biol 1992; 14: 73–100. 41. Rall WF, Meyer TK. Zona fracture damage and its avoidance during the cryopreservation of mammalian embryos. Theriogenology 1989; 31: 683–92. 42. Stachecki JJ, Willadsen SM. Cryopreservation of mouse oocytes using a medium with low sodium content: effect of plunge temperature. Cryobiology 2000; 40: 4–12. 43. Stachecki JJ, Cohen J, Schimmel T, Willadsen SM. Fetal development of mouse oocytes and zygotes cryopreserved in a nonconventional freezing medium. Cryobiology 2002; 44(1): 5–13.



HPE_Chapter10.qxp

7/13/2007

4:50 PM

Page 134


44. Moreira da Silva F, Metelo R. Relation between physical properties of the zona pellucida and viability of bovine embryos after slow-freezing and vitrification. Reprod Domest Anim 2005; 40(3): 205–9. 45. Iwasaki S, Yoshikane Y, Li X, Watanabe S, Nakahara T. Effects of freezing of bovine preimplantation embryos derived from oocytes fertilized in vitro on survival of their inner cell mass cells. Mol Reprod Dev 1994; 37(3): 272–5. 46. Wilson JM, Caceci T, Potter GD, Kraemer DC. Ultrastructure of cryopreserved horse embryos. J Reprod Fertil Suppl 1987; 35: 405–17. 47. Wiemer KE, Dale B, Hu Y et al. Blastocyst development in co-culture: development and morphological aspects. Hum Reprod 1995; 10(12): 3226–32. 48. Escriba MJ, Escobedo-Lucea C, Mercader A et al. Ultrastructure of preimplantation genetic diagnosis-derived human blastocysts grown in a coculture system after vitrification. Fertil Steril 2006; 86(3): 664–71. 49. Marquez YC, Galina CS, Moreno N et al. Seasonal effect on zebu embryo quality as determined by their degree of apoptosis and resistance to cryopreservation. Reprod Domest Anim 2005; 40(6): 553–8. 50. Van den Abbeel E, Camus M, Van Waesberghe L, Devroey P, Van Steirteghem AC. Viability of partially damaged human embryos after cryopreservation. Hum Reprod 1997; 12(9): 2006–10. 51. Cohen J, Elsner C, Kort H et al. Impairment of the hatching process following IVF in the human and improvement of implantation by assisting hatching using micromanipulation. Hum Reprod 1990; 5(1): 7–13. 52. Cohen J, Feldberg D. Effects of the size and number of zona pellucida openings on hatching and trophoblast outgrowth in the mouse embryo. Mol Reprod Dev 1991; 30(1): 70–8. 53. Cohen J, Alikani M, Reing AM et al. Selective assisted hatching of human embryos. Ann Acad Med Singapore 1992; 21(4): 565–70. 54. Cohen J. Assisted hatching: indications and techniques. Acta Eur Fertil 1993; 24(5): 215–9. 55. Wiemer KE, Hu Y, Cuervo M, Genetis P, Leibowitz D. The combination of coculture and selective assisted hatching: results from their clinical application. Fertil Steril 1994; 61(1): 105–10. 56. Rienzi L, Nagy ZP, Ubaldi F et al. Laser-assisted removal of necrotic blastomeres from cryopreserved embryos that were partially damaged. Fertil Steril 2002; 77(6): 1196–201. 57. Nagy ZP, Taylor T, Elliott T et al. Removal of lysed blastomeres from frozen-thawed embryos improves implantation and pregnancy rates in frozen embryo transfer cycles. Fertil Steril 2005; 84(6): 1606–12.



58. Rienzi L, Ubaldi F, Iacobelli M et al. Developmental potential of fully intact and partially damaged cryopreserved embryos after laser-assisted removal of necrotic blastomeres and post-thaw culture selection. Fertil Steril 2005; 84(4): 888–94. 59. Rieger D, Bruyas JF, Lagneaux D, Bezard J, Palmer E. The effect of cryopreservation on the metabolic activity of day-6.5 horse embryos. J Reprod Fertil Supplement 1991; 44: 411–7. 60. Rieger D, Loskutoff NM, Betteridge KJ. Developmentally related changes in the metabolism of glucose and glutamine by cattle embryos produced and co-cultured in vitro. J Reprod Fertil 1992; 95(2): 585–95. 61. Gardner DK, Pawelczynski M, Trounson AO. Nutrient uptake and utilization can be used to select viable day 7 bovine blastocysts after cryopreservation. Mol Reprod Dev 1996; 44(4): 472–5. 62. Van Blerkom J. Maturation at high frequency of germinal-vesicle-stage mouse oocytes after cryopreservation: alterations in cytoplasmic, nuclear, nucleolar and chromosomal structure and organization associated with vitrification. Hum Reprod 1989; 4(8): 883–98. 63. Rall WF, Fahy GM. Ice-free cryopreservation of mouse embryos at ⫺196 degrees C by vitrification. Nature 1985; 313(6003): 573–5. 64. Whittingham DG. Some factors affecting embryo storage in laboratory animals. Ciba Found Symp 1977; (52): 97–127. 65. Trad FS, Toner M, Biggers JD. Effects of cryoprotectants and ice-seeding temperature on intracellular freezing and survival of human oocytes. Hum Reprod 1999; 14(6): 1569–77. 66. Mazur P. Equilibrium, quasi-equilibrium, and nonequilibrium freezing of mammalian embryos. Cell Biophys 1990; 17(1): 53–92. 67. Kroener C, Luyet B. Formation of cracks during the vitrification of glycerol solutions and disappearance of the cracks during rewarming. Biodynamica 1966; 10(198): 47–52. 68. Mazur P, Schmidt JJ. Interactions of cooling velocity, temperature, and warming velocity on the survival of frozen and thawed yeast. Cryobiology 1968; 5(1): 1–17. 69. Luyet B. Physical changes occurring in frozen solutions during rewarming and melting. In: Wolstenholme G, O’Connor JA, editors. The Frozen Cell. London: Churchill, 1970; 27–50. 70. Kasai M, Niwa K, Iritani A. Survival of mouse embryos frozen and thawed rapidly. J Reprod Fertil 1980; 59(1): 51–6. 71. Kasai M, Ito K, Edashige K. Morphological appearance of the cryopreserved mouse blastocyst as a tool to identify the type of cryoinjury. Hum Reprod 2002; 17(7): 1863–74.

HPE_Chapter11.qxp

7/13/2007

4:50 PM

Page 135

11. Manipulating embryo development Jacques Cohen

INTRODUCTION

The merging of different areas of biology, engineering, and physics can be expected to bring exciting developments to any field of research. This was evident when micromanipulation was introduced to experimental biology a century ago by Marshall Barber,1 and the approach has shown remarkable value in its application to clinical assisted reproductive technology (ART). Cellular surgery in ART had its beginnings in the early days of promoting fertilization by breaching the zona pellucida;2 this led on to offering real alternatives for male factor infertility by intracytoplasmic injection (ICSI),3 and then expanded to the use of assisted hatching and polar body and embryo biopsy for the purpose of preconception and preimplantation genetic diagnosis (PGD).4,5 Since the time that micromanipulation techniques were first cautiously applied in the treatment of infertile patients, hundreds of thousands of babies have been born worldwide as a result of these methods. Thousands of embryologists are now proficient in one or more micromanipulation techniques. The main emphasis in this chapter is to discuss the use of micromanipulation as a tool for improving embryo selection. Consequently, assisted hatching as well as some principles related to embryo biopsy will be evaluated. No review would be complete without referring to alternative opinions, and it should be acknowledged that some clinicians and scientists regard these procedures as controversial; the reader is referred to an opinion published in 2004 by Jim Cummins, which presents some of these technologies in a quite different perspective.6

ASSISTED HATCHING GENERAL CONSIDERATIONS

The premise behind assisted hatching is based on a hypothesis put forward nearly 20 years ago, suggesting that a modification of the human zona pellucida might promote hatching or implantation of embryos that are otherwise unable to escape intact from the zona pellucida.7 The modification could be carried out either by its elimination, by drilling a hole, by thinning, or by altering its stability. This argument is based on data from eggs obtained from follicular stimulation and in vitro observations involving IVF, and therefore none of the work suggests that there is a true disease-specific condition causing infertility due to impaired hatching from the zona pellucida. The hypothesis relates only to eggs that are fertilized in vitro. However, the technology has become a controversial conundrum, with only a minority of supporters. Fifteen years ago, it was demonstrated that the implantation potential of day 3 embryos undergoing initial compaction could be improved by creating relatively large openings in their zonae, (15–20 ␮m on the inside to 30–50 ␮m on the outside) by drilling with acidified Tyrode’s solution.7 Zona drilling increased the rate of implantation in patients whose embryos had thick zonae, a phenomenon which is known to reduce implantation, as well as in patients whose embryos developed slowly.8 The procedure was most beneficial in patients over the age of 38, and in those with elevated basal follicle stimulating hormone (FSH) levels.9 Selective assisted hatching has now been implemented in many IVF patients, with its application being dependent on individual embryonic variables,



HPE_Chapter11.qxp

7/13/2007

4:50 PM

Page 136


maternal age, patient’s previous history, and basal FSH levels. An evaluation of even the most recent literature shows that a great deal of confusion still exists about the usefulness of assisted hatching, and to which patient population it should be applied. There is no general consensus of agreement about the appropriate technique, or stage of development. In 2006 assisted hatching was reported to be of no confirmed benefit in good prognosis patients;10 however, this had already been acknowledged by the team who developed the original technique in 1992.8 Since that time there have been numerous papers arguing that there is no benefit (but no detriment either!) to the use of assisted hatching in young and first time patients. Due to the complexity of patient etiology and diversity in technique, metaanalyses11–15 have failed to demonstrate a gain from artificially opening the zonae of fresh embryos. This is hardly surprising, since few (if any) studies allowable for meta-analysis are comparable – many are unsuitable for analyses because of different selection and technical criteria. Conclusions vary from demonstrating significantly improved odds of increasing the clinical pregnancy rate, to no evidence of an increase in birth rates. One single independent meta-analysis study did show a demonstrable effect.15 Although the different hatching techniques seem comparable, minor technical differences could account for the contradictory results. Over the past 2 years alone, two prospective randomized studies were carried out on the use of laser guided assisted hatching in frozen–thawed embryos. Balaban et al.16 showed a clear improvement in pregnancy and implantation rates, whereas Ng et al.17 in an apparently similar study, showed no effect at all. Three other recent comparable non-randomized studies of frozen embryo assisted hatching also disagreed; one argued that assisted hatching had a positive effect,18 but this was not confirmed by two other groups.19,20 A critical analysis suggests that these studies were performed with substandard approaches to assisted hatching. In the first study19 a quarter of the zona pellucida was removed before embryo freezing. Zona opening before freezing at the cleavage stage is known to significantly decrease implantation after thawing.21 The other group20 used a proteolytic 

enzyme to thin the zona pellucida – a technique that is rarely applied and therefore not comparable to assisted hatching involving a full breach of the zona, since the toxicological consequences of enzyme breaching are unknown in experimental models. In contrast to the published literature for ICSI, only approximately 275 scientific papers have been published about assisted hatching; it is therefore assumed that the technique is not widely practiced. Whereas there is a general consensus about the efficacy of ICSI, any review of the literature demonstrates that this is clearly not the case with assisted hatching. A large proportion of studies have evaluated minor changes in the technical protocol without emphasis on clinical efficacy. Five metaanalyses on assisted hatching have been reported: four from the same group, published over a period of 4 years (three in the same journal), to some extent re-evaluating a small proportion of acceptable prospective randomized studies.22–26 The consensus of earlier reviews11–15 supports our conclusions from four randomized trials carried out in 1993:9 assisted hatching may improve outcomes in poor prognosis patients, particularly in the case of maternal aging. Two further meta-analyses11,12 conclude that there is no clinical benefit to assisted hatching. Proof of efficacy was only attainable when patients were selected based on maternal age or prior failed attempt. Conflicting opinions may be due to the problems inherent in randomizing unselected or partially selected populations. In our experience with over 10 000 patients, the effect of selective assisted hatching in certain low prognosis patients is not noticeable because we are unable to select against suitable control patients; with experience gained from randomized studies, differing selection criteria and a drift in technology may cause a deviation in overall results. In addition, applying selective assisted hatching to a much larger number of embryos and patients than was indicated by the initially accepted criteria may result in a dilution effect. Based on the supposition that a detrimental effect of assisted hatching has never been shown, practitioners may have loosened their criteria over time. In comparing controlled studies, there is no effective means of compensating for study design differences such as those caused by

HPE_Chapter11.qxp

7/13/2007

4:50 PM

Page 137

MANIPULATING EMBRYO DEVELOPMENT

age cut-off, opening, thinning, and subtle variation in methodologies and number of attempts. Hence, the comparisons become blurred, in the same manner that results from attempting to evaluate the parameters of a single large IVF database with specific groups represented in both experimental and control arms. In addition, embryo culture technology and follicular stimulation, which are both the most likely factors affecting embryonic health and associated zona changes, have improved spectacularly during the past 15 years. This has possibly reduced the need for assisted hatching in sub-groups of patients; however, the exact contribution of this change in technology on hatching behavior needs to be determined. Papers reviewing assisted hatching reveal that there is a lack of understanding regarding the nuances of techniques within the different methodologies used. There are only a certain number of appropriate ways of carrying out each technique, whether it involves the use of laser, acidified Tyrode’s solution, enzymes, mechanical opening, or some other derivative technology. However, no guidelines for this exist in the literature. Indeed, this is the case not only for assisted hatching, but for any applied medical technology. Most practitioners realize that there are ways of doing suboptimal IVF and good IVF, and likewise, there is good assisted hatching technique and suboptimal assisted hatching technique. Meta-analysis evaluations are not able to weigh these important aspects. Indeed nor can this reviewer provide a more exact evaluation of the problem, but nevertheless as the one who introduced assisted hatching, I would like to take this opportunity to share some general observations, without evaluating each of the assisted hatching papers one by one. Previous attempts to do so illustrate both the difficulty in determining the efficacy of assisted hatching, and the lack of consensus among practitioners.22,23 The most common hatching method used is zona drilling, opening the zona pellucida with acidified Tyrode’s solution.9 The zona reacts to a pH of about 2.3: at pH ⬍2.3 the zona will disappear as a coherent structure, whilst it remains intact at pH ⬎2.3. In this respect it is important to consider the buffer solution that the embryos are maintained in during the procedure. Buffers such as HEPES are excellent

in locally maintaining pH and have high buffering capacity. These solutions are more forgiving than bicarbonate buffered systems, which have reduced buffering capacity. Another consideration is that in being aware of the aggressive nature of acidified solution, technologists may not necessarily consider the physics of the system, and therefore often like to perform the procedure gently and carefully. Also, a hierarchy of teaching from developer to practitioner, as was the case with ICSI is largely absent; zona opening has unfortunately been interpreted as a very simple procedure, and the technique was applied directly by reading the original publications, without communicating with the experienced groups. While the first descriptions had some depth, it has become apparent over time that subtle aspects were not described in detail. Hence a number of interpretations of the procedure have been applied, and its perception as an easy technique has largely prevailed. One example is that many embryologists will not deposit the acidified solution directly on the zona pellucida by keeping the microneedle pressed on the zona. This may seem aggressive, but it has the specific purpose of limiting the release of acidified solution further. While the zona is dissolving, the microneedle should be moved into the thinning area. Releasing acidified solution from a microneedle that is kept at distance causes a broad stream of acidified medium that will not sufficiently lower the pH below 2.3, with the result that larger quantities of acidified solution are released because of perceived carefulness. This is likely to affect cells adjacent to the manipulated area. Similar considerations can be applied to the use of a laser; original clinical applications do not refer to possibly detrimental effects due to details of methodology. The same arguments can be applied to opening the zona pellucida for biopsy purposes. Less than half of all the embryos created after assisted reproduction appear genetically normal (see Chapter 18), but implantation rates are generally lower than that, indicating that a number of other factors must be involved. Assisted hatching promotes earlier implantation and may therefore elevate the chance of implantation by optimizing the implantation window.24 

HPE_Chapter11.qxp

7/13/2007

4:50 PM

Page 138


Assisted hatching can potentially be applied to any embryo, but in the absence of consistent results, many clinics have been slow to introduce it, or have abandoned its use. Although our own practice indicates that assisted hatching improves implantation, controversy continues to impede its widespread and dependable application. Inconsistent results are invariably due to technical variations and subtle changes to the originally published protocols; the initial learning curve with this technique must be emphasized, because individual and team results may improve considerably after experience. Complete removal of the zona pellucida prior to compaction may lead to loss of cells due to the absence of structural space. At the blastocyst stage, the proportion of human expanded blastocysts growing in vitro that will hatch may be related to the quality of the culture system, and the frequency of successful hatching in vitro is enhanced by zona opening in at least a proportion of in vitro studies. Common technical problems that we have identified include: (1) Lack of immunosuppression25 in patients whose embryos are zona drilled at the cleavage stage; (2) Excessive use of acidified Tyrode’s solution by keeping the micropipette more than 5 ␮m away from the zona pellucida, thereby lowering the pH over a greater area than is required for zona piercing as described above; (3) Use of hand-controlled suction devices for release of acidified solution – insufficient visualization of the fluid impedes its controlled release; (4) Creating holes smaller than 10 ␮m, which may trap the embryos during hatching; (5) Creating holes larger than 25 ␮m, which may lead to cell loss during or after embryo transfer; (6) Inappropriate trans-cervical embryo transfer method, so that excess zona pressure is created during replacement; (7) Inappropriate or inconsistent use of laser. At least five different assisted hatching methods have been used, with varying degrees of success: (1) Mechanical partial zona dissection on day 2, 3 or 5; 

(2) Zona drilling with acidified Tyrode’s solution on day 2 or 3; (3) Laser assisted drilling on day 3 using an infrared non-contact laser; (4) Zona drilling with acid or laser at the blastocyst stage.26 (5) Piezo-mediated drilling has also been used in animal and human models, and might in future be considered an alternative for clinical application.64 However, results in our laboratory have shown that excessive use of piezo devices is detrimental to embryo development in the mouse. In our experience, the first method of mechanical partial zona dissection tended to produce gaps which were too small, possibly resulting in cell separation during the escape from the zona pellucida; this approach was therefore abandoned.27 Mouse experiments and clinical studies from polar body biopsies suggest that creating a second perpendicular gap may be beneficial.28 Larger mechanical openings were also studied to some extent, and may be helpful.29 If artificial gaps are too small, the blastocyst may become trapped during hatching; recent findings following embryo biopsy demonstrated a link between gap size and monozygotic twinning, but the rate of monozygotic twinning was not increased when the incidence was compared with the frequency in the natural population.30 Gaps created for biopsy are usually 2 or 3 times larger than those created for assisted hatching. SELECTION FOR ASSISTED HATCHING

Our current policy is to apply assisted hatching to a large proportion of both IVF and ICSI patients, with the exception of those in whom traumatic transfers are anticipated, and those who do not consent to the procedure. A number of programs have simplified their selection criteria by using a patient age cut-off limit. Other selection criteria include all patients who have previously failed IVF or ICSI, or who are otherwise considered to have a poor prognosis for achieving a pregnancy. Guidelines for selecting individual embryos for assisted hatching depend upon a number of variables.

HPE_Chapter11.qxp

7/13/2007

4:50 PM

Page 139


Hypothetical cut-off levels above or below which the chance of implantation is considerably reduced are related to maternal age, elevated FSH, zona thickness, percentage fragmentation, and number of blastomeres. However, ultimately the selection process remains a clinical rather than a scientific decision. One guideline suggests that embryos from patients with elevated basal FSH levels should always be manipulated, regardless of other evaluations, but this argument is not supported by solid data. Embryos from patients 38 years or older may also benefit from zona drilling. A truly thick zona is defined as having a mean zona pellucida greater than 18 ␮m, but this number is dependent on the duration of follicular stimulation and maternal age as well as the instrument used for observation. Assisted hatching can also be used for embryos with a detrimental degree of fragmentation. Patients with previous failed treatment after apparently normal embryo transfer without zona drilling are also considered for assisted hatching. TECHNIQUES FOR ZONA OPENING: ACIDIFIED SOLUTION

Acidified Tyrode’s solution is used in embryo biopsy procedures, and therefore it is also widely used for zona drilling. The diameter of needles used for directing the acidified solution ranges from 10 to 12 ␮m. The microneedle is front-loaded with acidified Tyrode’s solution before each hatching event, precisely controlled with mouth suction. This precise control is especially important, because the meniscus of the acidic fluid is difficult to manipulate. Although heated stages may be used, the optimal temperature for this or any other type of embryo micromanipulation is not known. The embryo should be pre-aligned with the holding pipette only, so that there is an area directly subjacent to the region to which the acidified solution will be directed (usually the 3 o’clock position) that shows either an open area between blastomeres, an area of unusually large perivitelline space, or an area with a concentration of fragments. The likelihood of a small amount of acidified solution expelled into the perivitelline space coming into

immediate contact with the surface of a blastomere before being aspirated back into the hatching needle is therefore diminished. When the embryo is appropriately positioned with the use of the single tool, the micropipette filled with acidified solution is lowered into the medium and brought adjacent to the target area as quickly as possible. This is intended to avoid diluting the acidified solution with medium of normal pH: releasing fluid that is insufficiently acidic will jeopardize the embryo without affecting the zona pellucida. The key to successful assisted hatching is producing a gap in the zona whilst minimizing the impact of exposure to acidified solution on the embryo. Since the acidified Tyrode’s solution is aspirated through a needle of very small diameter, considerable residual suction (lower pressure) remains in the hatching needle even after it is removed from the reservoir drop of acid solution. When the hatching needle enters the droplet of medium containing the embryo to be hatched, this residual suction will aspirate culture medium, so that a column of neutral pH medium will therefore be at the tip of the hatching needle, diluting the upstream acidified Tyrode’s solution. This technical problem can be prevented by preparing the system in advance, so that the hatching needle will be in precisely the correct position relative to the embryo when it is lowered into the medium droplet. There should be a delay of no more than 2 seconds between the time that the hatching needle enters the drop until the fluid is expelled to initiate hatching. A slight positive pressure in the hatching needle, applied as soon as the needle breaks the surface of the drop when it is lowered, will also serve to counteract the residual suction. The procedure should be ceased if thinning is not immediately obvious. The acid solution should be expelled forcefully, so that the hole is made as quickly as possible, minimizing the time of acid exposure. The total time required to breach the zona should not exceed a couple of seconds; most zonae will yield in fewer than 5 seconds. The needle should be applied directly to the zona pellucida, massaging the area to be opened while the hole is being made, with the narrowest point created at the interior surface of the zona. The massaging motion should create a hole 

HPE_Chapter11.qxp

7/13/2007

4:50 PM

Page 140


that is nearly rectangular. The inside layer of the zona pellucida is frequently more resistant to reduced pH than the outer layers, and care should therefore be taken to ensure that zona breakthrough occurs over a sufficiently wide area of at least 20 ␮m, and not at a single small point. The flow through the assisted hatching needle must be immediately reversed as soon as the zona is breached. This is most effectively achieved via the use of a mouth controlled suction tube, and for this reason mouth pipetting may be preferable for assisted hatching. All of the expelled acidified solution must be aspirated, paying particular attention to any solution that may have entered the perivitelline space. The embryo must be simultaneously moved to another area of the droplet, away from the area of reduced pH.

zona ablation with no obvious detrimental effects reported.33,34 Two recent randomized studies show no real consensus;16,35 one study suggests that laser assisted thinning is of no benefit in patients older than 36 years, whereas the other study shows a significant improvement in patients whose embryos have been thawed. Earlier studies did show a possible benefit in terms of implantation and clinical pregnancy rate, and a relatively large group of healthy babies have now been born following laser-mediated assisted hatching. However, basic safety studies are rare, and those that exist32 mainly do not involve clinical applications. Whether these tools are safe and can improve pregnancy rates is yet to be determined.

ASSISTED HATCHING WITH LASER

Small fragments or lysed cells can be aspirated with the hatching needle after the gap in the zona pellucida has been created. Using a mouse model, in 1993 Alikani et al were the first to explore the possibility of removing degenerate blastomeres from embryos before compaction.36 They also described that clinical pregnancies could be established after removing lysed cells from human embryos, but the efficacy of this procedure was not demonstrated. Data sets of embryos transferred with lysed cells remaining inside their zonae were recently compared with embryos the lysed cells of which had been removed, and this comparison showed a remarkable improvement in implantation after fragment removal.37 The clinical benefit of this procedure is still debated,36,38 but promising results have now been reported after removing lysed cells from thawed embryos before transfer.37,39 The technique requires a high level of skill in order to remove all or most of the fragments and lysed material from an embryo via a single hole without causing damage. Great caution must be exercised, since even the slightest touch of the hatching needle on the membrane of a blastomere can result in the loss of membrane integrity. Since 1993, we have removed fragments and lysed cells from the embryos of more than 3000 patients. A 12 ␮m diameter needle is preferred for fragment removal, compared with the approximate

Non-contact infrared (IR) laser has emerged as the methodology perhaps best suited to mammalian zona cutting applications.31 Several FDA approved commercial systems are now available that use IR diode lasers, and these have been used in human clinical embryology procedures such as assisted hatching and biopsy. However, some of the basic models that have been used to investigate the infrared delivery system are questionable, and localized effects such as heat have been assessed only in other laser systems (for review see Malter et al).32 Early clinical studies have been generally hampered by lack of appropriate controls. FDA studies have been conducted, but the study designs have in some cases been suboptimal, as comparisons were made between laser manipulated embryos and nonmanipulated control embryos.33 This confounds the effect of zona opening per se with any effects of the laser. A more appropriate study would be to compare the laser technique with other methods for mechanical or chemical zona opening. Recent studies using laser applications have focused only on the efficacy of the method, with safety evaluated in terms of pregnancy rates. The results reported appear promising, and apparently demonstrate simple, repeatable, and appropriate 

REMOVAL OF FRAGMENTS AND LYSED CELLS AFTER CRYOPRESERVATION

HPE_Chapter11.qxp

7/13/2007

4:50 PM

Page 141


10 ␮m of a regular assisted hatching needle. If the membrane of a blastomere appears to be reacting in any way during the procedure, suction must be instantly ceased. Some fragments and lysed debris may be firmly attached to blastomeres, and removal may be counter-productive in such cases. The procedure can be time-consuming, and should be done gently and patiently. Since fragments are not all in the same focal plane as the pipette tip and artificial opening, the microscope must be continually adjusted in order to focus on target material and adjacent blastomeres. The zona should be turned after removing some of the fragments, as a different angle of approach increases the likelihood that more fragments can be removed. Fragments in between blastomeres and those opposite the aperture should be removed last. Removing fragments from areas between cells is especially advantageous, because this apparently improves cell–cell interaction during compaction.38 Fragment and debris removal should only be practiced at the highest magnification, using a stateof-the-art interference microscope. An awareness of the distinct patterns of fragmentation is essential, along with continuous re-focusing and moving the zona around while determining the best approach. Most of the benefits of assisted hatching and fragment removal lie in creating the artificial gap in the zona pellucida. The benefit derived from fragment removal over and above that of assisted hatching alone is real, but modest in comparison. It is probably unwise to spend a great deal of time attempting to remove only a few fragments on the periphery of an embryo, whereas it is more appropriate to take time in removing larger numbers of fragments, especially if they are present in an orientation that may interfere with normal cell–cell contact and impede compaction. MONOZYGOTIC TWINNING AND ASSISTED HATCHING

The incidence of monozygtic twinning is quadrupled after follicular stimulation with or without assisted reproduction,30 and this has not been described in relation to all aspects of ART. Follicular stimulation,

poor embryo development, assisted hatching, and blastocyst development have all been described as risk factors, but evidence of direct associations is largely missing. In the case of blastocyst transfer, the association may be related to changes in the zona pellucida or the ability of the embryo to prepare for hatching, but this may be due to subtle aspects of embryo culture and could be remedied with improved culture conditions. The incidence of identical twinning after blastocyst transfer may also be reduced by removing the zona mechanically or enzymatically, or by performing a partial yet vigorous opening procedure. The association between assisted hatching and monozygotic twinning is complicated by confounding factors such as selection of patients and embryos. Again, the association between monozygotic twinning and micromanipulation may be due to technical variations – a more constraining zona opening is likely to lead to trapped or split embryos.38 This is illustrated by the observation that only 1/140 (0.7%) embryos that were biopsied developed into monozygotic twins.30 Adapting the procedure technically so that a larger zona opening is created might therefore reduce nearly all the factors that are associated with trapping and splitting.

EMBRYO BIOPSY

PGD can now be performed at nearly any stage between egg maturation and cavitation of the blastocyst. The validity of PGD and its scope of application in relation to embryo selection is a complex topic that requires broad discussion, and is described elsewhere in this book by Munné and by Kuliev and Verlinsky. The general interest in the biopsy technique itself and in understanding aspects of the culture system that may play a role during micromanipulation for PGD is rather limited. Few studies have emphasized these aspects, in spite of a genuine and broad interest in the genetic aspects of PGD. Biopsy of the cleaved embryo on day 3 is the method preferred by most clinics that are interested in PGD, and this is discussed briefly here. The zona can be opened for embryo biopsy at the cleavage stage in three different ways, and at least two methods 

HPE_Chapter11.qxp

7/13/2007

4:50 PM

Page 142


can be used to extract a blastomere through the artificially created gap.40–44 The zona can be opened mechanically (but this is rarely applied) via the use of acidified solution (see Assisted hatching section above) or with the use of a laser. The targeted blastomere can be extracted using a biopsy needle that absorbs the cell partially or completely and releases it elsewhere in the droplet, or it can be freed from the embryo by applying pressure to the zona pellucida with a thin tool that either enters, or does not enter the zona pellucida. The latter is a single tool technique, and is often used in combination with a laser. Surprisingly, there are no randomized studies that compare any of these techniques, although serial observations have been described.44 The circumstances of biopsy have not been studied in great detail. Despite its widespread use, there is no information available from the literature regarding the optimal use of Ca- and Mg-free medium and any safety aspects. There is a consensus of agreement about its technical advantages, since blastomere survival is clearly enhanced. Other aspects of the technique which lack information include the use of buffered culture systems, optimal temperature, and appropriate guidelines for targeting a certain blastomere, although it is agreed that the blastomere should not be dividing and should have a single clear nucleus. Whether polarity exists that predisposes blastomeres to the inner cell mass and trophoblast at the 8 cell stage is uncertain. Cryopreservation studies involving models of cell survival indicate that early allocation is either absent or has little clinical effect after biopsy.45,46 The most important question regarding embryo biopsy is the possible effect on further development and outcome if an associated genetic diagnosis was not performed. In other words, does a single blastomere biopsy affect the embryo and is this loss compensated in one way or another by the subsequent genetic diagnosis? Certainly, the notion that PGD would be successful and result in acceptably high pregnancy rates after testing for single gene disorders was assumed many years ago, but results in clinical practice have been disappointing. Although the effects of cell loss after biopsy can be tested by performing prospective analysis of intact and 

biopsied embryos without a genetic diagnosis, such experiments present an ethical challenge. The proposition that blastocysts obtained after single cell biopsy have cell numbers proportional to their initial cell count is often quoted as showing that the embryo can lose a cell without detriment, but those experiments were performed with embryos showing good development obtained from young donors.47 IVF specialists often defer that the number of cells on day 3 is not 8 cells on average, but a lower number that is dependent on patient selection and other factors. A discussion about this and the consequences of losing one instead of two cells from embryos has been discussed in the chapter by Munné in the context of PGD efficiency.

OTHER CONSIDERATIONS

Reproductive medicine is singled out in most countries as the only form of clinical practice that requires legislation. Many apparently secular governments appear to introduce contemporary limitations that are sanctioned by organized religion. The field of reproductive cell surgery apparently causes great concern among lawmakers and ethicists alike, especially since it involves the tools that enable scientists to perform nuclear transplantation. However, there are advantages to be gained from the use of similar technologies. Transferring a nucleus into an oocyte, such as is the case with ICSI, is cloning derived technology, as are the attempts by reproductive specialists to transfer haploid sperm precursor cells. Removing cells from embryos during PGD (a technology now prohibited in several European countries) is yet another application where experience from cloning technology has been advantageous. The opportunity to treat infertility using micromanipulation for embryo selection should not be confused with or mistaken for cloning or alteration of the genome in an as yet undefined and possibly eugenic direction (genetic engineering). Nothing could be further from the truth. These procedures aim to enhance development and reduce multiple pregnancy, consequently improving conditions for the developing fetus and the mother.

HPE_Chapter11.qxp

7/13/2007

4:50 PM

Page 143


Just as the success of advanced surgery depends upon the skill of the surgeon, successful duplication of embryology procedures depends in great measure on mastering the skills of micromanipulation, and on the experience and expertise of the individual embryologist. This appears to be especially the case with procedures such as assisted hatching, fragment removal, and embryo biopsy. The field of micromanipulation in embryology is just as exciting as that of surgery, perhaps even more so – and may ultimately lead to more widespread use of some of the techniques described here. It is naturally to be expected that new, and perhaps surprising, applications will be discovered that may extend beyond infertility. This timeline is not controlled by patients or their doctors, and it is not dependent on scientists or their lack of funding. It is almost entirely constrained by the attitude of powerful religious extremists.

9.

10.

11.

12.

13.

14.

15. 16.

ACKNOWLEDGMENTS

I am grateful to Santiago Munné, John Garrisi, Mina Alikani, Tim Schimmel, Steen Willadsen, and Henry Malter for countless contributions. I am thankful for editorial comments by Rashmi Dalai. REFERENCES 1. Korzh V, Strahle U. Marshall Barber and the century of microinjection: from cloning of bacteria to cloning of everything. Differentiation 2002; 70: 221–6. 2. Cohen J, Malter H, Fehilly C et al. Implantation of embryos after partial opening of oocyte zona pellucida to facilitate sperm penetration. Lancet 1988; 16: 162. 3. Palermo G, Joris H, Devroey P, Van Steirteghem A. Pregnancies after intracytoplasmic sperm injection of single spermatozoon into an oocyte. Lancet 1992; 340: 17–8. 4. Handyside AH, Kontogianni EH, Hardy K, Winston RML. Pregnancies from biopsied human preimplantation embryos sex by Y-specific DNA amplification, Nature 1990; 344: 768–70. 5. Verlinsky Y, Ginsberg N, Lifchez A et al. Analysis of the first polar body: preconception genetic diagnosis. Hum Reprod 1990; 5: 826–9. 6. Cummins JM. Can and should human embryos be rescued from developmental demise? Methods and biological basis. In: Van Blerkom J, Gregory L, eds. Essential IVF: Basic Research and Clinical Applications, Norwell, Massachusetts: Kluwer Academic Publishers, 2004: 555–75. 7. Cohen J, Elsner C, Kort H et al. Impairment of the hatching process following IVF in the human and improvement of implantation by assisted hatching using micromanipulation. Hum Reprod 1990; 5: 7–13. 8. Cohen J, Inge KL, Suzman M, Wiker SR, Wright G. Videocinematography of fresh and cryopreserved embryos: a retrospective

17.

18.

19.

20.

21. 22.

23.

24. 25.

26.

27.

analysis of embryonic morphology and implantation. Fertil Steril 1989; 51: 820–7. Cohen J, Alikani M, Trowbridge J, Rosenwaks Z. Implantation enhancement by selective assisted hatching using zona drilling of human embryos with poor prognosis. Hum Reprod 1992; 7: 685–91. Sagoskin AW, Levy MJ, Tucker MJ, Richter KS, Widra EA. Laser assisted hatching in good prognosis patients undergoing in vitro fertilization-embryo transfer: a randomized controlled trial. Fertil Steril 2007; 87: 283–7. Seif MM, Edi-Osagie EC, Farquhar C, Hooper L, Blake D, McGinlay P. Assisted hatching on assisted conception (IVF & ICSI). Cochrane Database Syst Rev. 2006; (1): CD001894. Review. Seif MM, Edi-Osagie EC, Farquhar C, Hooper L, Blake D, McGinlay P. Assisted hatching on assisted conception (IVF & ICSI). Cochrane Database Syst Rev. 2005; (1): CD001894. Review. Update in Cochrane Database Syst Rev. 2006; (1): CD001894. Edi-Osagie EC, Hooper L, McGinlay P, Seif MW. Effect(s) of assisted hatching on assisted conception (IVF & ICSI). Cochrane Database Syst Rev. 2003; (4): CD001894. Review. Update in Cochrane Database Syst Rev. 2005; (4): CD001894. Edi-Osagie E, Hooper L, Seif MW. The impact of assisted hatching on live birth rates and outcomes of assisted conception: a systematic review. Hum Reprod 2003; 18: 1828–35. Sallam HN, Sadek SS, Agameya AF. Assisted hatching – a meta-analysis of randomized controlled trials. J Assist Reprod Genet. 2003; 20: 332–42. Balaban B, Urman B, Yakin K et al. Laser-assisted hatching increases pregnancy and implantation rates in cryopreserved embryos that were allowed to cleave in vitro after thawing: a prospective randomized study. Hum Reprod 2006; 21: 2136–40. Ng EH, Naveed F, Lau EY, Yeung WS et al. A randomized double-blind controlled study of the efficacy of laser-assisted hatching on implantation and pregnancy rates of frozen-thawed embryo transfer at the cleavage stage. Hum Reprod 2005; 20: 979–85. Gabrielsen A, Fedder J, Agerholm I. Parameters predicting the implantation rate of thawed IVF/ICSI embryos: a retrospective study. Reprod Biomed Online 2006; 12: 70–6. Kung FT, Lin YC, Tseng YJ et al. Transfer of frozen-thawed blastocysts that underwent quarter laser-assisted hatching at the day 3 cleaving stage before freezing. Fertil Steril 2003; 79: 893–9. Sifer C, Sellami A, Poncelet C et al. A prospective randomized study to assess the benefit of partial zona pellucida digestion before frozenthawed embryo transfers. Hum Reprod 2006; 21: 2384–9. Magli MC, Gianaroli L, Grieco N et al. Cryopreservation of biopsied embryos at the blastocyst stage. Hum Reprod 2006; 21: 2656–60. Wright G and Jones A. Assisted hatching in clinical IVF. In: Van Blerkom J and Gregory L, eds. Essential IVF: Basic Research and Clinical Applications, Norwell, Massachusetts: Kluwer Academic Publishers, 2004: 441. Practice Committee of the American Society for Reproductive Medicine. The role of assisted hatching in in vitro fertilization: a review of the literature. Fertil Steril 2004; 82(Suppl 1): S164. Liu H-C, Cohen J, Alikani M, et al. Assisted hatching facilitates earlier implantation. Fertil Steril 1993; 60: 871–5. Cohen J, Malter H, Elsner C, Kort H, Massey J, and Mayer MP. Immunosuppression supports implantation of zona pellucida dissected human embryos, Fertil Steril 1990; 53: 662–5. Vanderzwalmen P, Bertin G, Debauche CH et al. Vitrification of human blastocysts with the Hemi-Straw carrier: application of assisted hatching after thawing. Hum Reprod 2003; 18: 1504–11. Cohen J, Feldberg D. Effects of the size and number of zona pellucida openings on hatching and trophoblast outgrowth in the mouse embryo. Mol Reprod Dev 1991; 30: 70–8.



HPE_Chapter11.qxp

7/13/2007

4:50 PM

Page 144


28. Cieslak J, Ivakhnenko V, Wolf G et al. Three-dimensional partial zona dissection for preimplantation genetic diagnosis and assisted hatching. Fertil Steril 1999; 71: 308–13. 29. Lyu QF, Wu LQ, Li YP et al. An improved mechanical technique for assisted hatching. Hum Reprod 2005; 20: 1619–23. 30. Alikani M, Cekleniak NA, Walters E et al. Monozygotic twinning following assisted conception: an analysis of 81 consecutive cases. Hum Reprod 2003; 18: 1937–43. 31. Germond M, Nocera D, Senn A et al. Microdissection of mouse and human zona pellucida using a 1.48-microns diode laser beam: efficacy and safety of the procedure. Fertil Steril 1995; 64: 604–11. 32. Malter HE, Schimmel T, Cohen J. Zona dissection by infrared laser: developmental consequences in the mouse, technical considerations, and controlled clinical trial. Reprod Biomed Online 2001; 3: 117–23. 33. Primi MP, Senn A, Montag M et al. A European multicentre prospective randomized study to assess the use of assisted hatching with a diode laser and the benefit of an immunosuppressive/antibiotic treatment in different patient populations. Hum Reprod 2004; 19: 2325–33. 34. Ebner T, Moser M, Tews G. Possible applications of a non-contact 1.48 micron wavelength diode laser in assisted reproduction technologies. Hum Reprod Update. 2005; 11: 425–35. 35. Frydman N, Madoux S, Hesters L et al. A randomized double-blind controlled study on the efficacy of laser zona pellucida thinning on live birth rates in cases of advanced female age. Hum Reprod 2006; 21: 2131–5. 36. Alikani M, Olivennes F, Cohen J. Microsurgical correction of partially degenerate mouse embryos promotes hatching and restores their viability. Hum Reprod 1993; 8: 1723–8. 37. Nagy ZP, Taylor T, Elliott T, et al. Removal of lysed blastomeres from frozen-thawed embryos improves implantation and pregnancy rates in frozen embryo transfer cycles. Fertil Steril 2005; 84: 1606–12. 38. Alikani M, Sadowy S, Cohen J. Human embryo morphology and developmental capacity. In: A Van Soom and M Boerjan, eds. Assessment of Mammalian Embryo Quality, Invasive and non-Invasive Techniques. Kluwer Academic Publishers: The Netherlands, 2002: 1–24. 39. Rienzi L, Nagy ZP, Ubaldi F et al. Laser-assisted removal of necrotic blastomeres from cryopreserved embryos that were partially damaged. Fertil Steril 2002; 77: 1196–201.



40. Wilton LJ, Trounson AO. Biopsy of preimplantation mouse embryos: development of micromanipulated embryos and proliferation of single blastomeres in vitro. Biol Reprod 1989; 40: 145–52. 41. Grifo JA, Boyle A, Fischer E et al. Pre-embryo biopsy and analysis of blastomeres by in situ hybridization. Am J Obstet Gynecol 1990; 163: 2013–9. 42. Grifo JA, Tang YX, Cohen J et al. Pregnancy after embryo biopsy and co-amplification of DNA from X and Y chromosomes. JAMA. 1992; 12(268): 727–9. 43. Inzunza J, Iwarsson E, Fridstrom M. et al. Application of single–needle blastomere biopsy in human preimplantation genetic diagnosis. Prenat Diagn 1998; 18: 1381–88. 44. Joris H, De Vos A, Janssens R et al. Comparison of the results of human embryo biopsy and outcome of PGD after zona drilling using acid Tyrode medium or a laser. Hum Reprod 2003; 18: 1896–902. 45. Edgar DH, Bourne H, Jericho H, McBain JC. The developmental potential of cryopreserved human embryos. Mol Cell Endocrinol 2000; 27, 169: 69–72. 46. Edgar DH, Bourne H, Speirs AL, McBain JC. A quantitative analysis of the impact of cryopreservation on the implantation potential of human early cleavage stage embryos. Hum Reprod 2000; 15: 175–9. 47. Hardy K, Martin KL, Leese HJ et al. Human preimplantation development in vitro is not adversely affected by biopsy at the 8-cell stage. Hum Reprod 1990; 5: 708–14. 48. Cohen J, Wells D, Munne S. Removal of 2 cells from cleavage stage embryos is likely to reduce the efficacy of chromosomal tests that are used to enhance implantation rates. Fertil Steril 2006; 87: 496–503. 49. Munne S, Sandalinas M, Escudero T et al. Improved implantation after preimplantation genetic diagnosis of aneuploidy. Reprod Biomed Online 2003; 7: 91–7. 50. Munne S, Chen S, Fischer J et al. Preimplantation genetic diagnosis reduces pregnancy loss in women aged 35 years and older with a history of recurrent miscarriages. Fertil Steril 2005; 84: 331–5. 51. Staessen C, Platteau P, Van Assche E et al. Comparison of blastocyst transfer with or without preimplantation genetic diagnosis for aneuploidy screening in couples with advanced maternal age: a prospective randomized controlled trial. Hum Reprod 2004; 19: 2849–58. 52. Platteau P, Staessen C, Michiels A et al. Preimplantation genetic diagnosis for aneuploidy screening in patients with unexplained recurrent miscarriages. Fertil Steril 2005; 83: 393–7.

HPE_Chapter12.qxp

7/13/2007

4:51 PM

Page 145

12. Assessment of soluble human leukocyte antigen G in human embryos Jeffrey D Fisch, Levent Keskintepe, and Geoffrey Sher

INTRODUCTION

Implantation is one of the least well understood aspects of assisted reproduction. Over the past 20 years embryologists have improved culture techniques and physicians have refined stimulation protocols and optimized transfer techniques. Still, the fate of embryos after transfer remains largely a matter of speculation and hope. ‘Implantation’ is actually a complex, choreographed dialog between the embryo and the waiting endometrium. Prostaglandins, cytokines, integrins, and growth factors are all reported to play a role. The suppression of the maternal immune system is critical to successful implantation. The body discriminates self from non-self by the human leukocyte antigen (HLA) on the major histocompatability complex (MHC), encoded by a series of grouped genes on chromosome 6.1 The MHC elicits T cell mediated immune responses to allogenic (nonself antigens expressed by another individual of the same species) MHC antigens. This phenomenon, first recognized in tissue transplantation experiments in mice, may also play a role in the natural state, since the mammalian fetus is a semi-allograft in the maternal uterus due to paternal MHC antigens. During normal pregnancy, the maternal immune system undergoes changes that lead to fetal tolerance. A growing body of literature suggests that HLA-G, produced predominantly by extravillous cytotrophoblasts (which are the only fetal cells in direct contact with the maternal decidual cells) confers immunotolerance through interaction with maternal decidual lymphocytes.2 Immunolocalization studies of trophoblast invasion indicate that invading trophoblasts express HLA-G in a highly regulated process. The effects of growth factors and

cytokines on trophoblast invasion suggest that molecules of uterine origin can modify the process of implantation.3 HLA-G is expressed on placental extravillous cytotrophoblast cells in direct contact with maternal tissues. Circumstantial evidence suggests HLA-G protects the fetus from the maternal immune response and appears to be the functional analog of mouse Qa-2 protein.4 Using reverse transcriptase polymerase chain reaction (RT-PCR) on surplus embryos and unfertilized oocytes from IVF patients, HLA-G heavy chain mRNA was detected in 40% of 148 blastocysts tested. HLA-G was also detected in unfertilized oocytes and early embryos, but not in control cumulus oophorus cells. HLA-G levels in the preimplantation period correlated with the cleavage rate of embryos. In contrast to classical HLA-A and -B class Ia genes that are down regulated in human trophoblast cells, HLA-G is expressed by the placenta throughout gestation. In addition to extravillous cytotrophoblast, HLA-G expression was also observed in endothelial cells in the chorionic villi, as well as in amnion cells and amniotic fluid.5 Both membranebound and soluble HLA-G isoforms were identified, raising the possibility that membrane-bound HLA-G is involved in regulation of chorionic villi angiogenesis and that soluble HLA-G (sHLA-G) isoforms may act as specific immunosuppressors during pregnancy5 (Figure 12.1). This chapter reviews the role of sHLA-G in implantation and how assessment of sHLA-G secretion from preimplantation embryos can influence the outcome from assisted reproduction. We first briefly review the identification and characterization of HLA-G, examine the role of HLA-G on implantation, and, finally, evaluate the effect of 

HPE_Chapter12.qxp

7/13/2007

4:51 PM

Page 146


HLA class Ia

L

α1

α2

α3

Tm

C

3⬘UT

HLA-G1

L

α1

α2

α3

Tm

C

3⬘UT

HLA-G2

L

α1

α3

HLA-G3

L

α1

HLA-G4

L

α1

α2

sHLA-G1

L

α1

α2

sHLA-G2

L

α1

α3

Tm

C

Tm

C

3⬘UT

3⬘UT Tm

C

3⬘UT

α3 int. 4

int. 4 Tm

C

Tm

C

3⬘UT

3⬘UT

Figure 12.1 Schematic diagram of HLA class Ia and major HLA-G transcriptional isoforms. HLA-G1, -G2, -G3, and -G4 isoforms encode potentially membrane-bound proteins, whereas sHLA-G1 and sHLA-G2 encode soluble forms. L, exon 1 encoding the leader sequence; ␣1, ␣2, ␣3, exons 2, 3, and 4 encoding the ␣1, ␣2, ␣3 external domains, respectively; Tm, exon 5 encoding the transmembrane domain; C, exons 6 and 7 encoding the cytoplasmic region; 3⬘UT: exon 8 encoding the 3⬘ untranscribed region. The hexagon indicates the presence of ‘stop codons.’ Modified from Le Bouteiller et al.5

sHLA-G measurement on assisted reproductive technology (ART) outcomes.

CHARACTERIZATION OF HUMAN LEUKOCYTE ANTIGEN G

HLA-G was first cloned in 1987,6 but not until 1990 was the atypical HLA class I molecule present in abundant amounts at the maternal–fetal interface determined to be HLA-G.7 Alternative splicing of the HLA-G mRNA yields different membranebound and soluble variants of the HLA-G protein8 and a limited number of variable sites in the DNA sequence of the HLA-G gene have been reported.9,10 Because HLA-G has a low level of nucleotide sequence polymorphisms, it has been postulated as a prerequisite for maintenance of maternal immune tolerance. The sHLA-G molecule is strongly expressed during the first trimester of gestation and then decreases through the remainder of pregnancy, 

suggesting a potential role in implantation and immunoprotection of the developing embryo. An association has been reported between HLA-G expression in human embryos and other factors known to influence IVF outcome. Using non-transferred embryos at the 2–4-cell stage, Jurisicova et al11 found sibling embryos from successful cycles were more likely to express HLA-G than were those from non-successful cycles. Cytotoxicity assays determined that the primary HLA-G gene transcript is alternatively spliced into five main mRNA forms: HLA-G1 (full length), HLAG2 (minus exon 3) which codes for a membranebound isoform associated with B2 microglobulin, HLA-G3 (minus exons 3 and 4), HLA-G4 (minus exon 4), and HLA-G5 (plus intron 4) which codes for a soluble form of the HLA-G antigen. All HLA-G isoforms are capable of inhibiting natural killer (NK) cell activity. These findings suggest that HLA-G is the public ligand for NK cell inhibitory receptors.

HPE_Chapter12.qxp

7/13/2007

4:51 PM

Page 147

ASSESSMENT OF SOLUBLE HUMAN LEUKOCYTE ANTIGEN G IN HUMAN EMBRYOS

Immunohistochemical expression of classical and non-classical HLA class I proteins in human placentas using monoclonal antibodies demonstrated classical HLA class I proteins in all nontrophoblast cells, including fetal and maternal cells.13 Syncytiotrophoblast did not show any HLA class I staining, but extravillous cells expressed high amounts of HLA-G. The HLA-G gene message is alternatively spliced into multiple transcripts, three of which encode soluble isoforms. Morales et al14 produced soluble recombinant (r)HLA-G1, rHLA-G2 in human embryonic kidney 293 cells and characterized the proteins. Both isoforms were glycosylated and formed disulfide-bonded oligomers. The rG1 antibody associated with B2 microglobulin, while rG2 did not. Mouse monoclonal antibodies generated to rsG1 (specific to sG1) and to rsG2 (identifies both soluble and membrane-bound G2) were used for immunohistochemical isoform mapping studies on placental tissue sections. Soluble G1 was abundant in many subpopulations of trophoblast cells, while membrane (m)/sG2 protein was present exclusively in extravillous cytotrophoblast cells. Immunostaining localized m/sG2 only to the invasive trophoblast subpopulation. Analysis of function by Northern and Western blotting demonstrated both rsG1 and rsG2 inhibit CD8 alpha expression on peripheral blood mononuclear cells (PBMC) without changing CD3 delta expression or causing apoptotic cell death. Collectively, these studies indicate that both sG1 and sG2 are produced in placentae and that transcription and translation are linked for sG1, but not for sG2. While expression of sG2 is exclusively associated with the invasive phenotype, the two isoforms of sG may promote semiallogenic pregnancy by reducing expression of CD8, a molecule required for functional activation of cytotoxic T lymphocytes (CTL). The HLA-G molecule has unique properties of low polymorphism and restricted tissue distribution mainly to the extravillous cytotrophoblast (EVT). These EVT cells vigorously penetrate into maternal decidual tissues and are found in contact with maternal lymphocytes, mainly with NK cells. The HLA-G molecule inhibits effector function of maternal

NK cells via interaction with KIR2DL4 and ILT-2 inhibitory NK receptors. Efficient binding and function of ILT-2 is dependent on the presence of HLA-G complexes on the cell surface. Their presence on the cell surface suggests these complexes may also be present in a soluble form.15 This may be a particular mechanism to increase the avidity of NK receptors to HLA-G, resulting in better protection of the fetus from maternal NK cell attack (Figure 12.2). Soluble class Ib HLA-G glycoproteins synthesized in the placenta are abundant in the pregnant uterus and circulate in the maternal blood throughout pregnancy. Hunt et al17 established the immunogenicity of these proteins by testing sera of 64 women with at least one successful pregnancy, 21 women who had never been pregnant, and 54 males for antibodies to epitopes present on recombinant sHLA-G isoforms (G1/G2) derived from HLA 6.0 cDNA (HLA-G*0101 allele). Indirect enzyme linked immunosorbent assay (ELISA) identified antibodies to sHLA-G isoforms in six sera, all from multigravid women, while all other sera were negative. These results indicate that tolerance to HLA-G is the usual condition, as antibodies to HLA-G were not detected in 91% (58/64) of multigravid women. Pregnancy stimulated the loss of tolerance in only 9% (6/64). However, all six women delivered healthy babies, demonstrating that antibodies to HLA-G do not prevent successful pregnancy outcome. Both membrane-bound and soluble HLA-G isoforms are proposed to influence the outcome of pregnancy. Aberrant HLA-G expression is reported in pre-eclampsia and spontaneous abortion. Placental hypoxia following the immature remodeling of spiral arteries by extravillous cytotrophoblasts (CTs) has been investigated as a pathogenic mechanism for the development of pre-eclampsia.18,19 HLA-G expression is also decreased at the protein and mRNA levels in pre-eclamptic placenta.19

HUMAN LEUKOCYTE ANTIGEN G AND IMPLANTATION

Pregnancy is an immune balancing act between maternal tolerance of paternal major histocompatabilty 

HPE_Chapter12.qxp

7/13/2007

4:51 PM

Page 148


α1 α2 β2m α3

α1 α1 α3 α3

sHLA-G1 sHLA-G2 α1 α2 β2m α3

HLA-G1

Golgi

ER Cytosol

α1 α2

α1 α1

β2m α3

α3 α3

α1 α1

-G2

-G3

-G1

α1 α2

-G4

Peptide Figure 12.2 Schematic of sHLA-G secretion from cytotrophoblast cells. Only the full length HLA-G1 isoform is expressed at the cell surface whereas the HLA-G2, -G3, and -G4 forms are sequestered in the endoplasmic reticulum (ER). Both soluble HLA-G1 and -G2 isoforms can be secreted. Adapted from Le Bouteiller et al.16

antigens and maintaining normal immune competence. The placenta separates maternal and fetal blood/lymphatic systems. Fetal trophoblast plays a major role in evading recognition by the maternal immune system. Trophoblast cells fail to express MHC class I or II molecules and extravillous cytotrophoblasts strongly express non-classical MHC genes encoding HLA-G, which appear to down regulate NK cell activity. Trophoblast expresses Fas ligand, conferring immune privilege: maternal immune cells expressing Fas will undergo apoptosis at the placental–decidual interface20 (Figure 12.3). Natural killer cells in the decidua may control trophoblast migration during implantation through recognition of HLA-G/HLA-C by killer inhibitor/ activator receptors (KIR/KAR). Monoclonal antibodies demonstrated HLA-G in extravillous trophoblast expressed in both B2 microglobulin associated form and as free heavy chains.22 Lack of classical HLA molecules in the human placenta 

prevents recognition and lysis by maternal T lymphocytes, but leaves the conceptus susceptible to NK cell-mediated lysis. NK cells are known to express a number of HLA-class I-specific inhibitory receptors, including members of the Ig superfamily (p58, p70, p140), characterized by a defined allele specificity and CD94/NKG2A with broad specificity for HLA class I molecules. Analysis of NK cell clones expressing CD94/NKG2A as identified by Z199 monoclonal antibodies, displayed markedly reduced cytolytic activity against 221/G. Monoclonal antibodies against CD94/NKG2A completely restored target cell lysis.23 Receptors mediating NK cell inhibition were evaluated using monoclonal antibodies against known NK inhibitory receptors: CD158a, CD158b, and CD94.24 In at least one-third of NK cells inhibited by HLA-G, these antibodies alone or in combination failed to reverse inhibition, suggesting the presence of a third major unidentified receptor for HLA-G.

HPE_Chapter12.qxp

7/13/2007

4:51 PM

Page 149


Maternal blood/decidua Fas

Activated T cell CD8+

Fas-L

Apoptosis

CD8

Lysis α1 α2

sHLA-G1

β2m α3

Villous syncytiotrophoblast Extravillous cytotrophoblast

Figure 12.3 Role of soluble HLA-G1 in induction of specific apoptosis of activated CD8⫹ T cells through CD8 interaction. Such function is Fas/Fas ligand mediated. In vivo, soluble HLA-G1 may be produced by syncytiotrophoblast and, possibly, by extravillous cytotrophoblast, inducing apoptosis of activated CD8⫹ T cells present in maternal blood of the intervillous space and/or in the decidua. Modified from Solier et al.21

Class I negative syncytiotrophoblast escapes NK lysis by maternal peripheral blood leukocytes (PBL), while HLA-G expressing transfectants of LCL.722.221 cells are protected from lymphokine-activated killer lysis. Extravillous cytotrophoblast cells and HLA-G expressing choriocarcinoma cells (CC) are not protected. Avril et al25 showed both JAR (HLA class I negative) and JEG-3 (HLA-G and HLA-Cw4 positive) cells were resistant to NK cell lysis by PBL and were equally lyzed by interleukin (IL)-2 stimulated PBL isolated from a given donor. Down regulating HLA-class I on JEG-3 cells by acid treatment did not affect NK or lymphokine activated killer cell (LAK) lysis of CC. Soluble HLA-G produced at the maternofetal interface by cytotrophoblasts and circulating in the body fluids shows a capacity analogous to membranebound structures to inhibit NK cells. Fuzzi et al26 used a specific ELISA to investigate the presence of

sHLA-G molecules in culture supernatant of early embryos obtained by IVF. Analysis of 285 supernatants derived from embryos cultured in groups, corresponding to 101 ART procedures (43 IVF/58 ICSI) identified two groups of patients based on sHLA-G expression. No differences in clinical parameters were identified, but implantation occurred only in women showing sHLA-G molecules in the culture supernatants. They conclude that HLA-G expression is a mandatory, but not sufficient requirement on its own, for the development of successful pregnancy. During pregnancy the fetus represents semiallograft to the mother. The relevance of sHLA-G levels in maternal circulation in relation to the occurrence of characteristic pregnancy disorders were evaluated by measuring plasma sHLA-G levels in women with normal and pathological pregnancies. Compared with normal pregnancy, significantly increased sHLA-G levels were detected in women 

HPE_Chapter12.qxp

7/13/2007

4:51 PM

Page 150


delivered preterm due to intrauterine activation (uncontrollable labor, rupture of fetal membranes, cervical insufficiency, and hemolysis, elevated liver enzymes, low platelets (HELLP) syndrome). Soluble HLA-G levels in women with placental abruption were more than three times lower than in normal pregnancies. Non-parametric discriminant analysis showed women with sHLA-G levels below 9.95 ng/mL had a 7.1 relative risk for development of placental abruption. These findings suggest that the occurrence of pregnancy-associated disease is strongly influenced by maternal sHLA-G plasma levels.27 Pre-eclampsia, a common disorder of first pregnancies, is thought to result from poor placentation and may reflect abnormal maternal immune reaction to the hemiallogenic fetus. HLA-G from fetal derived extravillous cytotrophoblasts may protect from maternal–fetal immune intolerance and allow these cells to invade the uterus. In normal placenta, HLA-G was shown to be expressed in the anchoring extravillous trophoblasts with an increasing gradient of expression in the more invasive cells, while in 9/10 pre-eclamptic placentae HLA-G expression was absent or reduced.28 Trophoblasts lacking HLA-G may be vulnerable to attack by the maternal immune system. A subset of recurrent spontaneous abortions may involve immunological mechanisms. Aberrant T helper 1 (CD4⫹)/T helper 2 (CD8⫹) cytokine profiles have been observed, which are not present in uncomplicated pregnancies. Studies of classical HLA class I and II antigens in relation to recurrent spontaneous abortion are inconclusive. HLA-G is expressed on invasive cytotrophoblasts and exists in both membrane-bound and soluble forms. However, the hypothesis of histoincompatability between placenta and mother does not appear to be supported by the data, although an altered expression profile of HLA-G isoforms or reduced expression of certain isoforms is still possible.29 Coordinated communication between trophoblast cells and their decidual environment is important during the implantation process response. Numerous studies have shown that MHC proteins play an important role in reproduction and development by down regulating the maternal immune response.30 

Initially a great deal of controversy surrounded the issue of whether or not class I MHC antigens are expressed on preimplantation embryos. Class I MHC antigens have been demonstrated on preimplantation mouse embryos through a variety of techniques, including electron microscopy,31,32 125I-lactoperoxidase labeling,33 complement-dependent cytotoxicity,34 cell mediated cytotoxicity assays,35 and a specific ELISA procedure.36,37 Using RT-PCR, mRNA for class I MHC antigens has been detected in preimplantation mouse embryos.38 A comparison of paternally and maternally derived transcripts has shown that active transcription of the class I MHC genes begins at the 2-cell stage of development.39 HLA-G expression has been demonstrated during early preimplantation stages of human development and expression of HLA-G can be correlated with the embryo cleavage rate. Transcripts of mRNA from both the maternal and paternal haplotypes were demonstrated in embryos from the 1-cell zygote to the late blastocyst stage of development in congenic strains of mice.40 These data clearly show that both paternally and maternally inherited MHC class I genes are transcribed from the earliest stages of embryonic development, and suggest that developmental regulation of expression of their protein products is principally at the posttranscriptional level. HLA-G is reputed to prevent allorecognition by maternal cytotoxic lymphocytes and to protect against NK cell mediated lysis of target cells.41 In fact, HLA-G expression on extravillous cytotrophoblast cells probably plays a compulsory role in the development of pregnancy. Soluble HLA-G has been identified in the culture medium of early embryos cultured in groups. Its concentration in the media surrounding groups of embryos has been found to correlate with the rate of embryo cleavage and clinical pregnancy potential.26,42 Direct evidence supporting the role of HLA-G in protecting cytotrophoblasts against NK cytolysis under physiological conditions has been reported by Rouas-Freiss et al.43 In six semi-allogenic combinations of maternal uterine NK cells and their own trophoblast counterparts, as well as in 20 allogenic combinations of maternal uterine NK cells and trophoblasts from different mothers, they showed

HPE_Chapter12.qxp

7/13/2007

4:51 PM

Page 151


HLA-G-mediated protection was abolished by treatment of cytotrophoblasts with HLA-G-specific monoclonal antibodies. An HLA class I negative K562 cell line transfected with the predominant HLA-G1 isoform resulted in similar protection from maternal uterine NK cell lysis. The immune tolerance inducing effects of sHLAG1 on NK cells and T cells were investigated by Fang et al.44 A recombinant plasmid expressing sHLA-G1 was transfected into human lymphoblastoid cells. HLA-G1 dose-dependently inhibited NK cell cytotoxicity as well as dose-dependently inhibited proliferation of activated T cells and induced T cell apoptosis with a dose-saturation character and without antigen specificity. The relationship between sHLA-G expression in trophoblast cells and early embryo development was studied by Wang et al45 who evaluated trophoblasts from 14 cases of spontaneous abortion after IVF/ET and six cases of choriocarcinoma, compared with ten cases of trophoblast cells from normal placenta. In the spontaneous abortion group only 1/14 had positive sHLA-G expression, while all cases of choriocarcinoma had strong sHLA-G expression. All cases of normal villous were also positive for sHLA-G expression. The expression level of sHLA-G decreased in the spontaneous abortion group compared with the normal placenta group. In individual cells, staining on choriocarcinoma cells was stronger than on normal villous cells. High expression in normal villi and low expression in trophoblast cells from spontaneous abortions, suggests that sHLA-G from trophoblast cells does have a relationship to early embryo development. Differences in sHLA-G protein levels were analyzed in women with uncomplicated compared with pre-eclamptic pregnancies. Both serum and placental HLA-G levels decreased significantly in the preeclamptic group compared with the normal pregnant women. High correlation was found between serum and placental HLA-G levels (r ⫽ 0.6). Reduced expression of placental HLA-G and reduced release of protein into the maternal circulation in pre-eclampsia may alter the maternal–fetal immune relationship and thus be involved in the cause of the disorder.46

HUMAN LEUKOCYTE ANTIGEN G AND ASSISTED REPRODUCTIVE TECHNOLOGIES

The role of sHLA-G in ART has only recently been evaluated. Sher et al47 report data on a retrospective cohort of 594 embryos transferred into 201 women aged 28–44 years. IVF/ICSI was performed with embryo transfer (ET) of at least two 7–10-cell embryos 72 hours after retrieval. Media from individually cultured embryos was collected 46 hours postICSI. Group A consisted of 159 women aged ⬍39 years. Group A1 consisted of 101 women with at least one embryo with sHLA-G⫹ (within one SD of the geometric mean), while group A2 consisted of 58 women with all transferred embryos below range. Group B consisted of 42 women aged 39–44 years. Group B1 consisted of 29 women with at least one embryo with sHLA-G⫹, while group B2 consisted of 13 women with all transferred embryos below range. The clinical pregnancy and implantation rates in group A1 were 71% (72/101) and 38%, respectively, compared with those in group A2 which were 22% (13/58) and 9%, respectively. The clinical pregnancy and implantation rates for group B1 were 52% (15/29) and 25%, respectively, compared with those in group B2 which were 15% (2/13) and 5%, respectively. These data suggest that selecting embryos for transfer based on individual sHLA-G expression can maximize pregnancy rates from the transfer of fewer embryos, which can help reduce the incidence of high-order multiple gestation. The same group reported the results of a prospective evaluation of 107 women aged ⬍39 years, with normal ovarian reserve, normal uterine cavity, and two embryos transferred on day 3 scoring GESⱖ70.48 In group A (n ⫽ 51) all embryos expressed sHLA-G within one SD of the geometric mean optical density 0.190 ⫾ 0.006. The pregnancy and implantation rates were 75% (38/51) and 44% (51 sacs/116 embryos), respectively. In group B (n ⫽ 56) all embryos transferred were sHLA-G negative (OD: ⬍0.184). The pregnancy and implantation rates were 23% (13/56) and 14% (20/143), respectively. 

HPE_Chapter12.qxp

7/13/2007

4:51 PM

Page 152


In a further prospective trial, Sher et al49 evaluated 482 women aged 28–43 years with transfer of 7–9-cell embryos. In women ⱕ38 years (n ⫽ 387), when all transferred embryos were HLA-G-negative (n ⫽ 150), the pregnancy and implantation rates were 25% and 13%, respectively. When one transferred embryo was HLA-G-positive (n ⫽ 95), the pregnancy and implantation rates were 63% and 32%, respectively. When all transferred embryos were HLA-G-positive (n ⫽ 142), pregnancy and implantation rates were 69% and 36%, respectively. In women aged 39–43 years (n ⫽ 95), when all transferred embryos were HLA-G-negative (n ⫽ 38), the pregnancy and implantation rates were 29% and 11%, respectively. When one transferred embryo was HLA-G-positive (n ⫽ 21), pregnancy and implantation rates were 38% and 15%, respectively. When all transferred embryos were HLA-G-positive (n ⫽ 36), pregnancy and implantation rates were 61% and 26%, respectively. Using sHLA-G measurement in culture media at 46 hours postICSI to select embryos for transfer provided a mechanism to lower the incidence of multiple gestations. Noci et al50 evaluated sHLA-G using an ELISA employing monoclonal antibody MEM-G9 in 318 media of single embryo culture after 48 hours of culture. They also correlated the presence of sHLA-G with embryo morphology and pregnancy obtained in that cycle. No correlation between sHLA-G and embryo morphology was noted. Pregnancy was observed only when one embryo was sHLA-G⫹ with a 1 ng/mL limit of sensitivity (mean 4 ng/mL); 26/66 patients had no sHLA-G expression and no pregnancy occurred (although overall, pregnancy occurred in only 9/66). Of seven patients with no sHLA-G expression in the first cycle who did a second cycle, 4/5 had sHLA-G⫹ embryos and all four became pregnant. Yie et al51 report a retrospective evaluation of media from 386 grouped embryo cultures assayed for sHLA-G after 72 hours culture. Soluble HLA-G was detected in 270 samples. Secretion was independent of embryo grade or patient age. The cleavage rate of embryos secreting sHLA-G was significantly higher than for those lacking it (mean 6.7 vs 5.8 cells on day 3). The live birth rate in embryos with 

sHLA-G was higher than in those without (48% vs 17%). Combining secretion of HLA-G and cleavage rate was most predictive of pregnancy. The authors concluded that embryonic secretion of sHLA-G is variable, sHLA-G secretion is associated with higher cleavage and pregnancy rates, and sHLA-G is a better independent predictor than cleavage alone. Finally, the combination of sHLA-G detection and high cleavage rate was best predictor of outcome. Taken together these studies provide strong support to the notion that sHLA-G measurement in the supernatant of cultured embryos can be used to predict embryo viability. In summary, the accumulated literature supports the following conclusions. (1) HLA-G is a non-classical MHC protein expressed on extravillous cytotrophoblasts, which appears to modulate maternal immune function by down regulating the actions of decidual NK cells through interaction with the CD94 killer inhibitory receptor. (2) Both membrane-bound and soluble forms of HLA-G appear to be immunoprotective for the conceptus. (3) Specific monoclonal antibodies have been developed to detect HLA-G expression. (4) Measurement of sHLA-G correlates with pregnancy outcome. Presence of sHLA-G in embryo culture media predicts the outcome of IVF/ET. Combining sHLA-G secretion with other established embryo markers may be most clinically useful. REFERENCES 1. Fernandez N, Cooper J, Sprinks M et al. A critical review of the role of the major histocompatability complex in fertilization, preimplantation development and feto-maternal interactions. Hum Reprod Update 1999; 5: 234–48. 2. Bjorkman PJ, Saper MA, Samraoui B et al. The foreign antigen binding site and T-cell recognition regions of class I histocompatibility antigens. Nature 1987; 329: 512–18. 3. McMaster MT, Bass KE, Fisher SJ. Human trophoblast invasion. Autocrine control and paracrine modulation. Ann NY Acad Sci 1994; 734: 122–31. 4. Jurisicova A, Casper RF, MacLusky NJ, Mills GB, Librach CL. HLA-G expression during preimplantation human embryo development. Proc Natl Acad Sci USA 1996; 93: 161–5. 5. Le Bouteiller P, Solier C, Proell J et al. Placental HLA-G protein expression in vivo: where and what for? Hum Reprod Update 1999; 5: 223–33.

HPE_Chapter12.qxp

7/13/2007

4:51 PM

Page 153


6. Geraghty DE, Koller BH, Orr HT. A human histocompatibility complex class I gene that encodes a protein with a shortened cytoplasmic segment. Proc Natl Acad Sci USA 1987; 85:227–31. 7. Ellis SA, Palmer MS, McMichael AJ. Human trophoblasts and the human choriocarcinoma cell line BeWo express a truncated HLA class I molecule. J Immunol 1990; 144: 731–5. 8. Le Bouteiller P, Blaschitz A. The functionality of HLA-G is emerging. Immunol Rev 1999; 167: 233–44. 9. Ober C, Aldrich C. HLA-G polymorphisms: neutral evolution or novel function? J Reprod Immunol 1997; 36: 1–21. 10. van der Ven K, Skrablin S, Ober C, Krebs D. HLA-G polymorphisms: ethnic differences and implications for potential molecule function. Am J Reprod Immunol 1998; 40: 145–57. 11. Jurisicova A, Casper RF, MacLusky NJ, Librach CL. Embryonic human leukocyte antigen-G expression: possible implications for human preimplantation development. Fertil Steril 1996; 65: 997–1002. 12. Moreau P, Paul P, Rouas-Freiss N et al. Molecular and immunologic aspects of the nonclassical HLA class I antigen HLA-G: evidence for an important role in the maternal tolerance of the fetal allograft. Am J Reprod Immunol 1998; 40: 136–44. 13. Blaschitz A, Hutter H, Dohr G. HLA class I protein expression in the human placenta. Early Pregnancy 2001; 5: 67–9. 14. Morales PJ, Pace JL, Platt JS et al. Placental cell expression of HLA-G2 isoforms is limited to the invasive trophoblast phenotype. J Immunol 2003; 171: 6215–24. 15. Gonen-Gross T, Gazit R, Achdout H et al. Special organization of the HLA-G protein on the cell surface. Hum Immunol 2003; 64: 1011–6. 16. Le Bouteiller P, Legrand-Abravanel F, Solier C. Soluble HLA-G1 at the materno-foetal interface – a review. Placenta 2003; 24 (Suppl A): S10–5. 17. Hunt JS, Pace JL, Morales PJ, Ober C. Immunogenicity of the soluble isoforms of HLA-G. Mol Hum Reprod 2003; 9: 729–35. 18. Hviid TV, Hylenius S, Rorbye C, Nielsen LG. HLA-G allelic variants are associated with differences in the HLA-G mRNA isoform profile and HLA-G mRNA levels. Gynecol Obstet Fertil 2003; 31: 782–5. 19. Nagamatsu T, Fujii T, Yamashita T et al. Hypoxyia does not reduce HLA-G expression on extravillous cytotrophoblasts. J Reprod Immunol 2004; 63: 85–95. 20. Weetman AP. The immunology of pregnancy. Thyroid 1999; 9: 643–6. 21. Soller C, Aguerre-GM, Lenfent F, et al. Secretion of pro-apototic intron 4-retaining soluble HLA-G1 by human villous trophoblast. Eur J Immunol 2002; 32(2): 3576–86. 22. King A, Hiby SE, Verma S, Burrows T, Gardner L, Loke YW. Uterine NK cells and trophoblast HLA class I molecules. Am J Reprod Immunol 1997; 37: 459–62. 23. Pende D, Sivori S, Accame L et al. HLA-G recognition by human natural killer cells. Involvement of CD94 both as inhibitory and as activating receptor complex. Eur J Immunol 1997; 27: 1875–80. 24. Mandelboim O, Pazmany L, Davis DM et al. Multiple receptors for HLA-G on human natural killer cells. Proc Natl Acad Sci USA 1997; 94: 14666–70. 25. Avril T, Jarousseau AC, Watier H et al. Trophoblast cell line resistance to NK lysis mainly involves an HLA class I-independent mechanism. J Immunol 1999; 162: 5902–9. 26. Fuzzi B, Rizzo R, Criscouli L et al. HLA-G expression in early embryos is a fundamental prerequisite for the obtainment of pregnancy. Eur J Immunol 2002; 32: 311–5. 27. Steinborn A, Rebmann V, Scarf A, Sohn C, Grosse-Wilde H. Placental abruption is associated with decreased maternal plasma levels of soluble HLA-G. J Clin Immunol 2003; 23: 307–14.

28. Goldman-Wohl DS, Ariel I, Greenfield C et al. Lack of human leukocyte antigen-G expression in extravillous trophoblasts is associated with pre-eclampsia. Mol Hum Reprod 2000; 6: 88–95. 29. Hviid TV, Hylenius S, Hoegh AM, Kruse C, Christiansen OB. HLA-G polymorphisms in couples with recurrent spontaneous abortions. Tissue Antigens 2002; 60: 122–32. 30. Warner CM, Brownell MS, Ewoldsen MA. Why aren’t embryos immunologically rejected by their mothers? Biol Reprod 1988; 38: 17–29. 31. Searle RF, Sellens MH, Elson J, Jenkinson EJ, Billington WD. Detection of alloantigens during preimplantation development and early trophoblast differentiation. J Exp Med 1977; 143: 348–359. 32. Warner CM, Spannaus DJ. Demonstration of H-2 antigens on preimplantation mouse embryos using conventional antisera and monoclonal antibody. J Exp Zool 1984; 230: 37–52. 33. Cozad KM, Warner CM. Detection of H-2 antigens on 8-cell mouse embryos. J Exp Zool 1982; 221: 213–17. 34. Ewoldsen MA, Ostlie NS, Warner CM. Killing of mouse blastocyst stage embryos by cytotoxic T lymphocytes directed to major histocompatability complex antigens. J Immunol 1987; 138: 2764–70. 35. Goldbard SB, Gollnick SO, Warner CM. A highly sensitive method for the detection of cell surface antigens on preimplantation mouse embryos. J Immunol Methods 1984; 68: 137–46. 36. Warner CM, Gollnick SO, Flaherty L, Goldbard SB. Analysis of Qa-2 antigen expression by preimplantation mouse embryos: possible relationship to the Ped gene product. Biol Reprod 1987; 36: 611–16. 37. Tian Xu Y, Warner CM. Removal of Qa-2 antigen alters the Ped gene phenotype of preimplantation mouse embryos. Biol Reprod 1992; 47: 271–6. 38. Arcellana-Panlilio MY, Schultz GK. Expression of MHC class I genes during preimplantation mouse embryo development. Presented at Program of the 31st Annual Meeting of the Society for Cell Biology, Boston, MA, 1991: Abstract 303. 39. Jin P, Meyer TE, Warner CM. Control of embryo growth by the Ped gene: use of reverse transcriptase polymerase chain reaction (RT-PCR) to measure mRNA in preimplantation embryos. Assisted Reprod Technol Androl 1992: 3377–83. 40. Sprinks MT, Sellens MH, Dealtry GB, Fernandez N. Preimplantation mouse embryos express MHC class I genes before the first cleavage division. Immunogenetics 1993; 38: 35–40. 41. Kanai T, Fujii T, Kozuma S et al. Soluble HLA-G influences the release of cytokines from allogenic peripheral blood mononuclear cells in culture. Mol Hum Reprod 2001; 7: 195–200. 42. Menicucci A, Noci I, Fuzzi B et al. Non-classic sHLA class I in human oocyte culture medium. Hum Immunol 1999; 60: 1054–7. 43. Rouas-Freiss N, Goncalves RM, Menier C, Dausset J, Carosella ED. Direct evidence to support the role of HLA-G in protecting the fetus from maternal uterine natural killer cytolysis. Proc Natl Acad Sci USA 1997; 94: 11520–5. 44. Fang CY, Wu XW, Liang ZH et al. Immune tolerance inducing effects of soluble human leukocyte antigen G1 on natural killer cells and T cells. Zhonghua Yi Xue Zhi 2003; 83: 584–7. 45. Wang Q, Zhuang G, Zhou C et al. Relationship between trophoblast cell’s human leukocyte antigen G expression and early embryo development. Zhonghua Fu Chan Ke Za Zhi 2002; 37: 723–5. 46. Yie SM, Li LH, Li YM, Librach C. HLA-G protein concentrations in maternal serum and placental tissue are decreased in preeclampsia. Am J Obstet Gynecol 2004; 191: 525–9. 47. Sher G, Keskintepe L, Nouriani M, Roussev R, Batzofin J. Expression of sHLA-G in supernatants of individually cultured 46-h embryos: a potentially valuable indicator of ‘embryo competency’ and IVF outcome. Reprod Biomed Online 2004; 9: 74–8.



HPE_Chapter12.qxp

7/13/2007

4:51 PM

Page 154


48. Sher G, Keskintepe L, Fisch JD et al. Soluble human leukocyte antigen G expression in phase I culture media at 46 hours after fertilization predicts pregnancy and implantation from day 3 embryo transfer. Fertil Steril 2005a; 83: 1410–3. 49. Sher G, Keskintepe L, Fisch JD et al. Influence of early ICSI-derived embryo sHLA-G expression on pregnancy and implantation rates: a prospective study. Hum Reprod 2005b; 20: 1359–63.



50. Noci I, Fuzzi B, Rizzo R et al. Embryonic soluble HLA-G as a marker of developmental potential in embryos. Hum Reprod 2005; 20: 138–46. 51. Yie SM, Balakier H, Motamedi G, Librach CL. Secretion of human leukocyte antigen-G by human embryos is associated with a higher in vitro fertilization pregnancy rate. Fertil Steril 2005; 83: 30–6.

HPE_Chapter13.qxp

7/13/2007

4:52 PM

Page 155

13. Immunological aspects of embryo development Carol M Warner

INTRODUCTION

Preimplantation embryos express paternal as well as maternally derived proteins, yet they are not rejected by the maternal immune system. This enigma was first recognized by Medawar over 50 years ago1 when he wrote ‘. . . how does the pregnant mother contrive to nourish within itself, for many weeks or months, the foetus that is an antigenically foreign body? Is pregnancy accompanied by any physiological changes which may in some degree prevent the foetus, qua tissue homograph, from immunizing the mother against itself?’ In spite of extensive research in this area, our knowledge of how this comes about is still incomplete. This review focuses on immunological aspects of the preimplantation embryo and not on later stages of development. The reason for this is that embryos are relevant to clinics performing assisted reproductive technologies (ART) such as in vitro fertilization (IVF) and intracytoplasmic sperm injection (ICSI) only at the preimplantation stage. There are many limitations associated with research on human embryos, including the scarcity of material, the outbred nature of the human population making every embryo unique, as well as lack of federal funding in the United States and limited funding in other countries. Therefore, animal models are required for extensive research on preimplantation embryos. The mouse is a particularly good model system because of the availability of genetically homogeneous inbred strains, congenic strains, knockout strains, and transgenic strains. In addition, mice are relatively inexpensive compared with other animal models such as primates or domestic animal species. For these reasons research on the mouse model system will be highlighted in this review. The first objective of this review is to give the reader a brief overview of the immune system,

including a discussion of innate and adaptive immunity, cells and molecules involved in the immune system, and the fundamentals of the major histocompatibility complex (MHC). A brief overview of preimplantation development is then presented in the context of MHC expression by preimplantation embryos and considerations of the embryo as an allograft. A discussion of the discovery of an MHC encoded gene, the Ped (preimplantation embryo development) gene, follows, along with a summary of the properties of the Ped gene product, Qa-2 in the mouse, and its homolog, HLA-G, in humans. Both Qa-2 and HLA-G have a role in enhancing embryo survival. Clinically relevant properties of the mouse Ped gene, during both the preimplantation period and later in life, are discussed. Concluding remarks challenge the reader to think about how new methods in nanomedicine might be applied to preimplantation embryos in order to gain a further understanding of the relevance of the immune system to reproduction.

OVERVIEW OF THE IMMUNE SYSTEM

The major function of the immune system is to protect an organism from damage induced by exogenously introduced foreign pathogens such as bacteria, viruses, and fungi, and from endogenously produced cancer cells. This is accomplished by both immune cells and their products, called cytokines, as well as by specialized molecules such as antibodies that recognize foreign substances, termed antigens. For convenience, immunologists have divided the immune system into two major categories, the innate immune system and the adaptive immune system. Several excellent textbooks are available that 

HPE_Chapter13.qxp

7/13/2007

4:52 PM

Page 156


discuss all aspects of the field of immunology.2,3 For a more in-depth discussion of the immune system with references to the primary literature, the reader is directed to Paul’s Fundamental Immunology.4 Briefly, the innate immune system is non-specific, present before exposure to any pathogen, and provides the first line of defense against disease. Some aspects of the innate immune system are maintaining a mechanical barrier (e.g. the skin and mucous membranes), a physiological barrier (e.g. temperature, pH, and certain chemicals, such as lysozyme and complement), a cellular barrier (e.g. macrophages, neutrophils, and natural killer (NK) cells), and an inflammatory barrier (e.g. acute-phase proteins). On the other hand, the adaptive immune system is not triggered until the organism receives an antigenic challenge. Unlike innate immunity, adaptive immunity displays four special characteristics: specificity for antigen, diversity, memory, and self/non-self recognition. The adaptive immune system involves T and B lymphocytes and four major classes of membranebound proteins: membrane-bound antibodies on B cells, T cell receptors, and class I MHC proteins, class II MHC proteins. (Class I MHC proteins are also involved in the innate immune system.) In addition, antigen presenting cells such as macrophages and dendritic cells are part of the adaptive immune system. We have recently proposed an interesting theory5 which suggests that the mammalian immune system may have developed as an accidental consequence of a mechanism whereby semi-allogeneic fetuses can escape destruction by the mother. Thus, a mechanism to distinguish self from non-self and to tolerate the allogeneic fetus may have resulted in the ability to recognize foreign pathogens. However, regardless of whether this theory is correct, both the innate and the adaptive immune systems are intimately involved in the protection of the preimplantation embryo from immunological destruction by the mother.

leading to blastocyst formation. Images showing the stages of mouse preimplantation development are shown in Figure 13.1. Outstanding images of human preimplantation embryos can be found in reference 6. This cleavage process takes 4–5 days in the mouse and 5–7 days in humans. The images shown in Figure 13.1 are from differential interference contrast (DIC) microscopy and represent what the clinician is used to seeing in the laboratory. However, the DIC pictures are somewhat deceptive because the surrounding zona pellucida is optically clear so that the observer looks right through the zona pellucida when using DIC microscopy. This point is emphasized when one examines scanning electron

Oocyte

Zygote

2-cell

8-cell

Morula

Blastocyst

THE IMMUNE SYSTEM AND PREIMPLANTATION DEVELOPMENT

After fertilization the resulting preimplantation embryo undergoes a series of cleavage divisions 

Figure 13.1 Images of a mouse oocyte and preimplantation mouse embryos. The images are from differential interference contrast microscopy.

HPE_Chapter13.qxp

7/13/2007

4:52 PM

Page 157

IMMUNOLOGICAL ASPECTS OF EMBRYO DEVELOPMENT

A

B

Figure 13.2 Scanning electron microscope images of a mouse blastocyst with the zona pellucida intact (A) and the zona pellucida removed (B). Reprinted with slight modification and permission from reference 7.

microscope images of a mouse blastocyst with the zona pellucida intact (Figure 13.2A) and with the zona pellucida removed (Figure 13.2B). In many ways the zona pellucida is a physical barrier to an immunological attack on the embryo, in a similar way that the skin is a barrier to invasion by pathogens in the innate immune system. THE ZONA PELLUCIDA

Extensive studies have been performed on the biochemical composition of the zona pellucida in the mouse,8–10 and more recently in humans.11,12 In the mouse, the zona pellucida is composed of three proteins, designated mZP1, mZP2, and mZP3. Recently a fourth zona pellucida protein was discovered in humans11 and it was originally named ZPB, but has more recently been renamed ZP4.12 Thus, the four human zona pellucida proteins are ZP1, ZP2, ZP3, and ZP4. The existence of ZP4 in human oocytes has been shown at both the mRNA level by using reverse transcriptase polymerase chain reaction (RT-PCR) and the protein level by using tandem mass spectrometry.11 There are very few data in the literature defining the function of the four human ZP proteins. In the mouse, it was thought for many years that the

carbohydrate moieties of mZP2 and mZP3 were involved in sperm–egg binding and fertilization.8,9 However, this view has recently been challenged,10 by analyzing data from knockout mice, and it is unclear whether fertilization depends on carbohydrate– carbohydrate interactions, carbohydrate–protein interactions, protein–protein interactions, or some combination of all three mechanisms.12 The realization that the zona pellucida is involved in fertilization has led to the idea of developing a contraceptive vaccine based on immunization with zona pellucida proteins.13 However, there is evidence that anti-zona antibodies cause ovarian dysfunction and infertility,14 so this approach needs more investigation before it is ready for clinical trials. Aside from a putative role in fertilization, the zona pellucida has been used to evaluate human embryos in ART clinics. Evidence has been presented that surface morphology15 and thickness variation, which was originally proposed by Cohen et al16 and recently confirmed by Sun et al,17 as well as amino acid sequence of the ZP proteins18 are related to pregnancy outcome after ART procedures. In addition, a number of laboratories have engaged in partial disruption of the zona pellucida, in a procedure called ‘assisted hatching’ to increase the pregnancy rate in ART. Disruption of the zona pellucida can be accomplished by mechanical, chemical, or laser methods. A recent meta-analysis of 23 randomized trials involving 2668 women has shown that although assisted hatching improves the odds of a clinical pregnancy, there is no effect on live birth rates.19 Thus, it seems that, in general, disruption of the zona pellucida in vitro during the preimplantation period does not affect pregnancy outcome after transfer to the recipient mothers. A further discussion of the zona pellucida as a mechanical barrier to the destruction of the embryo in vivo by the maternal immune system is in order. The zona pellucida is a very loose matrix, and allows quite large molecules to traverse it. For example, even IgM antibody with a molecular weight of approximately 950 000 can traverse the zona pellucida. The zona pellucida is, however, protective against attack by immune cells, as depicted in Figure 13.3. Figure 13.3 shows that molecules can traverse the zona pellucida, 

HPE_Chapter13.qxp

7/13/2007

4:52 PM

Page 158


MF

NK cell

EMBRYO

CTL

Molecules

Zona pellucida

Figure 13.3 Schematic of the protection of the preimplantation embryo from attack by immune cells: NK, natural killer cell; CTL, cytotoxic T lymphocyte; M⌽, macrophage. The zona pellucida is porous to molecules, including large macromolecules, but impervious to attack by cytotoxic cells.

but immune cells from both the innate and adaptive immune systems are blocked from direct interaction with the membranes of the embryonic cells. We have shown this experimentally in an elegant series of experiments in which cytotoxic T lymphocytes (CTL) directed to major histocompatibility complex (MHC) antigens were able to kill mouse blastocysts with the zona pellucida removed, but not with the zona pellucida intact.20 These experiments not only confirmed that the zona pellucida is a barrier to immune cells, but also showed unequivocally that preimplantation mouse embryos express MHC antigens. To the best of my knowledge, similar studies to determine the effect of the zona pellucida on possible killing of embryos by macrophages or natural killer (NK) cells have not yet been reported. In support of the idea that the zona pellucida protects the embryo from destruction by maternal immune cells is a report that shows that the implantation rate for human embryos with a small hole in their zona pellucida was higher when the recipient mothers were immunosuppressed.21 At the end of the preimplantation period the embryo hatches from the zona pellucida and implants in the uterus. Thus, there is a period of time, between 

hatching and implantation, when the protection of the zona pellucida against immunological attack by cells is not present. Due to the difficulty of designing a good experimental system, very little research has been done on the period of development just after hatching and before implantation. WHAT IS UNDER THE ZONA PELLUCIDA?

The real challenge in understanding the immunological aspects of preimplantation embryos is to generate a topological map of all proteins that are expressed under the zona pellucida, on the embryonic cell surface. In spite of the fact that we now know that humans and mice have 20 000–25 000 genes in their genomes, very few of the proteins encoded by these genes have been identified on the embryonic cell surface. One problem is the large size of preimplantation embryos, which makes proteins that are expressed at low levels difficult to detect because they may be far apart on the embryonic cell surface. Immunofluorescence is a relatively insensitive technique, because there is no enzymatic amplification step. Enzyme linked immunosorbent assay (ELISA)22 and Immuno-PCR23,24 are more sensitive procedures for detection of proteins on the embryonic cell surface. However, all of these protein detection techniques require antibody to the specific protein of interest. In contrast to protein detection, mRNA expression studies need no antibody and are relatively simple to perform; DNA sequence data allow the design of primers, and RT-PCR allows amplification of specific mRNAs in single embryos. In addition, the use of microchips has allowed the identification of thousands of genes that are transcribed during preimplantation mouse development.25–29 However, whereas the absence of mRNA precludes protein expression, the presence of mRNA does not guarantee protein expression. Therefore, the ability to detect as many proteins on the embryonic cell surface as possible would be highly desirable, in order to fully understand the mechanisms that allow embryos to escape surveillance by the maternal immune system. Proteins of the MHC are of particular relevance to this review.

HPE_Chapter13.qxp

7/13/2007

4:52 PM

Page 159


THE MAJOR HISTOCOMPATIBILITY COMPLEX

The MHC is a set of linked genes, located on human chromosome 6 and mouse chromosome 17, which encode three classes of proteins, class I, class II, and class III. An up-to-date review of the MHC can be found in a chapter by Margulies and McCluskey.30 Simplified diagrams of the human MHC (HLA) and mouse MHC (H-2) are shown in Figure 13.4. It is clear from these diagrams that the MHC is not orthologous between species, because there is no one-to-one correspondence between the human genes and the mouse genes. The MHC class I genes are a major focus of this review. Figure 13.4 shows that there are two classes of MHC class I genes, designated class Ia and class Ib. The MHC class Ia genes are the ‘classical’ genes the protein products of which are involved in both innate and adaptive immunity. In humans, the MHC class Ia protein products are HLA-A, HLA-B, and HLA-C. In the mouse, the MHC class Ia protein products are H-2K and H-2D, found in all mouse strains, and H-2L (encoded in the D region) found in only some mouse strains. The location of the K region in the mouse MHC, displaced from the rest of the MHC class Ia genes, has no known significance. In humans, all of the MHC class I genes, classical (Ia) and non-classical (Ib), are clustered at the telomeric end of the HLA complex.

Human – Chromosome 6 Region

II

III

B

C

E

A

G

F

Class of protein

II

III

Ia

Ia

Ib

Ia

Ib

Ib

HLA-G Mouse – Chromosome 17 Region

K

I

S

D

Q

TL

Class of protein

Ia

II

III

Ia

Ib

Ib

Qa-2

Figure 13.4 Schematic of the major histocompatibility complex (MHC) genes of the human and the mouse.

The MHC class Ib genes differ significantly between mouse and man. The protein products of the MHC class Ib genes have functions in both the innate immune response and other non-immunological venues. In humans, there are three MHC class Ib genes that encode the protein products HLA-E, HLA-F, and HLA-G. In the mouse, there are 15 known MHC class Ib genes in the Q region and 24 known class Ib genes in the TL region, with the number of genes in each region varying among mouse strains. Because there is no direct gene-to-gene correlation between mouse and man (orthologs), it is difficult to decipher which, if any, of these mouse MHC class Ib genes are functional homologs with human MHC class Ib genes. Of especial interest to this review are Qa-2 protein encoded in the Q region of the mouse MHC and HLA-G encoded in the G region of the human MHC. Qa-2 and HLA-G MHC class Ib proteins have been shown to be functional homologs31,32 and are discussed in detail later in this review. Interestingly, one consequence of deciphering the complete genome sequences of mouse and man was the realization that only 80% of mouse and human genes are orthologs and the other 20% exist as functional homologs. The MHC belongs to this latter group of genes. The MHC is so important in the understanding of the immune response that not one, but two Nobel prizes have been awarded to researchers working on the role of the MHC in the immune response. The first, awarded in 1980 to George Snell, Jean Dausset, and Baruj Benacerraf, was for the discovery that the MHC was involved in the immune response to foreign antigens. The second, awarded in 1996 to Peter Doherty and Rolf Zinkernagel, was for the discovery of the role of the MHC in antigen recognition by T cells. However, we now know that the MHC is also involved in reproduction. The MHC class I proteins are most directly relevant to immunological rejection. Because the MHC contains dozens of genes, for simplicity the whole group of genes within an MHC on a single chromosome is called a ‘haplotype.’ It should be noted that each class I MHC gene has many polymorphic variants in the population, leading to the existence of 

HPE_Chapter13.qxp

7/13/2007

4:52 PM

Page 160


Mother A/B

A/C

X

A/D

Father C/D

B/C

B/D

Figure 13.5 Inheritance of MHC haplotypes. Hypothetical haplotypes are represented by A, B, C, and D. Each embryo receives one haplotype of maternal origin and one haplotype of paternal origin.

thousands of different haplotypes. During reproduction, each embryo receives one MHC haplotype of maternal origin and one MHC haplotype of paternal origin, as depicted in Figure 13.5. In this diagram each haplotype is represented by a capital letter, A, B, C, or D. As can be seen from Figure 13.5, all offspring are histoincompatible (semi-allogeneic) with the mother, as stated by Medawar and quoted in the opening paragraph of this review. An easy answer to the lack of immunological rejection during the preimplantation period would be that preimplantation embryos do not express MHC class I proteins. However, this is not true. MHC CLASS I PROTEINS ON PREIMPLANTATION EMBRYOS

Initially there was a great deal of controversy about whether or not preimplantation mouse embryos express MHC class I proteins. It now seems that these initial studies, based on using immunofluorescence, were simply not sensitive enough to detect MHC class I proteins on preimplantation embryos. The first clear demonstration of MHC class I proteins on preimplantation embryos was by Searle et al33 using electron microscopy, and was later confirmed by our group34 using a similar electron microscopic technique. A variety of techniques other than electron microscopy have since been used to 

demonstrate that MHC class I proteins are expressed on preimplantation mouse embryos. These studies include 125I-lactoperoxidase labeling,35 complementdependent cytotoxicity,36,37 cell-mediated cytotoxicity,20 ELISA,22, 38–40 and Immuno-PCR.23,41 Early studies on human preimplantation embryos reported the absence of MHC class I proteins.42,43 However, although this idea has persisted,44 proper, highly sensitive techniques have never been used to systematically evaluate preimplantation human embryos for MHC class I expression. Moreover, at least one MHC class I protein, HLA-G, has been shown to be expressed by preimplantation embryos.45 In addition, human embryonic stem cells created from human blastocysts have been shown to express MHC class I proteins.46 Therefore, it seems highly likely that with the use of appropriate techniques and experimental protocols, expression of MHC class I proteins on human preimplantation embryos will be found, as they are found on mouse preimplantation embryos. MEMBRANE-BOUND VERSUS SOLUBLE MHC CLASS I PROTEINS ON PREIMPLANTATION EMBRYOS

The big question, then, is why are MHC class I proteins expressed by preimplantation embryos? The answer to this question is entangled in the fact that preimplantation embryos not only express membrane-bound MHC class I molecules, but also express soluble MHC class I molecules. Of particular clinical relevance are the recent reports from ART clinics that preimplantation human embryos produce soluble HLA-G and that the presence of soluble HLA-G enhances the chance of pregnancy success.47–52 However, at least two groups have been unable to detect HLA-G in the supernatants of preimplantation human embryos.53,54 These provocative findings have been the subject of debate, and are addressed in a recent issue of Molecular Human Reproduction with comments from the Editor-inChief 55 and an excellent editorial by Sargent.56 The clinical relevance of a test for soluble HLA-G in preimplantation embryo supernatants and its role in preimplantation embryo development and pregnancy outcome after ART56 is reviewed in

HPE_Chapter13.qxp

7/13/2007

4:52 PM

Page 161


Chapter 12 of this book, and will not be pursued further here. In addition, a thoughtful and complete review of the role of both membrane-bound and soluble HLA-G in human reproduction has recently been published.57 Whether or not soluble HLA-G turns out to be a clinically relevant marker for the prediction of pregnancy success, the existence of soluble isoforms of MHC class I proteins is of direct relevance to the embryo’s escape from destruction by the maternal immune system. MHC class I proteins other than HLA-G are known to exist in soluble form, but to my knowledge no attempt has been made to assay for the presence of soluble HLA-A, HLA-B, or HLA-C in the supernatants of preimplantation embryos. This is consistent with the previous discussion about the lack of rigorous experiments to test for the presence of membrane-bound MHC class I molecules other than HLA-G on preimplantation human embryos. As pointed out previously, the zona pellucida protects the preimplantation embryo from direct contact and killing by cytotoxic cells. Moreover, the zona pellucida is porous and allows the exchange of macromolecules between the embryo and its environment, whether that environment is the culture medium in the ART clinic or the environment in the uterus. Several reports from animal models indicate that soluble MHC class I molecules suppress the immune response and prolong the survival of tissue allografts.58,59 (The embryo is an allograft, as shown in Figure 13.5.) Several reports in human systems have also shown that binding of soluble MHC class I proteins to cytotoxic T lymphocytes (CTL) induces them to undergo apoptosis.60,61 In addition, it has been shown that soluble HLA-G inhibits the activity of both CTLs (CD8⫹ cells) and natural killer (NK) cells, prevents proliferation of helper T cells (CD4⫹ cells), and produces immunological tolerance in dendritic cells.62–66 Therefore, cells of both the innate and adaptive immune systems are inhibited by soluble MHC class I molecules. This may be an important mechanism in protecting the embryo from attack by the maternal immune system during the window of time between hatching and implantation, when the zona pellucida no longer presents a physical barrier to attack by immune cells.

THE PED GENE

One of the major criteria used to determine which embryos should be transferred to the mother after ART procedures is the embryo’s rate of development.67 The transfer of faster developing embryos generally leads to a greater chance of pregnancy success than does transfer of slower developing embryos. The rate at which preimplantation embryos cleave is dependent on both environmental and genetic parameters. The Ped (preimplantation embryo development) gene, which was discovered in my laboratory,68 has a profound influence on the rate of development of preimplantation mouse embryos. Of particular relevance to the topic of this review is that the Ped gene product is an MHC class Ib protein, Qa-2 in the mouse and HLA-G in humans, and these proteins are encoded in the Q region of the mouse MHC and the G region of the human MHC (Figure 13.4). Embryos that express Qa-2/HLA-G on their cell surface cleave at a faster rate than embryos that do not express these proteins, and therefore are more likely to lead to a successful pregnancy. As discussed previously, it is difficult to image MHC class I proteins on preimplantation embryos because of their low concentration per surface area. However, we have recently succeeded in visualizing the Ped gene product, Qa-2 protein, on the mouse blastocyst cell surface by immunogold labeling, as shown in Figure 13.6 (Newmark and Warner, unpublished). Based on the analysis of representative electron microscopic samples, we estimate that there are approximately 2000 Qa-2 protein molecules on the surface of a typical mouse blastocyst. This is the first direct demonstration of Qa-2 protein on the embryonic cell surface and emphasizes that even a small amount of cell-surface protein can have a very big effect on development and reproduction. MECHANISM OF ACTION OF THE PED GENE

The major challenge in elucidating the role of the Ped gene in reproduction is to understand how the presence of a cell surface molecule signals that the 

HPE_Chapter13.qxp

7/13/2007

4:52 PM

Page 162


we have recent evidence that the pathway to activation is via phosphatidylinositol-3 (PI-3) kinase (De Fazio and Warner, unpublished). Figure 13.7 shows a hypothetical scheme for Qa-2/HLA-G activation of cells, based upon using cross-linking of Qa-2 on T cells to induce proliferation of the cells (unpublished results). In these experiments we showed that cross-linking of Qa-2 with anti-Qa-2 antibody, in the presence of a second signal provided by phorphol myristate acetate (PMA), induces activation of PI-3 kinase and phosphorylation of Akt. In addition, we have shown that Fyn co-immunoprecipitates with anti-Qa-2 antibody, implying that Qa-2 and Fyn are in close physical proximity. This suggests, but not yet proves, that Fyn is involved in the activation scheme. Thus, we are well on the way towards understanding the pathways by which Qa-2 enhances cell cycle progression. 200 nm PRESENCE VERSUS ABSENCE OF QA-2/HLA-G

Figure 13.6 Image of Qa-2 on the surface of a mouse blastocyst. The image is from transmission electron microscopy with immunogold labeling of Qa-2 protein. The arrow points to the gold particle that identifies the location of Qa-2.

preimplantation embryo should increase its cleavage rate. We have shown that Qa-2 must be expressed on the cell surface in order for the protein to mediate a rapid rate of preimplantation development. This is based on the fact that enzymatic cleavage of Qa-2 from the embryonic cell surface slows the rate of preimplantation development.38 Neither Qa-2 nor HLA-G reach far enough into the inside of the cell to act independently as signaling molecules. Qa-2 is attached to the outer leaflet of the cell membrane by a glycosylphosphatidylinositol (GPI) linkage, whereas HLA-G traverses the cell membrane but has a truncated cytoplasmic tail of only six amino acid residues. Therefore, Qa-2 and HLA-G must bind to accessory molecules(s) to transmit a signal from the cell surface to the inside of the cell. We also know that both Qa-2 and HLA-G are located in lipid rafts,31 implying a signaling role for these molecules. Although we do not yet know the nature of the Qa-2/HLA-G transmembrane binding partner(s), 

AND EMBRYO SURVIVAL

One curious aspect of the Ped gene product, Qa-2 in the mouse, and HLA-G in humans, is that although the absence of these proteins is compatible with embryo survival, their presence enhances pregnancy success.57,69,70 The majority of mouse strains and humans have genes that can potentially be transcribed and translated to produce Qa-2/HLA-G proteins. In humans about 3% of the population has a ‘null allele’ at the DNA level, which results in the lack of HLA-G protein expression.71 We have recently shown that the genes encoding Qa-2 in the wild mouse population are absent at a similar frequency (Byrne, Jones, and Warner, unpublished). We speculate that the non-expression of Qa-2/HLA-G has been retained through evolution because it is advantageous to have embryos that develop at a slower rate under some circumstances. In order for implantation to occur, the timing of cleavage divisions of the preimplantation embryo and the preparation of the uterus for implantation must be coordinated. Embryos that develop at a slow rate can wait for the uterus to catch up, but embryos that have missed the opportunity to implant because they are developing too rapidly will die. Therefore, it seems

HPE_Chapter13.qxp

7/13/2007

4:52 PM

Page 163


Antibody to Qa-2

Y Qa-2

α α α

α

α

α α

β

α

β

Hypothetical transmembrane signaling partner

PIP3

Fyn? p85

p110

PI-3 kinase PMA

Akt P Thr P Ser

Cell cycle progression Inhibition of apoptosis Figure 13.7 Schematic of the putative mechanism of Qa-2 signaling. After antibody-mediated cross-linking of two Qa-2 protein molecules, Fyn may be activated, possibly via a transmembrane protein partner that interacts with Qa-2 on the cell exterior and with Fyn inside the cell. Fyn can directly interact with the regulatory domain (p85) of phosphatidylinositol-3 (PI-3) kinase. PI-3 kinase catalyzes the formation of phosphatidylinositol 3,4,5 triphosphate (PIP3), which results in phosphorylation (P) of a specific threonine residue (Thr) in Akt. Phorbol myristic acetate (PMA) completes the activation of Akt by inducing phosphorylation of a specific serine residue (Ser). Once activated, Akt promotes a number of events that support cell proliferation and inhibit apoptosis.

that a range of developmental rates would be advantageous for species with litters, so that at least some embryos could implant. In humans the same argument can be made for successive reproductive cycles. RECENT FINDINGS USING A MOUSE MODEL TO STUDY THE PED GENE

Animal models provide an excellent opportunity for studying the immunological aspects of early pregnancy. Many experiments can be carried out in animal embryos that cannot be performed on human embryos. As mentioned previously, of the many possible animal models, the mouse is particularly valuable because of the availability of inbred, congenic, transgenic, and knockout strains. Preimplantation mouse embryos can be cultured in chemically defined media, and research can thus be performed that is directly comparable with human embryos that are grown in vitro in the ART clinic. In the mouse, the

congenic strains B6.K1 and B6.K2 have been a valuable resource for studies on the Ped gene.69,70 These two strains differ genetically only in that B6.K1 lacks the Qa-2 encoding genes, and these are present in B6.K2. We recently tested whether the sex of preimplantation embryos is a confounding factor in mediating the rate of preimplantation development of B6.K1 and B6.K2 embryos, and found that sex has no effect on the rate of development of embryos from both B6.K1 and B6.K2 strains of mice.72 Therefore, Qa-2 protein, and only Qa-2 protein (the Ped gene product), is responsible for regulating the rate of preimplantation development in the B6.K1/B6.K2 mouse model. Another interesting recent finding has an impact on the mechanism that enables B6.K1 embryos to survive without the presence of Qa-2 protein. We cultured embryos from B6.K1 and B6.K2 mice from the 2-cell to the blastocyst stage, and measured the amount of platelet activating factor (PAF) in the 

HPE_Chapter13.qxp

7/13/2007

4:52 PM

Page 164


culture supernatant. PAF is known to have a role in many reproductive events, including ovulation, fertilization, implantation, and parturition. We found that B6.K1 embryos produce twice as much PAF as B6.K2 embryos.73 The conclusion is that B6.K1 preimplantation embryos may enhance their chance of survival in order to compensate for the lack of Qa-2 expression by increasing PAF production. It seems logical to hypothesize that embryonic expression of molecules other than PAF may also be affected by the presence or absence of Qa-2 protein. Such molecules may be proteins, but may also be other molecules. Studies to identify these other putative molecules are underway in my laboratory, using the B6.K1/B6.K2 mouse model system. It remains to be determined which, if any, of these hypothesized molecules are involved in mediating embryo survival and protecting the embryo from immune destruction by the mother. CLINICALLY RELEVANT PROPERTIES OF THE MOUSE PED GENE

Our basic research on the mouse Ped gene is ripe for transition from laboratory to clinic. A large body of literature on the properties of the Ped gene and its protein product, Qa-2 in the mouse and HLA-G in humans, has recently been reviewed.5,57,69,70 The original definition of the Ped gene phenotype was its relation to rapid or slow development during the preimplantation period. Thus, it is likely that selection of rapidly developing embryos in the ART clinic will select for the ‘fast’ allele of the human Ped gene, as well as for other as yet unidentified genes that affect the rate of preimplantation development. As shown in Table 13.1, there are many parameters other than rate of preimplantation development that are affected by the presence of Qa-2 protein in mouse embryos. In Table 13.1, clinically relevant parameters have been separated into the effects of Qa-2 protein on preimplantation embryos and effects of Qa-2 protein later in life. Some of the data in Table 13.1 are based on our studies in the B6.K1/B6.K2 mouse model system described above. It is fascinating to note that the presence of the Ped gene product, Qa-2 protein, has effects beyond 

Table 13.1 Clinically relevant properties of the mouse Ped gene product, Qa-2 protein. Data are based on the B6.K1 (Qa-2 negative)/B6.K2 (Qa-2 positive) mouse model system Effect

Reference

Preimplantation effects of the presence of Qa-2 protein Earlier ovulation Comiskey and Warner, unpublished Earlier 1st cleavage Comiskey and Warner, unpublished Faster rate of development Reviewed in 69 and 70 Lower levels of platelet 73 activating factor Earlier implantation Comiskey and Warner, unpublished Later effects of the presence of Qa-2 protein Enhanced survival to birth Reviewed in 69 and 70 Higher birth weight Reviewed in 69 and 70 Enhanced survival to weaning Reviewed in 69 and 70 Higher weaning weight Reviewed in 69 and 70 Lower adult blood pressure 75 Lower adult levels of 75 angiotensin converting enzyme

the preimplantation period into adult life. These results are consistent with the ‘Barker Hypothesis’,74 which states that growth of the embryo during the preimplantation period affects health during adulthood, including effects on body weight, heart disease, and blood pressure. Our recent study75 has indeed shown that both blood pressure and levels of angiotensin converting enzyme (ACE) are lower in B6.K2 (Qa-2 positive) mice compared with B6.K1 (Qa-2 negative) mice. It has been speculated that perhaps the lifespan of B6.K2 mice is shorter than the lifespan of B6.K1 mice, but this has not been tested experimentally.76 However, this speculation is at odds with the Barker Hypothesis, so the potential outcome of a longevity study on the B6.K1 and B6.K2 mice is not clear. The overall conclusion is that selection of fast-developing embryos for transfer to the mother after ART may have health implications that are far beyond the preimplantation period of development.

CONCLUSIONS AND FUTURE DIRECTIONS

I now return to the quote by Medawar at the beginning of this article, in which he wonders why the

HPE_Chapter13.qxp

7/13/2007

4:52 PM

Page 165


allogeneic embryo is not rejected by the maternal immune system. In this chapter I have shown that not only do preimplantation embryos express allogeneic MHC proteins, but that the expression of at least one of them, Qa-2 in the mouse and its homolog, HLA-G in humans, enhances pregnancy outcome. An interesting recent article77 suggests that HLA-G protein must be expressed at just the right level to enhance reproduction, too much HLA-G or too little HLA-G are detrimental to reproductive success. The validity of this hypothesis awaits testing in a clinical setting. It also remains to be determined if all human embryos that express cell surface HLA-G also produce soluble HLA-G. It has been shown that embryos that express cell surface HLA-G cleave faster than embryos that lack cell surface HLA-G,44 but whether the rate of development is correlated with the amount of soluble HLA-G found in the supernatant of the cultured embryos has not yet been rigorously tested. The proteins encoded by the MHC are only one of many classes of proteins involved in the immune response. It has been estimated that perhaps 10% of the 20 000–25 000 human genes are involved in immune function. Therefore, the challenge is not only to identify these proteins on the surface of the embryo and in culture supernatants from preimplantation embryos, but also to ascertain the amount of protein that is indicative of a healthy embryo. In addition, diagnostic methods could also be applied to the parents donating eggs and sperm in the ART clinic. For example, it has been shown that there is an association between the mother’s HLA-G genotype and success of IVF and pregnancy outcome.78 In the future it may be possible to determine the genotype of both parents and to predict which couples will have the greatest chance of a successful outcome after IVF. This approach should apply to all genes and not only to those involved in the immune system. SINGLE EMBRYO TRANSFER

In the United States about 1% of all births result from ART procedures, with the percentage reaching 2–3% in many European countries and in Australia. At the present time there is a worldwide effort towards

single embryo transfer in order to reduce the frequency of multiple births, which are detrimental to both the mothers and the babies. In the United States the multiple birth rate is 35% after ART compared with 1% after natural conception. A recent paper79 presents data on the first prospective study in which single embryo transfer on day 3 (cleavage stage embryos) is compared with embryo transfer on day 5 (blastocyst stage embryos): the delivery rate was 22% for day 3 transfer, and 32% for the day 5 transfer. The embryos in this study were all produced by women under 36 years of age, thereby excluding half of the women who visit the ART clinic. Clearly future prospective studies need to be performed on women in the older age group. The fact that even the best procedure gave only a 32% success rate emphasizes the issue of identifying which embryos should be transferred to the mother. In this study, only morphological methods (DIC microscopy) were used for embryo evaluation on both day 3 and day 5, and the number of blastomeres (cells) and the degree of fragmentation were evaluated. Day 5 embryos (blastocysts) were also evaluated on the basis of size of the inner cell mass (ICM), and the degree of expansion of the blastocoel cavity. It seems clear that if additional data were available, such as the nature of the molecules secreted by these embryos, the methods by which embryos could be selected for single embryo transfer could be greatly enhanced. BETTER REPRODUCTION THROUGH CHEMISTRY

How can we identify the proteins on the surface of preimplantation embryos and those secreted into the culture medium? After identification, how can we quantitate the levels of these proteins and correlate the levels with embryo health? How can we determine the genotype of the embryos and their parents? Clearly new methodologies for genomics and proteomics are needed. There is growing interest among chemists in the application of nanotechnologies to biomolecular detection and medical diagnostics.80 The long range goal is to provide personalized medicine based on the screening of biological fluids such as blood and urine. It seems 

HPE_Chapter13.qxp

7/13/2007

4:52 PM

Page 166


perfectly reasonable to suggest that these same nanotechnologies can and will be applied to preimplantation embryos and culture supernatants in the ART clinic. For instance, high-throughput multiantigen microfluidic fluorescence immunoassays that can detect 100 proteins simultaneously are under development.81 This amount of information should give us a much better chance of evaluating embryo health, compared with evaluating a single protein at a time, such as HLA-G. The major challenge is to have access to a sufficient number of embryos and sufficient funding to conduct this research. BETTER REPRODUCTION THROUGH POLITICS

At the present time there are hundreds of thousands of human embryos left over from ART procedures stored in liquid nitrogen tanks. In order to conduct carefully controlled hypothesis driven research that will help to define the optimal levels of immunerelated and other molecules that constitute a healthy embryo, access to these embryos and money to support the research is needed. Unfortunately, there is a crisis in the United States with respect to the ability to conduct research on these embryos, since it is illegal to use federal funds for this endeavor. This has been pointed out elegantly by Chris Mooney in his recent book The Republican War on Science.82 A review of this book by Paul Berg, Nobel Laureate in Chemistry, states that ‘If left unchallenged, the Bush administration’s deliberate misrepresentation and frequent outright disregard of science advisory processes will have serious consequences for the nation’s economy, health and security. Chis Mooney has opened a window to reveal the extent of the antiscience bias in government policy making.’ Another review, by James McCarthy, the Director of the Museum of Comparative Zoology, Harvard University, states that ‘Chris Mooney’s thorough research into recent political efforts to distort and suppress scientifically grounded knowledge demonstrate that our nation is veering sharply from the path that enlightened men like Thomas Jefferson and Benjamin Franklin envisioned for us.’ Not only is the present antiscience attitude of the leadership in the United States hampering better 

ways to evaluate embryos in the ART clinic, but it is also impeding research on new contraceptive methods for an over-populated world and research on the very promising field of embryonic stem (ES) cells, both of which rely on the availability of federal funding for human embryo research. My final challenge is to the next generation of researchers to work toward the implementation of a rational, science oriented government that fosters, rather than blocks human embryo research. ACKNOWLEDGMENTS

Funding from the NIH (HD39215) and the NSF (EEC-9986821) is gratefully acknowledged. My sincere thanks go to the members of my research group, Michael Byrne, Martina Comiskey, Sally De Fazio, Carmit Goldstein, Paula Lampton, and Judith Newmark, and to Julian Fleischman for the critical reading of this manuscript. I especially thank Judith Newmark and Sally De Fazio for the preparation of the figures. I am indebted to Paula Leventman for alerting me to the ‘must read’ book, The Republican War on Science. REFERENCES 1. Medwar PB. Some immunological and endocrinological problems raised by evolution of viviparity in vertebrates. Symp Soc Exp Biol 1953; 7: 320–38. 2. Goldsby RA, Kindt TJ, Osborne BA et al. Kuby Immunology, 6th edn. New York: WH Freeman, 2007: 574. 3. Janeway CA, Travers P, Walport M et al. Immunobiology, 6th edn. New York: Garland Science, 2005: 823. 4. Paul WE. Fundamental Immunology, 5th edn. Philadelphia: Lippincott Williams & Wilkins, 2003: 1701. 5. Comiskey M, Warner CM, Schultz DJ. MHC molecules of the preimplantation embryo and trophoblast. In: Mor G, ed. Immunology of Pregnancy. New York: Landes Bioscience, 2006: 67–84. 6. Veeck LL, Zaninovic´ N. An Atlas of Human Blastocysts. New York: Parthenon Publishing, 2003: 286. 7. Warner CM, Brownell MS, Ewoldsen MA. Why aren’t embryos immunologically rejected by their mothers? Biol Reprod 1988; 38: 17–29. 8. Wasserman PM. Contribution of mouse egg zona pellucida glycoproteins to gamete recognition during fertilization. J Cell Physiol 2005; 204: 388–91. 9. Wasserman PM, Jovine L, Qi H et al. Recent aspects of mammalian fertilization research. Mol Cell Endocrinol 2005; 234: 95–103. 10. Clark GF, Dell A. Molecular models for murine sperm-egg binding. J Biol Chem 2006; 281: 13853–6. 11. Lefièvre L, Conner SJ, Salpekar A et al. Four zona pellucida glycoproteins are expressed in the human. Hum Reprod 2004; 19: 1580–6.

HPE_Chapter13.qxp

7/13/2007

4:52 PM

Page 167


12. Conner SJ, Lefièvre L, Hughes DC et al. Cracking the egg: increased complexity in the zona pellucida. Hum Reprod 2005; 20: 1148–52. 13. Naz RK, Gupta SK, Gupta JC et al. Recent advances in contraceptive vaccine development: a mini-review. Hum Reprod 2005; 20: 3271–83. 14. Koyama K, Hasegawa A, Mochida N et al. Follicular dysfunction induced by autoimmunity to the zona pellucida. Reprod Biol 2005; 5: 269–78. 15. Nottola SA, Makabe S, Stallone T et al. Surface morphology of the zona pellucida surrounding human blastocysts obtained after in vitro fertilization. Arch Histol Cytol 2005; 68: 133–41. 16. Cohen J, Inge KL, Suzman M et al. Videocinematography of fresh and cryopreserved embryos: a retrospective analysis of embryonic morphology and implantation. Fertil Steril 1989; 52: 877–8. 17. Sun YP, Xu Y, Cao T et al. Zona pellucida thickness and clinical pregnancy outcome following in vitro fertilization. Int J Gynaecol Obstet 2005; 89: 258–62. 18. Manniko M, Mormala RM, Tuuri T et al. Association between sequence variations in genes encoding human zona pellucida glycoproteins and fertilization failure in IVF. Hum Reprod 2005; 20: 1578–85. 19. Seif M, Edi-Osagie E, Farquhar C et al. Assisted hatching on assisted conception (IVF & ICSI). Cochrane Database Syst Rev 2006; 1: 1–68. 20. Ewoldsen MA, Ostlie NS, Warner CM. Killing of mouse blastocyst stage embryos by cytotoxic T lymphocytes directed to major histocompatibility complex antigens. J Immunol 1987; 138: 2764–70. 21. Cohen J, Malter H, Elsner C et al. Immunosuppression supports implantation of zona pellucida dissected human embryos. Fertil Steril 1990; 53: 662–5. 22. Goldbard SB, Gollnick SO, Warner CM. A highly sensitive method for the detection of cell surface antigens on preimplantation mouse embryos. J Immunol Meth 1984; 68: 137–46. 23. McElhinny AS, Warner CM. Detection of major histocompatibility complex class I antigens on the surface of a single murine blastocyst by Immuno-PCR. BioTechniques 1997; 23: 660–2. 24. Niemeyer CM, Adler M, Wacker R. Immuno-PCR: high sensitivity detection of proteins by nucleic acid amplification. Trends Biotech 2005; 23: 208–16. 25. Sharov AA, Piao Y, Matoba R et al. Transcriptome analysis of mouse stem cells and early embryos. PLos Biol 2003; 1: 410–19. 26. Carter MG, Piao Y, Dudekula DB et al. The NIA cDNA project in mouse stem cells and early embryos. C R Biol 2003; 326: 931–40. 27. Hamatani T, Carter MG, Sharov AA et al. Dynamics of global gene expression changes during mouse preimplantation development. Dev Cell 2004; 6: 117–31. 28. Ko MS. Embryogenomics of pre-implantation mammalian development: current status. Reprod Fertil Dev 2004; 16: 79–85. 29. Ko MS. Molecular biology of preimplantation embryos: primer for philosophical discussions. Reprod Biomed Online 2005; 10(Suppl 1): 80–87. 30. Margulies DH, McCluskey J. The major histocompatibility complex and its encoded proteins. In: Paul WE, ed. Fundamental Immunology 5th edn. Philadelphia: Lippincott Williams & Wilkins, 2003: 571–612. 31. Comiskey M, Goldstein CY, De Fazio SR et al. Evidence that HLA-G is the functional homolog of mouse Qa-2, the Ped gene product. Hum Immunol 2003; 64: 999–1004. 32. Clements CS, Kjer-Nielsen L, Kostenko L et al. Crystal structure of HLA-G: a nonclassical MHC class I molecule expressed at the fetalmaternal interface. Proc Natl Acad Sci USA 2005; 102: 3360–5. 33. Searle RF, Sellens MH, Elson J et al. Detection of alloantigens during preimplantation development and early trophoblast differentiation. J Exp Med 1976; 143: 348–59. 34. Warner CM, Spannaus DJ. Demonstration of H-2 antigens on preimplantation mouse embryos using conventional antisera and monoclonal antibody. J Exp Zool 1984; 230: 37–52.

35. Webb CG, Gall WE, Edelman GM. Synthesis and distribution of H-2 antigens in preimplantation mouse embryos. J Exp Med 1977; 146: 923–31. 36. Krco CJ, Goldberg EM. Major histocompatibility antigens on preimplantation mouse embryos. Trans Proc 1977; IX: 1367–70. 37. Cozad KM, Warner CM. Detection of H-2 antigens on 8-cell mouse embryos. J Exp Zool 1982; 221: 213–17. 38. Tian Z, Xu Y, Warner CM. Removal of Qa-2 antigen alters the Ped gene phenotype of preimplantation mouse embryos. Biol Reprod 1992; 47: 271–6. 39. Warner CM, Almquist CD, Toulimat MH et al. Induction of embryonic major histocompatibility complex antigen expression by ␥-IFN. J Reprod Immunol 1993; 24: 111–21. 40. Warner CM, Gollnick SO et al. Expression of H-2K major histocompatibility antigens on preimplantation mouse embryos. Bio Reprod 1993; 48: 1082–7. 41. McElhinny AS, Kadow N, Warner CM. The expression pattern of the Qa-2 antigen in mouse preimplantation embryos and its correlation with the Ped gene phenotype. Mol Hum Reprod 1998; 4: 966–71. 42. Desoye G, Dohr GA, Motter W et al. Lack of HLA class I and class II antigens on human preimplantation embryos. J Immunol 1988; 140: 4157–9. 43. Roberts JM, Taylor CT, Melling GC et al. Expression of the CD46 antigen, and absence of class I MHC antigen, on the human oocyte and preimplantation blastocyst. Immunology 1992; 75: 202–5. 44. Manyonda IT. The Immunology of Human Reproduction. London: Taylor & Francis, 2006: 26–7. 45. Jurisicova A, Casper RF, MacLusky et al. HLA-G expression during preimplantation human embryo development. Proc Natl Acad Sci USA 1996; 93: 161–5. 46. Drukker M, Katz G, Urbach A et al. Characterization of the expression of MHC proteins in human embryonic stem cells. Proc Natl Acad Sci USA 2002; 99: 9864–9. 47. Fuzzi B, Rizzo R, Criscuoli L et al. HLA-G expression in early embryos is a fundamental prerequisite for the obtainment of pregnancy. Eur J Immunol 2002; 32: 311–15. 48. Sher G, Keskintepe L, Nouriani M et al. Expression of sHLA-G in supernatants of individually cultured 46-h embryos: a potentially valuable indicator of ‘embryo competency’ and IVF outcome. Reprod Biomed Online 2004; 9: 74–8. 49. Noci I, Fuzzi B, Rizzo R et al. Embryonic soluble HLA-G as a marker of developmental potential in embryos. Hum Reprod 2005; 20: 138–46. 50. Sher G, Kestintepe L, Batzofin J et al. Influence of early ICSI-derived embryo sHLA-G expression on pregnancy and implantation rates: a prospective study. Hum Reprod 2005; 20: 1359–63. 51. Sher G, Kestintepe L, Fisch JD et al. Soluble human leukocyte antigen G expression in phase I culture media at 46 hours after fertilization predicts pregnancy and implantation from day 3 embryo transfer. Fertil Steril 2005; 83: 1410–13. 52. Yie SM, Balakier H, Motamedi G et al. Secretion of human leukocyte antigen-G by human embryos is associated with a higher in vitro fertilization pregnancy rate. Fertil Steril 2005; 83: 30–6. 53. Van Lierop MJ, Wijnands F, Loke YW et al. Detection of HLA-G by a specific sandwich ELISA using monoclonal antibodies G233 and 56B. Mol Hum Reprod 2002; 8: 776–84. 54. Noriko S, Horotsugu H, Masanori Y et al. Are in vitro fertilized eggs able to secrete soluble HLA-G? Am J Reprod Immunol 2004; 52 (Suppl 1): P8. 55. Ivell R. Comment from the Editor-in-chief. Mol Hum Reprod 2005; 11: 693. 56. Sargent IL. Does ‘soluble’ HLA-G really exist? Another twist to the tale. Mol Hum Reprod 2005; 11: 695–8.



HPE_Chapter13.qxp

7/13/2007

4:52 PM

Page 168


57. Hviid TV. HLA-G in human reproduction: aspects of genetics, function and pregnancy complications. Hum Reprod Update 2006; 12: 209–32. 58. Geissler EK, Korzun WJ, Graeb C. Secreted donor-MHC class I antigen prolongs liver allograft survival and inhibits recipient antidonor cytotoxic T lymphocyte responses. Transplantation 1997; 64: 782–6. 59. Scherer MN, Graeb C, Tange S et al. Immunologic consideration for therapeutic strategies utilizing allogeneic hepatocytes: hepatocyteexpressed membrane-bound major histocompatibility complex class I sensitizes while soluble antigen suppresses the immune response in rats. Hepatology 2000; 32: 999–1007. 60. Fournel SM, Aguerre-Girr X, Huc F et al. Soluble HLA-G1 triggers CD95/CD95 ligand-mediated apoptosis in activated CD8⫹ cells by interaction with CD8. J Immunol 2000; 164: 6100–4. 61. Contini P, Ghio M, Merlo A et al. Apoptosis of antigen-specific T lymphocytes upon engagement of CD8 by soluble HLA class I molecules is Fas ligand/Fas mediated: Evidence for the involvement of p56lck, calcium calmoldulin kinase II, and calcium-independent protein kinase C signaling pathways for NF-␬B and NF-AT nuclear translocation. J Immunol 2005; 175: 7244–54. 62. Rouas-Freiss N, Marchal RE, Kirzenbaum M et al. The alpha 1 domain of HLA-G1 and HLA-G2 inhibits cytotoxicity induced by natural killer cells: is HLA-G the public ligand for natural killer cell inhibitory receptors? Proc Natl Acad Sci USA 1997; 94: 5249–54. 63. Riteau B, Menier C, Khalil-Daher I et al. HLA-G1 co-expression boosts the HLA class I-mediated NK lysis inhibition. Int Immunol 2001; 13: 193–201. 64. Riteau B, Rouas-Freiss N, Menier C et al. HLA-G2-G3 and -G4 isoforms expressed as nonmature cell surface glycoproteins inhibit NK and antigen-specific CTL cytolysis. J Immunol 2001; 166: 5018–26. 65. Park GM, Lee S, Park B et al. Soluble HLA-G generated by proteolytic shedding inhibits NK-mediated cell lysis. Biochem Biophys Res Commun 2004; 313: 606–11. 66. LeMaoult J, Rouas-Freiss N, Carosella ED. Immuno-tolerogenic functions of HLA-G: relevance in transplantation and oncology. Autoimmun Rev 2005; 4: 503–9. 67. Wharf E, Dimitrakopoulos A, Khalaf Y et al. Early embryo development is an indicator of implantation potential. Reprod Biomed Online 2004; 8: 212–18.



68. Verbanac KM, Warner CM. Role of the major histocompatibility complex in the timing of early mammalian development. In: Glasser SR, Bullock DW, eds. Cellular and Molecular Aspects of Implantation. New York: Plenum Publishers, 1981: 467–70. 69. Warner CM, Brenner CA. Genetic regulation of preimplantation embryo survival. In: Schatten GP, ed. Current Topics in Developmental Biology. San Diego: Academic Press, 2001; 52: 151–92. 70. Warner CM, Newmark JA, Comiskey M et al. Genetics and imaging to assess oocyte and preimplantation embryo health. Reprod Fertil Dev 2004; 16: 729–41. 71. Ober C, Aldrich C, Rosinsky B et al. HLA-G1 protein expression is not essential for fetal survival. Placenta 1998; 19: 127–32. 72. Byrne MJ, Newmark JA, Warner CM. Analysis of the sex ratio in preimplantation embryos from B6.K1 and B6.K2 Ped gene congenic mice. J Assist Reprod Genet 2006; 23: 321–8. 73. Purnell ET, Warner CM, Kort HI et al. Influence of the preimplantation embryo development (Ped) gene on embryonic platelet-activating factor (PAF) levels. J Assist Reprod Genet 2006; 23: 269–73. 74. Barker DJ. The developmental origins of chronic adult disease. Acta Paediatr Suppl 2004; 93: 26–33. 75. Watkins A, Wilkins A, Osmond C et al. The influence of mouse Ped gene expression on postnatal development. J Physiol 2006; 571: 211–20. 76. Tarin JJ. Do the fastest concepti have a shorter life span? Hum Reprod 1997; 12: 885–9. 77. Ober C, Billstrand C, Kuldanek S et al. The miscarriage-associated HLA-G -725G allele influences transcription rates in JEG-3 cells. Hum Reprod 2006; 21: 1743–8. 78. Hviid TVF, Hylenius S, Lindhard A et al. Association between human leukocyte antigen-G genotype and success of in vitro fertilization and pregnancy outcome. Tissue Antigens 2004; 64: 66–9. 79. Papanikolaou EG, Camus M, Kolibianakis EM et al. In vitro fertilization with single blastocyst-stage versus single cleavage-stage embryos. N Engl J Med 2006; 354: 1139–46. 80. Cheng M, Cuda G, Bunimovich YL et al. Nanotechnologies for biomolecular detection and medical diagnostics. Curr Opin Chem Biol 2006; 10: 11–19. 81. Kartalov EP, Zhong JF, Scherer A et al. High-throughput multi-antigen microfluidic fluorescence immunoassays. BioTechniques 2006; 40: 85–90. 82. Mooney C. The Republican War on Science. Cambridge, MA: Basic Books, 2005: 342.

HPE_Chapter14.qxp

7/13/2007

4:52 PM

Page 169

14. Nitric oxide regulation of the preimplantation embryo Yvette M Huet-Hudson

INTRODUCTION

Although we have a basic understanding of the biochemical and morphological changes that occur during mammalian preimplantation embryo development, from oocyte fertilization by spermatozoa to the zygote or 1-cell embryo, through early cleavage stages, compaction of the morula, and culminating in the formation of the blastocyst, large gaps in our knowledge remain. A great deal of literature has been published about nitric oxide (NO) (evidenced by over 76 000 papers produced by a Medline search), but much less is understood about its role in embryo development (only 94 papers). The current literature regarding its role in preimplantation embryo development has not been reviewed for several years.1 This chapter therefore reviews the role of NO as a regulator of cell cycle progression in embryo development from the 1-cell to the blastocyst stage, based upon literature published from 19982 through to the most recent advances in 2006.3 NITRIC OXIDE

NO is a gaseous free radical molecule that has been shown to be an intracellular messenger. NO is produced in a variety of tissue types and plays many important roles, including smooth muscle relaxation, vasodilation, neuronal signaling, and stimulation of the immune response, as well as regulating preimplantation embryo development.4–6 It has also been shown to regulate other reproductive processes including steroidogenesis, folliculogenesis, tissue remodeling, and angiogenesis, cervical ripening, and uterine contractility during pregnancy.7

NO freely diffuses across the cell membrane, and has a biological half-life of only a few seconds.8 Although it readily reacts with superoxide to form peroxynitrate anions, it is otherwise a relatively non-reactive molecule, and can thus diffuse intact into neighboring cells to alter target molecules and cellular responses.9 NO is produced from the conversion of L-arginine to L-citrulline by the enzyme nitric oxide synthase (NOS). L-arginine is the only available physiological nitrogen donor for NOS-catalyzed reactions, and therefore the availability of the amino acid L-arginine can determine rates of NO production.7 There are three isoforms of the NOS enzyme, neuronal (nNOS, bNOS, NOS1), endothelial (eNOS, NOS3), and inflammatory (iNOS, NOS2). These isoforms of NOS are products of three separate genes found on separate chromosomes, but they share 50–60% amino acid homology with one another.10 All isoforms are thought to function as homodimers with a carboxy-terminal reductase domain and an amino-terminal oxygenase domain linked in the middle by a calmodulin-binding domain.11 The amino-terminal oxygenase domain has binding sites for L-arginine, zinc, tetrahydrobiopterin (BH4) and heme, while the carboxyterminal reductase domain contains flavin adenine dinucleotide (FAD) and NADPH binding sites.12 There are clearly differences between the NOS isoforms, and typically they are subdivided based on calcium–calmodulin dependency. Endothelial (e)NOS and neuronal (n)NOS are thought to be produced constitutively, and are regulated via calcium–calmodulin binding; therefore their activation requires increases in intracellular calcium. The increase in intracellular calcium allows increased binding of calcium to calmodulin, forming 

HPE_Chapter14.qxp

7/13/2007

4:52 PM

Page 170


the calcium–calmodulin complex. This calcium– calmodulin complex then allows electron transfer from NADPH to one of the heme-containing active sites in the NOS homodimer. This facilitates the oxidation reaction converting oxygen and L-arginine into NO and L-citrulline.4,6 Although these isoforms are regulated by intracellular calcium levels and are constitutively expressed in many cells, it is also clear that their expression can be further regulated by other factors. One well-known example is the changes in activity of both eNOS and nNOS produced by phosphorylation. However, the result after phosphorylation of these two isoforms is not the same. Increased phosphorylation of eNOS results in an increase in calcium-independent NO production, whereas phosphorylation of nNOS leads to a decreased activity.13–17 There are also other examples of NOS activity regulated by circulating molecules, e.g. in the ovariectomized progesterone-treated mouse uterus, mRNA expression of eNOS is upregulated 2-fold after 1 hour of estrogen exposure (unpublished data). Regulation of eNOS expression in cultured endothelial cells is also demonstrated by several factors, including low concentrations of oxidized low density lipoprotein (LDL),18,19 lysophophatidylcholine,20 and tumor necrosis factor (TNF)- exposure.21 In addition, proliferating endothelial cells were shown to have 4–6 fold increased levels of eNOS mRNA during culture, relative to confluent cells.22 Inducible (i)NOS is structurally similar to the other NOS isoforms, but it is not calcium–calmodulin regulated, and is thus termed constitutively active. iNOS has calmodulin bound to each of its subunits, and does not require an increase in intracellular calcium for activation.12 Therefore, alterations in NO production by iNOS occur through alterations in the production of the enzyme. In macrophages, iNOS produces NO in large quantities, and thus is considered to have a more generalized action than does the production of NO from either eNOS or nNOS. These isoforms produce NO in small quantities, and their actions are considered to be more localized. In order to understand the role that NO plays in the regulation of mammalian development, we must elucidate its function in the 

regulation of the cell cycle as well as how it regulates preimplantation embryo development.

CELL CYCLE REGULATION

The transition from maternal control of development to zygotic control is an important event in the regulation of embryonic development. Initially, the oocyte contains proteins and maternal RNAs that are necessary to maintain and regulate oocyte function and the subsequent early development of the embryo.23–28 However, these stores have a limited lifespan and are degraded, so that embryonic genes must be turned on relatively early in embryonic life to take over control of development. The embryonic genome is activated in mammals at different embryonic stages (Table 14.1), but regardless of the stage at which the transition from maternal to embryonic control of gene expression occurs, continued development will not proceed without newly transcribed embryonic mRNAs. Once the embryonic genome is activated, maternal transcripts and protein can be degraded over time. Thus, normal embryonic development begins with the use of stored maternal components, which regulate the first mitotic division(s), and subsequent cellular divisions are influenced by embryo-derived factors. Eukaryotic cell division is highly regulated, and proceeds through an ordered set of events. This progression through the cell cycle requires intrinsic mechanisms that act in a cascade such that one cell cycle event depends on another. Progression through the cell cycle is mediated by the activation of a

Table 14.1 Timing of maternal to embryonic regulation of gene expression Species

Maternal to embryonic transition

Reference

Mouse Rat Cow Pig Rabbit Human

2-cell stage Late 2-cell stage 2-cell stage 4-cell stage 8-cell stage 4–8-cell stage

25 28 27 26 24 23

HPE_Chapter14.qxp

7/13/2007

4:52 PM

Page 171

NITRIC OXIDE REGULATION OF THE PREIMPLANTATION EMBRYO

highly conserved family of protein kinases, the cyclin-dependent kinases (cdks).29 Activation of cdks requires binding to positive regulatory subunits called cyclins.30,31 Cyclin protein levels fluctuate through the cell cycle, whereas the levels of cdks are relatively constant. Thus, regulated synthesis and destruction of each cyclin ultimately controls the activity of the cyclin–cdk complex.31 Working as a unit, the cyclin–cdk complexes are the regulators of progression through the cell cycle, with each specific cyclin–cdk complex controlling a specific transition between the subsequent phases in the cell cycle. Following association of a cyclin with its cdk, the cyclin–cdk complex is regulated through phosphorylation events that result in either activation or inhibition of kinase activity.31 Specific cyclin–cdk complexes are required at different stages of the cell cycle. For example, cyclin D1–cdk4 is upregulated by extracellular stimuli such as growth factors, and this allows quiescent cells to enter the cell cycle. The cyclin E–cdk2 complex regulates the transition from G1 to S phase and also progression through the S phase, allowing synthesis of cellular DNA.31,32 The transition from G2 to M, i.e. entry into mitosis phase, is under the control of cyclin B–cdc2 activation.29,31,33,34 It is important to remember that there are several points in the cell cycle at which progression can be stopped, in the event that DNA damage has occurred. These are called checkpoints, and are of particular importance in preventing DNA damage from being passed to the next phase of the cell cycle and eventually to the daughter cell. When DNA damage occurs, mammalian cells can arrest in the G1, S, or G2 phase of the cycle. The timing of arrest is dependent on the phase in which the damage is recognized. However, once a checkpoint has been activated, the cell will either repair the damage and continue progression through the cell cycle, or undergo apoptosis if the damage cannot be repaired.31,35

PREIMPLANTATION EMBRYO DEVELOPMENT

Preimplantation embryo development begins with the 1-cell embryo in the fallopian tube or oviduct.36

As the embryo divides, it progresses to the 2-cell stage without an increase in overall cellular mass, a process that takes approximately 18–20 hours, and it then continues to travel along the fallopian tube/ oviduct. Subsequent cellular divisions are faster, taking approximately 12 hours each. When the embryo reaches the morula stage, it is characterized as a ball of 8–16 cells. During the morula stage the embryo undergoes compaction, a calciummediated process. Once compaction occurs, the blastocoel cavity may begin to form, marking the beginning of the blastocyst stage, with blastomere differentiation for the first time. At the blastocyst stage the embryo is made up of 16–64 cells. During the morula to blastocyst transition, the embryo exits the fallopian tube/oviduct and enters the uterus, and it is at this stage of development that interactions between the uterus and the embryo initiate the process of implantation. This well-defined pattern of development in the preimplantation embryo occurs both in vivo and in vitro. Preimplantation embryos cultured in a simple medium can develop to the blastocyst stage and successfully hatch in vitro.37,38 However, various studies have shown that in vitro environmental conditions can alter the rate and successful development of embryos, and therefore the effect that molecules derived from the embryo, uterus, and environment have on the normal development of embryos is an area of great interest. The first studies carried out to investigate the role of NO in preimplantation embryo development were preceded by reports about the role of NO in rodent embryo implantation.39 It was thus already clear that NO does have a role in the uterus, but its role in embryo development was unknown. Early experiments in the mouse demonstrated that NO is produced by embryos in culture, and that its production could be decreased by known NO inhibitors.2 The data demonstrate that NO is produced by 1-cell mouse embryos through to blastocyst stage at fairly consistent levels, and that this production can be completely inhibited, altering development. In a mouse model of delayed implantation, the ovaries are removed from the mouse prior to the nidatory estrogen surge and the mouse 

HPE_Chapter14.qxp

7/13/2007

4:52 PM

Page 172


is then maintained in a state of delay by daily injection of progesterone. This model allows the effect of estrogen (which initiates implantation) on embryo production of NO to be determined. We observed that delayed blastocysts produced little NO when placed in culture after 24 hours. However, embryos recovered 1 hour after estrogen exposure in vivo and then cultured in vitro for 24 hours increased NO production over 800%. This indicates that the initiation of implantation is correlated with a large increase in embryo production of NO, prior to initial interactions with the maternal endometrium. It is not yet clear whether this increase acts in an autocrine manner on the embryonic cells, or whether it also has effects on the maternal endometrium. NO production can be inhibited by NG-nitro-Larginine (NLA), a non-convertible analog of arginine, and our laboratory demonstrated that culturing mouse embryos in media containing this inhibitor arrests development in over 74% of embryos.2 This arrest occurs only in embryos from the 2-cell stage onwards. NLA may prevent the embryos from developing beyond the 2-cell stage by triggering rapid enzyme inactivation of all three NOS isoforms and completely blocking NADPH-dependent heme reduction. This blocks both NO synthesis and superoxide (O2•) formation.10,40 Dissociation of the NLA from NOS is slow, allowing it to bind tightly to the NOS isoforms with prolonged inhibition of activity.10 This study was followed by several others using the NOS inhibitor N-omega-nitro-L-arginine methyl ester (L-NAME),40–42 all off which demonstrated inhibition of embryo development. Manser and Houghton3 also demonstrated that the effects of culture with NO inhibitors is dependent on the composition of the embryo culture media. However, these authors did not measure nitrite/nitrate production. Without this information, we must speculate that when embryos are cultured in media that does not contain arginine, the competitive NOS inhibitors result in little or no NO production, and embryos arrest prior to blastocyst formation. When the media does contain the amino acid arginine, the inhibitors for NOS may be unable to competitively inhibit NO production sufficiently to alter 

blastocyst formation significantly. However, it is very important to note that the number of cells in each blastocyst does decrease significantly, and therefore normal blastocyst development is nonetheless inhibited. It is clear that the normal progression of the cell cycle in mouse embryos can be delayed by a loss in the ability of NOS to produce NO. This delay can be reversed if NO production is resumed within less than 24 hours.2 However, prolonged inhibition of NO results in either a change in the developmental capacity of the embryo, or permanent arrest, depending on the culture environment. Normal embryo development beyond the 2-cell stage clearly requires a threshold level of NO. The studies performed by Manser and Houghton3 also demonstrate that culture in media containing different concentrations of amino acid results in altered oxygen consumption. The fact that arrest does not occur in the 1-cell mouse embryo but does occur in all other preimplantation stages, together with the fact that this change coincides with the transition from maternal to embryonic control of gene expression, suggests that NO regulates progression through the cell cycle through regulation of embryonic gene expression. Whether this alteration in gene expression is a direct effect of NO or a result of changes in oxygen consumption is yet to be determined. Excessive NO levels can also be induced by culturing embryos with the NO donor, sodium nitroprusside (SNP), and this also arrests development.40,41,43,44 This type of arrest differs from the arrest seen in embryos cultured with a NO inhibitor, in that the embryos are clearly, morphologically dead – this has been shown to be due to apoptosis.41,44 NO activates the p53 pathways and JNK/ SAPK, as well as the release of cytochrome c – all of which lead to apoptosis.45 In addition to NO production by the enzyme NOS, superoxide can also be produced. Superoxide and NO react to form peroxynitrite which can have multiple effects on the cell, including lipid peroxidation, DNA damage, and thiol nitrosylation.46 Some of these effects could activate apoptotic pathways. The exact mechanism by which SNP induces apoptosis is not clear, but due

HPE_Chapter14.qxp

7/13/2007

4:52 PM

Page 173


to the effect of NO on oxygen consumption and the ability of oxygen radicals to alter mitochondrial function, it is possible that mitochondrial damage and the release of cytochrome c may be part of the SNP-induced apoptosis pathway in embryos.

MODULATION OF NITRIC OXIDE SIGNAL TRANSDUCTION AND MITOSIS IN EMBRYOS

NO mediates its effects through a variety of pathways, including stimulation of cyclooxygenase enzymes, altering phosphodiesterases, and activating the inhibitory G subunit.40, 47, 48 Numerous studies have shown that NO also mediates many of its effects through the cyclic guanosine 3´,5´-monophosphate (cGMP) signal transduction system,47 by activating the enzyme soluble guanylyl cyclase, which then increases the production of cGMP. Guanylyl cyclase is a heterodimeric hemoprotein composed of and subunits. NO binds to the heme group, activating the enzyme and catalyzing the conversion of guanosine 5´-triphosphate (GTP) to cGMP. Increases in cGMP can alter the activity of cGMP-dependent protein kinase, cGMP-gated cation channels, and cGMP-regulated phosphodiesterases. It is well known that NO, via increased cGMP, decreases the proliferation of a variety of cells, including vascular smooth muscle cells,49 glial cells,50 T cells,51 brown adipocytes52 and a variety of tumor cells.53–55 However, it has also been shown that NO can have the opposite effect and induce proliferation.56–58 Clearly, the effect of NO on proliferation is cell-type dependent: in the case of embryos, its effect is promitotic when produced in the appropriate concentrations. NO could regulate mitosis either directly, or through alterations in cGMP levels; cGMP can regulate gene expression through both direct and indirect regulation of transcription factors. Transcription factors such as cAMP response-element binding protein (CREB) or activating transcription factor-1 (ATF-1) can be regulated via phosphorylation.59 cGMP can also alter the upstream pathways of transcription factors by inhibiting NF-B,60 inhibiting

serum response factor (SRF) through RhoA signaling,5 and in cardiomyocytes through inhibition of calcineurin signaling to the nuclear factor of activated T cells (NF/AT) transcription factor.61 Expression of other transcription factors is also directly affected by cGMP, including the early growth response gene, Egr-1, and c-fos and JunB, partners in the AP-1 transcription factor.62, 63 NO has been shown to modulate the expression of cyclins and the proteins that regulate their activity,64–67 but the exact effects of NO on the signaling cascades generated in the developing mouse embryo through to the blastocyst stage are unknown. Using real-time RT-PCR, we have determined that day 2 inhibited and culture control mouse embryos express cyclins A, B, and E. Statistical analysis shows that the expression of cyclins A and E was unchanged in the embryos cultured in NO inhibitor as compared with normally developing embryos. Cyclin B expression, however, was significantly down-regulated in the day 2 inhibited embryos, suggesting that NO may regulate the cell cycle during the transition from the G2 to the M stage. These data suggest that the main mechanism for NO regulation of mitotic division in the preimplantation embryo is regulation of cyclin B, which regulates the G2 to M transition that ultimately results in completed DNA replication. It must be noted, however, that these data are representative only of a 2-cell embryo. NO regulation of the cell cycle at different stages of embryo development may or may not be through cyclin B regulation, and this can only be determined by assessing data on each of the cyclins from each day of preimplantation embryo development. Checkpoint activity is another possible regulatory mechanism for NO. Checkpoint genes MAD (mitotic arrest deficient) and BUB (budding uninhibited by benomyl) code for the spindle checkpoint that regulates entry into anaphase.68 ATM (ataxia telangiectasia mutated) and ATR (ataxia telangiectasia and Rad 3 related) are DNA damage induced, and are involved in all checkpoints within the cell cycle with the exception of the spindle checkpoint.69 Using real-time RT-PCR, we have determined (unpublished data) that inhibition of NO had little or no effect on MAD, ATM, or BUB 

HPE_Chapter14.qxp

7/13/2007

4:52 PM

Page 174


expression on mouse embryos collected on days 2, 3, and 4. Decreased expression of cyclin B downregulates cdc2 kinase activity, and thus can bypass the ATM checkpoint.70 Based on our data described above on cyclin B, it is possible that NO may regulate the transition from the G2 to M phase via cyclin B downregulation, so that regulation of the checkpoint genes is not required. Days 2 and 3 control and inhibited embryos also showed no difference in ATR levels, but day 4 inhibited embryos showed a decrease in ATR expression at the blastocyst stage. This change in ATR expression may occur only in the blastocyst because at this stage the cells are differentiated and no longer totipotent, unlike those in earlier embryos. Thus, the mechanism by which NO regulates development in these embryos differs from earlier stages. However, the situation is complex, and further studies at all embryonic stages are required in

order to fully understand how NO regulates mitotic division.

NITRIC OXIDE SYNTHASE EXPRESSION IN EMBRYOS

Using real-time PCR, our laboratory has shown that eNOS mRNA is present in the 2-cell, morula, and blastocyst stages, iNOS in the morula and blastocyst stages, and nNOS in all three stages of preimplantation mouse embryo development.44 In addition, using immunocytochemistry, we have shown that protein for at least one of the NOS isoforms is present in each stage of preimplantation embryo development in the mouse (unpublished data). All three isoform proteins are present in day 2 and day 3 embryos. However, only iNOS and eNOS proteins are present in the day 4 embryo (Figure 14.1). In a

A

B

C

D

Figure 14.1 Immunocytochemistry of nNOS, iNOS, and eNOS in mouse blastocyst embryos. Negative and positive controls for each isoform were run alongside the embryos (brain for nNOS, activated macrophages for iNOS, and heart for eNOS). (A) Negative control; (B) nNOS; (C) iNOS; (D) eNOS. The red deposits (arrow) indicate the presence of protein. Embryo slides are shown at a 60 magnification.



HPE_Chapter14.qxp

7/13/2007

4:52 PM

Page 175


different set of experiments, we demonstrated that the localization of eNOS and iNOS is highest in trophoblast cells in delayed implanting blastocysts exposed to estrogen for 1 hour. Others have also looked at immunostaining with eNOS and iNOS antibodies, and with only one exception have found that both proteins are present in all preimplantation embryo stages studied.41,42,71 Gagioti et al did identify iNOS and eNOS staining in blastocysts, and the reason for being unable to see this in morula embryos is unclear. In blastocysts, however, there was evidence of more intense iNOS staining in the outer cells of the morula, and in the blastocyst trophoblasts.41,42 Ariel et al also found eNOS expression in human embryos;72 other isoforms were not studied. Interestingly, although we found nNOS mRNA to be present at all stages, immunocytochemical staining identified nNOS protein in 2-cell and morula embryos, but not in blastocysts. Abe et al73 also found evidence of nNOS mRNA in embryos from the 1- to 4-cell stage, followed by decreased expression thereafter. It would seem that either nNOS mRNA is regulated at a post-transcriptional level, or levels of nNOS protein are not detectable by our immunocytochemical techniques. The fact that Abe et al were unable to find any evidence of eNOS and iNOS mRNA in their studies is also puzzling. We have found that great care in the handling of samples is required in order to acquire mRNA that is of sufficient quality to produce consistent results in RT-PCR, and have also found it necessary to increase the amount of RNAse inhibitors used in sample preparation. There may be differential degradation of some of the NOS isoform mRNAs, which would explain the conflict between our real-time RT-PCR data and immunocytochemistry, and might also explain the conflict with reports by Chen et al, Nishikimi et al, and Abe et al.41,42,73 The discussion of current literature presented in this review clearly indicates that NO is required for embryo development, and the concentrations of NO must be tightly regulated. It is also clear that there is differential regulation of the NOS isoforms, and at least two, and usually all three isoforms can be found in all stages of mouse preimplantation

embryo development. However, these data are conflicting, since all NOS knockout lines currently available are still viable, albeit with impaired fertility and fecundity. Our laboratory therefore attempted to produce triple knockout animals in order to show that NO is absolutely required for embryo development. Our attempts to generate double knockouts resulted in only 12 double knockout animals instead of the expected out of 773 total pups born. Not only was there prenatal loss of double knockout animals, there was also preferential loss of males, as only two of the 12 pups born were male. This decrease in the number of males still does not account for the decreased production of double knockouts overall. However, another group reported the generation of NOS double knockout mice and do not indicate difficulties in breeding. While this group derived their animals from the same NOS1 mutant mouse line as was used in our experiment, they used a different NOS3 line.74 A triple knockout mouse line has also recently been reported, but these investigators obtained their NOS1 and NOS3 knockout lines from the group that produced the above double knockout line. In addition to the NOS3 line being different, their NOS2 line is also different to the line that we used, and therefore our line had a slightly dissimilar background, and areas of gene disruption. It is possible that these variations in background result in differences in gene segregation and changes in phenotype.75 Previous studies have also shown that splice variants for NOS isoforms can lead to low but persistent NOS activity, even in NOS knockout animals.76–78 We therefore attribute the apparent difference in production of double knockouts between our studies and those using other NOS / strains to a difference in the ability to produce NOS, as a result of alternative splice variants not found in our animals. In addition, our NOS2 NOS3 double mutants were produced from founder lines that were made using the same embryonic stem (ES) cell line (E14T62aES), which would allow for fewer strain difference interactions and give a clearer indication of the role of these NOS isoforms with respect to NO production and its role in embryonic development. However, our experiments did not determine at what point during embryonic development our 

HPE_Chapter14.qxp

7/13/2007

4:52 PM

Page 176


double knockout animals were lost, and in the absence of these data we cannot conclusively state that inhibiting NO production results in arrested embryo development at preimplantation stages. All of the evidence suggests that NO clearly has a role in embryonic development. Since NO signaling may occur through multiple mechanisms, a great deal remains to be investigated in order to understand the pathways that may be involved in NO regulation of the embryonic cell cycle, and whether other molecules can subsume the regulation by NO. Hopefully, a more complete picture should emerge with the elucidation of pathways used by NO and with additional work on knockout animals to determine the timing of embryos loss. In addition, information regarding downstream pathways could potentially be used to identify potential targets for novel IVF selection technologies. Such information could also allow for improvements in our ability to produce culture media that more adequately supports optimal levels of NO production and thus enhances the activation of downstream pathways to improve cleavage rates and quality of human IVF embryos. ACKNOWLEDGMENTS

Much of the unpublished data listed in this chapter is derived from work done by Dr Nury Steuerwald, Simone Hendrickson, and Jennifer Seegers, and I should like to express my gratitude to all members of the laboratory past and present for their contributions to it. Special thanks to Dr Laura W Schrum for her review of the manuscript. The work described has been supported by grants from the Charlotte Mecklenburg Hospital Authority, The University of North Carolina at Charlotte, and NIH. REFERENCES 1. Thaler CD, Epel D. Nitric oxide in oocyte maturation, ovulation, fertilization, cleavage and implantation: a little dab’ll do ya. Curr Pharm Des 2003; 9(5): 399–409. 2 Gouge RC, Marshburn P, Gordon BE, Nunley W, Huet-Hudson YM. Nitric oxide as a regulator of embryonic development. Biol Reprod 1998; 58(4): 875–9. 3 Manser RC, Houghton FD. Ca2-linked upregulation and mitochondrial production of nitric oxide in the mouse preimplantation embryo. J Cell Sci 2006; 119(Pt 10): 2048–55.



4. Moncada S, Higgs EA. Endogenous nitric oxide: physiology, pathology and clinical relevance. Eur J Clin Invest 1991; 21(4): 361–74. 5. Garthwaite J, Boulton CL. Nitric oxide signaling in the central nervous system. Annu Rev Physiol 1995; 57: 683–706. 6. Griffith OW, Stuehr DJ. Nitric oxide synthases: properties and catalytic mechanism. Annu Rev Physiol 1995; 57: 707–36. 7. Rosselli M, Keller PJ, Dubey RK. Role of nitric oxide in the biology, physiology and pathophysiology of reproduction. Hum Reprod Update 1998; 4(1): 3–24. 8. Forstermann U, Trogisch G, Busse R. Species-dependent differences in the nature of endothelium-derived vascular relaxing factor. Eur J Pharmacol 1984; 106(3): 639–43. 9. Denninger JW, Marletta MA. Guanylate cyclase and the NO/cGMP signaling pathway. Biochim Biophys Acta 1999; 1411(2-3): 334–50. 10. Nathan C, Xie QW. Regulation of biosynthesis of nitric oxide. J Biol Chem 1994; 269(19): 13725–8. 11. Wetzka B, Schafer WR, Stehmans A, Zahradnik HP. Effects of nitric oxide donors on the contractility and prostaglandin synthesis of myometrial strips from pregnant and non-pregnant women. Gynecol Endocrinol 2001; 15(1): 34–42. 12. Alderton WK, Cooper CE, Knowles RG. Nitric oxide synthases: structure, function and inhibition. Biochem J 2001; 357(Pt 3): 593–615. 13. Corson MA, James NL, Latta SE et al. Phosphorylation of endothelial nitric oxide synthase in response to fluid shear stress. Circ Res 1996; 79(5): 984–91. 14. Fisslthaler B, Dimmeler S, Hermann C, Busse R, Fleming I. Phosphorylation and activation of the endothelial nitric oxide synthase by fluid shear stress. Acta Physiol Scand 2000; 168(1): 81–8. 15. Hayashi Y, Nishio M, Naito Y et al. Regulation of neuronal nitric-oxide synthase by calmodulin kinases. J Biol Chem 1999; 274(29): 20597–602. 16. Komeima K, Hayashi Y, Naito Y, Watanabe Y. Inhibition of neuronal nitric-oxide synthase by calcium/calmodulin-dependent protein kinase IIalpha through Ser847 phosphorylation in NG108-15 neuronal cells. J Biol Chem 2000; 275(36): 28139–43. 17. Nakane M, Mitchell J, Forstermann U, Murad F. Phosphorylation by calcium calmodulin-dependent protein kinase II and protein kinase C modulates the activity of nitric oxide synthase. Biochem Biophys Res Commun 1991; 180(3): 1396–402. 18. Zembowicz A, Tang JL, Wu KK. Transcriptional induction of endothelial nitric oxide synthase type III by lysophosphatidylcholine. J Biol Chem 1995; 270(28): 17006–10. 19. Hirata K, Miki N, Kuroda Y et al. Low concentration of oxidized lowdensity lipoprotein and lysophosphatidylcholine upregulate constitutive nitric oxide synthase mRNA expression in bovine aortic endothelial cells. Circ Res 1995; 76(6): 958–62. 20. Freeman JE, Kuo WY, Drenger B et al. Analysis of lysophophatidylcholine-induced endothelial dysfunction. J Cardiovasc Pharmacol 1996; 28(3): 345–52. 21. Zhang J, Patel JM, Li YD, Block ER. Proinflammatory cytokines downregulate gene expression and activity of constitutive nitric oxide synthase in porcine pulmonary artery endothelial cells. Res Commun Mol Pathol Pharmacol 1997; 96(1): 71–87. 22. Arnal JF, Yamin J, Dockery S, Harrison DG. Regulation of endothelial nitric oxide synthase mRNA, protein, and activity during cell growth. Am J Physiol 1994; 267(5 Pt 1): C1381–8. 23. Braude P, Bolton V, Moore S. Human gene expression first occurs between the four- and eight-cell stages of preimplantation development. Nature 1988; 332(6163): 459–61. 24. Delouis C, Bonnerot C, Vernet M, Nicolas JF. Expression of microinjected DNA and RNA in early rabbit embryos: changes in permissiveness

HPE_Chapter14.qxp

7/13/2007

4:52 PM

Page 177


25.

26.

27. 28. 29. 30. 31. 32.

33.

34. 35. 36.

37. 38. 39.

40.

41.

42.

43.

44.

45.

46.

for expression and transcriptional selectivity. Exp Cell Res 1992; 201(2): 284–91. Flach G, Johnson MH, Braude PR, Taylor RA, Bolton VN. The transition from maternal to embryonic control in the 2-cell mouse embryo. EMBO J 1982; 1(6): 681–6. Jarrell VL, Day BN, Prather RS. The transition from maternal to zygotic control of development occurs during the 4-cell stage in the domestic pig, Sus scrofa: quantitative and qualitative aspects of protein synthesis. Biol Reprod 1991; 44(1): 62–8. Memili E, Dominko T, First NL. Onset of transcription in bovine oocytes and preimplantation embryos. Mol Reprod Dev 1998; 51(1): 36–41. Zernicka-Goetz M. Activation of embryonic genes during preimplantation rat development. Mol Reprod Dev 1994; 38(1): 30–5. Pines J. Cyclins and cyclin-dependent kinases: a biochemical view. Biochem J 1995; 308 (Pt 3): 697–711. Morgan DO. Principles of CDK regulation. Nature 1995; 374(6518): 131–4. Smits VA, Medema RH. Checking out the G(2)/M transition. Biochim Biophys Acta 2001; 1519(1-2): 1–12. Ma T, Van Tine BA, Wei Y et al. Cell cycle-regulated phosphorylation of p220(NPAT) by cyclin E/Cdk2 in Cajal bodies promotes histone gene transcription. Genes Dev 2000; 14(18): 2298–313. Guo K, Andres V, Walsh K. Nitric oxide-induced downregulation of Cdk2 activity and cyclin A gene transcription in vascular smooth muscle cells. Circulation 1998; 97(20): 2066–72. Nigg EA. Cyclin-dependent protein kinases: key regulators of the eukaryotic cell cycle. Bioessays 1995; 17(6): 471–80. Hartwell LH, Weinert TA. Checkpoints: controls that ensure the order of cell cycle events. Science 1989; 246(4930): 629–34. Hogan B, Constantini, F, Lacy E. Manipulating the Mouse Embryo: A Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Laboratory, 1986. Brinster RL. Nutrition and metabolism of the ovum, zygote, and blastocyst. Washington, DC: American Physiological Society, 1973. Whitten WK. Nutrition Requirements for the Culture of Preimplantation Embryos in vitro. Adv Biosci 1971; 6: 129-139. Novaro V, Rettori V, Gonzalez ET et al. Interaction between uterine PGE and PGF2 alpha production and the nitridergic system during embryonic implantation in the rat. Prostaglandins 1996; 51(6): 363–76. Sengoku K, Takuma N, Horikawa M et al. Requirement of nitric oxide for murine oocyte maturation, embryo development, and trophoblast outgrowth in vitro. Mol Reprod Dev 2001; 58(3): 262–8. Chen HW, Jiang WS, Tzeng CR. Nitric oxide as a regulator in preimplantation embryo development and apoptosis. Fertil Steril 2001; 75(6): 1163–71. Nishikimi A, Matsukawa T, Hoshino K et al. Localization of nitric oxide synthase activity in unfertilized oocytes and fertilized embryos during preimplantation development in mice. Reproduction 2001; 122(6): 957–63. Orsi NM. Embryotoxicity of the nitric oxide donor sodium nitroprusside in preimplantation bovine embryos in vitro. Anim Reprod Sci 2006; 91(3–4): 225–36. Tranguch S, Steuerwald N, Huet-Hudson YM. Nitric oxide synthase production and nitric oxide regulation of preimplantation embryo development. Biol Reprod 2003; 68(5): 1538–44. Chung HT, Pae HO, Choi BM, Billiar TR, Kim YM. Nitric oxide as a bioregulator of apoptosis. Biochem Biophys Res Commun 2001; 282(5): 1075–9. Nguyen T, Brunson D, Crespi CL et al. DNA damage and mutation in human cells exposed to nitric oxide in vitro. Proc Natl Acad Sci USA 1992; 89(7): 3030–4.

47. Dinerman JL, Lowenstein CJ, Snyder SH. Molecular mechanisms of nitric oxide regulation. Potential relevance to cardiovascular disease. Circ Res 1993; 73(2): 217–22. 48. Murad F. Regulation of cytosolic guanylyl cyclase by nitric oxide: the NO-cyclic GMP signal transduction system. Adv Pharmacol 1994; 26: 19–33. 49. Nunokawa Y, Tanaka S. Interferon-gamma inhibits proliferation of rat vascular smooth muscle cells by nitric oxide generation. Biochem Biophys Res Commun 1992; 188(1): 409–15. 50. Kawahara K, Gotoh T, Oyadomari S et al. Nitric oxide inhibits the proliferation of murine microglial MG5 cells by a mechanism involving p21 but independent of p53 and cyclic guanosine monophosphate. Neurosci Lett 2001; 310(2-3): 89–92. 51. Denham S, Rowland IJ. Inhibition of the reactive proliferation of lymphocytes by activated macrophages: the role of nitric oxide. Clin Exp Immunol 1992; 87(1): 157–62. 52. Nisoli E, Clementi E, Tonello C et al. Effects of nitric oxide on proliferation and differentiation of rat brown adipocytes in primary cultures. Br J Pharmacol 1998; 125(4): 888–94. 53. Buga GM, Wei LH, Bauer PM, Fukuto JM, Ignarro LJ. NG-hydroxy-Larginine and nitric oxide inhibit Caco-2 tumor cell proliferation by distinct mechanisms. Am J Physiol 1998; 275(4 Pt 2): R1256–64. 54. Pervin S, Singh R, Chaudhuri G. Nitric oxide-induced cytostasis and cell cycle arrest of a human breast cancer cell line (MDA-MB-231): potential role of cyclin D1. Proc Natl Acad Sci USA 2001; 98(6): 3583–8. 55. Hajri A, Metzger E, Vallat F et al. Role of nitric oxide in pancreatic tumour growth: in vivo and in vitro studies. Br J Cancer 1998; 78(7): 841–9. 56. Ulibarri JA, Mozdziak PE, Schultz E, Cook C, Best TM. Nitric oxide donors, sodium nitroprusside and S-nitroso-N-acetylpencillamine, stimulate myoblast proliferation in vitro. In Vitro Cell Dev Biol Anim 1999; 35(4): 215–8. 57. Du M, Islam MM, Lin L et al. Promotion of proliferation of murine BALB/C3T3 fibroblasts mediated by nitric oxide at lower concentrations. Biochem Mol Biol Int 1997; 41(3): 625–31. 58. Luczak K, Balcerczyk A, Soszynski M, Bartosz G. Low concentration of oxidant and nitric oxide donors stimulate proliferation of human endothelial cells in vitro. Cell Biol Int 2004; 28(6): 483–6. 59. Sauzeau V, Rolli-Derkinderen M, Marionneau C, Loirand G, Pacaud P. RhoA expression is controlled by nitric oxide through cGMP-dependent protein kinase activation. J Biol Chem 2003; 278(11): 9472–80. 60. Katsuyama K, Shichiri M, Marumo F, Hirata Y. NO inhibits cytokineinduced iNOS expression and NF-kappaB activation by interfering with phosphorylation and degradation of IkappaB-alpha. Arterioscler Thromb Vasc Biol 1998; 18(11): 1796–802. 61. Fiedler B, Lohmann SM, Smolenski A et al. Inhibition of calcineurinNFAT hypertrophy signaling by cGMP-dependent protein kinase type I in cardiac myocytes. Proc Natl Acad Sci USA 2002; 99(17): 11363–8. 62. Pilz RB, Suhasini M, Idriss S, Meinkoth JL, Boss GR. Nitric oxide and cGMP analogs activate transcription from AP-1-responsive promoters in mammalian cells. FASEB J 1995; 9(7): 552–8. 63. Thiriet N, Esteve L, Aunis D, Zwiller J. Immediate early gene induction by natriuretic peptides in PC12 phaeochromocytoma and C6 glioma cells. Neuroreport 1997; 8(2): 399–402. 64. Sharma RV, Tan E, Fang S, Gurjar MV, Bhalla RC. NOS gene transfer inhibits expression of cell cycle regulatory molecules in vascular smooth muscle cells. Am J Physiol 1999; 276(5 Pt 2): H1450–9. 65. Fukumoto S, Koyama H, Hosoi M et al. Distinct role of cAMP and cGMP in the cell cycle control of vascular smooth muscle cells: cGMP



HPE_Chapter14.qxp

7/13/2007

4:52 PM

Page 178


66.

67.

68.

69. 70.

71.

72.



delays cell cycle transition through suppression of cyclin D1 and cyclin-dependent kinase 4 activation. Circ Res 1999; 85(11): 985–91. Hanada S, Terada Y, Inoshita S et al. Overexpression of protein kinase G using adenovirus inhibits cyclin E transcription and mesangial cell cycle. Am J Physiol Renal Physiol 2001; 280(5): F851–9. Kronemann N, Nockher WA, Busse R, Schini-Kerth VB. Growthinhibitory effect of cyclic GMP- and cyclic AMP-dependent vasodilators on rat vascular smooth muscle cells: effect on cell cycle and cyclin expression. Br J Pharmacol 1999; 126(1): 349–57. Chan GK, Yen TJ. The mitotic checkpoint: a signaling pathway that allows a single unattached kinetochore to inhibit mitotic exit. Prog Cell Cycle Res 2003; 5: 431–9. McGowan CH, Russell P. The DNA damage response: sensing and signaling. Curr Opin Cell Biol 2004; 16(6): 629–33. Sibon OC, Laurencon A, Hawley R, Theurkauf WE. The Drosophila ATM homologue Mei-41 has an essential checkpoint function at the midblastula transition. Curr Biol 1999; 9(6): 302–12. Gagioti S, Scavone C, Bevilacqua E. Participation of the mouse implanting trophoblast in nitric oxide production during pregnancy. Biol Reprod 2000; 62(2): 260–8. Ariel I, Hochberg A, Shochina M. Endothelial nitric oxide synthase immunoreactivity in early gestation and in trophoblastic disease. J Clin Pathol 1998; 51(6): 427–31.

73. Abe KM, Inoue N, Taga M, Kato T. Messenger RNA of neuronal nitric oxide synthase is expressed and possibly functions in mouse oocytes and embryos during preimplantation development. Biomed Res (Tokyo) 1999; 20: 61–65. 74. Son H, Hawkins RD, Martin K et al. Long-term potentiation is reduced in mice that are doubly mutant in endothelial and neuronal nitric oxide synthase. Cell 1996; 87(6): 1015–23. 75. Phillips TJ, Hen R, Crabbe JC. Complications associated with genetic background effects in research using knockout mice. Psychopharmacology (Berl) 1999; 147(1): 5–7. 76. Huber A, Saur D, Kurjak M, Schusdziarra V, Allescher HD. Characterization and splice variants of neuronal nitric oxide synthase in rat small intestine. Am J Physiol 1998; 275(5 Pt 1): G1146–56. 77. Rothe F, Huang PL, Wolf G. Ultrastructural localization of neuronal nitric oxide synthase in the laterodorsal tegmental nucleus of wildtype and knockout mice. Neuroscience 1999; 94(1): 193–201. 78. Meng W, Ma J, Ayata C et al. ACh dilates pial arterioles in endothelial and neuronal NOS knockout mice by NO-dependent mechanisms. Am J Physiol 1996; 271(3 Pt 2): H1145–50.

HPE_Chapter15.qxp

7/14/2007

5:54 PM

Page 179

15. Uptake and release of metabolites in human preimplantation embryos Fabienne Devreker

INTRODUCTION

Selection of human embryos for transfer following in vitro fertilization (IVF) programs is routinely based on pronuclear or embryo morphology and cleavage rate.1 Despite numerous attempts to develop embryo scoring techniques that can predict embryo viability, implantation rates remain low.1 Not all embryos with good morphology will implant, and fragmented embryos can develop to the blastocyst stage in vitro.2 In fact, embryo selection based on morphology fundamentally eliminates only severely impaired embryos. To achieve acceptable pregnancy rates more than one embryo is routinely transferred, with an unavoidable risk of multiple gestations. Micromethods for assessing the metabolism of embryos were developed in order to improve the selection of embryos with the best developmental potential. Measurements of metabolite uptake and production assist in understanding how preimplantation embryos interact with their in vitro or in vivo conditions, and also give important information about components that can have beneficial or deleterious effects on embryo viability. In vitro culture conditions have been shown to affect the embryo during early and late development. Alterations of in vitro culture conditions can modify gene expression,3,4 cellular metabolism,5 or imprinting status6 during early development, and may also have profound effects on prenatal or postnatal development, such as the large offspring syndrome observed when bovine embryos are cultured in the presence of serum.6 Preimplantation embryo development is a complex process that produces a multicellular organism from a single cell. Cellular division and differentiation are associated with different metabolic pathways

that mainly focus on energy production and synthesis of metabolic precursors for macromolecules. Understanding early embryo metabolism is also essential in order to set up culture media and conditions better adapted to preimplantation development in vitro, reducing possible adverse effects during in vitro culture.3 Metabolic measurements may also help in the selection of embryos for transfer in IVF programs. This chapter focuses on embryo metabolism that can be measured by non-invasive techniques. It describes the main pathways that produce ATP from glucose or amino acids once they enter the cell, and presents the different techniques used to evaluate the uptake or release of metabolites by embryos in vitro and how this uptake or release can be correlated to embryo viability.

EARLY PREIMPLANTATION EMBRYO DEVELOPMENT

The preimplantation period begins with oocyte fertilization and finishes with the formation of a hatched blastocyst ready to implant. In humans, this period lasts for approximately 5–7 days. The mature oocyte contains sufficient maternal transcripts and proteins to support the process of fertilization and the first two cell divisions. The activation of the embryonic genome, which represents the transition from maternal to embryonic genome control, is thought to occur during the third cell division. At compaction, the first epithelium is formed and two distinct cell lines can be distinguished, one that will form the trophectoderm (TE) and the other that will form the fetus. These major events are associated with a drastic increase in embryo metabolism 

HPE_Chapter15.qxp

7/14/2007

5:54 PM

Page 180


that involves the production of ATP, mRNA, and proteins. Several experiments performed on preimplantation embryos from different species, including mouse, cow, rat, pig, sheep, and human, showed that embryo metabolism is different in the pre- and postcompaction stages, corresponding with before and after the fourth cell division. Before compaction, blastomeres are loosely joined together and therefore equally exposed to their environment. Precompaction stage embryos rely on maternal storage and have a relatively low level of biosynthesis,7 low respiratory rates,8,9 and a limited ability to metabolize glucose as a source of energy.10 On the other hand, postcompaction stage embryos have developed important gap-junctional complexes in the outer embryonic cells11 and cytoplasmic and membrane polarization have been observed. Two distinct apical and basal membranes, that are kept separated by zonular tight junctions, are formed. The formation of tight cellular junctions between the outer cells creates two different microenvironments for the inner and outer cells. The inner cells exposed to the inside microenvironment will form the inner cell mass (ICM), while outer cells will form the trophectoderm.12 Embryos then have a high rate of biosynthesis and an exponential increase in their demand for energy. Indeed, blastocyst formation requires new proteins or enzymes for the formation of cellular junctional complexes, cell differentiation, and the activation of a Na⫹/K⫹ ATPase pump for the formation of the blastocoel cavity. In mouse and rabbit embryos, the formation of the blastocoel is associated with a significant increase in the synthesis and activity of Na⫹/K⫹ ATPase.13 Postcompaction stage embryos become able to metabolize glucose and to actively control their cytoplasmic pH and composition (Table 15.1).10,14,15

TECHNIQUES FOR MEASURING EMBRYO METABOLISM IN VITRO

The metabolism of mammalian embryos from several species has been extensively studied, including 

mouse,10,16–21 hamster,22 cow,23,24 and rabbit.25,26 Much less is known about other species such as sheep,9,27 rat,28 pig,29 and human.2,10,14,18 Carbohydrate metabolism of embryos has been evaluated by measurements of substrate uptake or release using either radiolabeled substrates16–18,29,31 or noninvasive microfluorescence assays,1,2,10,14 and by comparing embryo development in vitro in the presence or absence of energy substrates in the culture medium (the ‘nutritional’ approach). Radioisotopic studies use different markers as tracers, including [14C]-glucose, [14C]-pyruvate, or [14C]-lactate and their products, such as 14CO2 or 14C-lactate, that can be collected or isolated chromatographically before measuring their radioactivity.10 This technique made it possible to both quantify substrate utilization and to identify the metabolic pathways used by the embryos. The disadvantage of this technique is that measurements are made for a group of embryos, and it is therefore not possible to correlate embryo metabolism with embryo development or viability post-transfer. Furthermore, radiolabeled substrates can have toxic effects on the embryo. Radiolabeled experiments have, for example, demonstrated that CO2 production is both stage and substrate specific. Incubating mouse embryos with [14C]-pyruvate or [14C]-lactate as the sole substrate showed that 14CO2 production by unfertilized oocytes, fertilized, and 8-cell embryos was low, and increased oocytes from the morula to blastocyst stages with both substrates. However, the

Table 15.1 Metabolic differences between cleavage stage and postcompaction stage embryos Cleavage stage

Postcompaction stage

Inability to metabolize glucose Pyruvate is essential Low level of ATP production Low level of mRNA production Low respiratory rate

Switch to glucose metabolism Pyruvate is less important Activation of anaerobic glycolysis High level of ATP production High level of mRNA High respiratory rate

Low protein synthesis Beneficial effect of non-essential amino acids High sensitivity to pH variations

Increase in protein synthesis Requirements for the 20 Eagle’s amino acids Less sensitivity to pH variations

HPE_Chapter15.qxp

7/14/2007

5:54 PM

Page 181

UPTAKE AND RELEASE OF METABOLITES IN HUMAN PREIMPLANTATION EMBRYOS

production of CO2 from lactate by post compaction embryos was 50% lower compared with the CO2 from pyruvate.16 The nutritional approach analyzes embryo development in vitro in the presence or absence of determined components in the culture medium, including carbohydrates, amino acids, oxygen, etc. In order to determine the toxicity or the beneficial effects of the tested component, several parameters are compared, such as the percentage of embryos reaching the blastocyst stage, implantation posttransfer, or more invasively, the cell numbers in the TE and ICM32 in resultant blastocysts. Once the new culture conditions have been proven to adequately support embryo development in vitro, these conditions can be used in clinical programs and the outcome of the resultant offspring can be assessed. Non-invasive microfluorescence assays make it possible to measure substrate depletion or release by individual embryos in culture media and to correlate metabolism with embryo viability.2,10,14,33 This method relies on the generation or consumption of the reduced pyridine nucleotides NADH and NADPH.34 These nucleotides have the property of absorbing light at 340 nm, whereas the oxidized forms, NAD⫹ and NADP⫹, emit fluorescence when activated with light at 340 nm. By choosing appropriate enzymatic reactions, the increase or decrease in the absorbance of fluorescence due to NADH or NADPH formation or oxidation can be related to the concentration of a metabolite or to the rate of an enzymatic reaction.14,20 The measurements have been facilitated by adapting these fluorometric assays to a Cobas-Bio autoanalyzer (Roche Instruments, UK).35 In the human, spare embryos are incubated singly in 4 or 5 ␮l droplets of culture medium for 24 hours, and 1 or 2 ␮l samples of the medium are then analyzed for nutritional content.32,35,36 By choosing appropriate enzymatic reactions, this method can be used to measure enzyme activities,37 glucose, pyruvate, lactate, or ammonium.38 For amino acids, the spent medium is analyzed by reverse-phase high performance liquid chromatography (HPLC).39 Ammonium excretion is measured in the spent medium by a quantitative enzymatic reaction using a Cobas-Bio autoanalyzer.

The method is based on the reductive amination of 2-oxoglutarate, using glutamate dehydrogenase and NADPH. To measure oxygen uptake, embryos are incubated in groups in 5 ␮l polymerase chain reaction micropipettes. The reaction is based on the fluorescence emitted by pyrene. Pyrene is a highly fluorescent non-toxic compound that is excited at 340 nm and emits light at 450 nm. The oxygen consumption is assessed by the change in pyrene’s fluorescence, measured by a fluorescence microscope with photomultiplier and photometer attachment.8 After the period of incubation, embryos are replaced in drops of fresh medium and cultured to later stages. Other methods are available that are less easy to perform, based on the conversion of oxyhemoglobin to hemoglobin or on the use of a microelectrode.8 Evaluation of ATP consumption is either measured through HPLC,8 or calculated on the basis that one mole of oxygen produces 6 moles of ATP, or that one mole of glucose produces 2 moles of ATP by anaerobic glycolysis.40

CARBOHYDRATE METABOLISM

Embryos use two main pathways to generate the ATP that is necessary for cellular metabolism: aerobic glycolysis or the tricarboxylic acid cycle (TCA or Krebs cycle) and anaerobic glycolysis by the EmbdenMeyerhof pathway (EMP). Glucose, pyruvate, and lactate occupy a central position as substrates for both energy production and for synthesis of complex molecules. Glucose enters the cell by either diffusion or facilitated transport (GLUT1).41 Once it enters the cell, glucose is directly phosphorylated into glucose 6-phosphate (G-6-P). At this stage glucose can enter into four major pathways (Figure 15.1).43 The first is the EMP, anaerobic glycolysis that converts one molecule of glucose into two molecules of pyruvate (a 3-carbon molecule) with the production of two molecules of ATP. In the presence of oxygen and the relative absence of lipids or other substrates, pyruvate enters the Krebs cycle for oxidative metabolism in mitochondria and is converted into acetyl-CoA, 

HPE_Chapter15.qxp

7/14/2007

5:54 PM

Page 182


3

Glucose

2

1 2 ATP Hexose phosphate

Hexosamines

Pentose phosphate

2 ADP + Pi Glyceraldehyde 3-phosphate Glycoproteins Glycolipids Glycosaminoglycans

4 ADP + Pi Triose phosphate 4 ATP

NADH

NAD+ + H+

Pyruvate

Cytosol Mitochondria

Ribose 5-phosphate NADPH

Lactate

Acetyl-CoA NAD+

32 ATP TCA Cycle

NADH

FADH

2Fe3+

Flavoprotein

Cytochrome

FAD+

2Fe2+

H2O

O2

Electron transport chain Figure 15.1 Diagram illustrating three catabolic pathways of glucose. (1) Through glycolysis, glucose is degraded into pyruvate with the production of 2 molecules of ATP. (2) Through transamination with amino acids, glucose participates in the synthesis of many glycoproteins and glycolipids. (3) The pentose-phosphate shunt uses glucose for the synthesis of DNA and RNA molecules. The storage of glucose under the form of glycogen is not represented in this diagram. Adapted from Rieger et al.

yielding 38 molecules of ATP for each molecule of glucose consumed. In the absence of oxygen, pyruvate is converted into lactate (Figure 15.1). The second pathway is the pentose phosphate shunt which produces NADPH and ribose 5-phosphate (Figure 15.1). The reduced equivalents of NADP⫹ are transferred to glutathione for the protection of the cell against peroxidation, and used in the synthesis of lipids or other complex molecules.42 Ribose 5-phosphate is an obligatory precursor for the synthesis of all nucleotides. The third pathway involves the transfer of one amine group (transamination) from glutamine to 

fructose, to form glucosamine 6-phosphate and glutamate. Glucose is therefore a precursor for glycoproteins, mucoproteins, and mucopolysaccharides (Figure 15.1). The fourth pathway is used when the glucose supply is in excess of requirements, and thus glucose is stored within the cell in the form of glycogen, a branched complex polymer of glucose molecules. In the mouse, pyruvate enters embryos by a combination of diffusion and facilitated transport.20,31 Pyruvate then enters into the TCA cycle to produce ATP. Lactate appears to enter into the embryo by rapid diffusion.18 Lactate has been shown

HPE_Chapter15.qxp

7/14/2007

5:54 PM

Page 183


to be the preferred substrate for 1-cell hamster embryos.43 In the mouse, lactate can be utilized as an energy source from the 2-cell stage onwards, and acts synergistically with pyruvate.44 Lactate is, however, mainly a product of glucose oxidation (Figure 15.1). PYRUVATE

Pyruvate is usually preferred to glucose during the early stages of development, and, using a nutritional approach, has been shown to be essential for the early preimplantation embryo development of mouse45 and human embryos.46 Pyruvate uptake by mouse embryos is already present at the 1-cell stage, increases throughout development and declines slightly at the blastocyst stage.14,31,35,36 Pyruvate is a major energy source during human oocyte maturation (Figure 15.2).47,48 After the activation of the embryonic genome, pyruvate uptake by mouse or human embryos exceeds glucose uptake from cleavage stage up to the blastocyst stage, to reach a maximum of 40–45 pmol/l per embryo/hour at the morula stage.2,35,49 If glucose is available in the culture medium, pyruvate uptake declines slightly at the morula stage.2,10,35

Pyruvate uptake has been correlated to embryo viability: embryos that developed to the blastocyst stage took up more pyruvate than those that arrested.2,35 Consumption of pyruvate was related to embryo viability in patients with tubal or polycystic ovary (PCO) infertility. Suboptimal ovulation regimens resulted in embryos with lower pyruvate uptake and lower blastocyst cell numbers.50 Pyruvate uptake is lower when embryos are cultured in suboptimal conditions that compromise their viability (Figure 15.3).51 The uptake of pyruvate during the first 24 h following fertilization in single human embryos that were derived from natural cycles was correlated to embryo morphology. Since a single embryo was transferred, the uptake of pyruvate could be directly correlated with the embryo. The results showed that embryos had a wide range of pyruvate uptake values (2–53 pmol/l per embryo/h), but that this variation was reduced significantly to an intermediate range of values in those embryos that were able to implant (10–30 pmol/l per embryo/h).52 Similarly, pyruvate uptake by human embryos during the second and third day of development displays a wide range of values. Embryos that implanted, however, took up less pyruvate than those that did not (22.9 ⫾ 1.0 and

Unfertilized Activated Fertilized

20

40

*

15

*

Pyruvate uptake (pmol/l per embryo/h)

Pyruvate uptake (pmol/l per embryo/h)

25

10 5 0 Oocyte fertilization

30

20

With glutamine

10

Without glutamine

First cleavage division

Figure 15.2 Pyruvate consumption during oocyte fertilisation and first cleavage division. First period represents measurements during 24H from the time of insemination until the assessment of fertilization status. Second period represents measurements during 24H from Day 1 until Day 2 post insemination, at the time of transfer. Pyruvate uptake is measured in pmoles/ embryo/ hour.45 *: p ⬍ 0.01. Adapted from Devreker et al.

0 4–7 cell

8–16 cell

M

B

Stage Figure 15.3 Pyruvate uptake by individual human embryos in the presence or absence of glutamine. M, morula; B, blastocyst. Adapted from Devreker et al.51 *: p ⬍ 0.05.



HPE_Chapter15.qxp

7/14/2007

5:54 PM

Page 184


27.1 ⫾ 0.6 pmol/l per embryo/h, respectively, on day 2; and 22.4 ⫾ 1.5 and 26.9 ⫾ 0.8 pmol/l per embryo/h, respectively, on day 3).49 However, the wide range of values precluded the use of pyruvate uptake to predict embryos that will implant. More recently, pyruvate uptake by human oocytes can be detected earlier, during fertilization and the first divisions.47 Fertilization and activation increased pyruvate uptake by the human embryo by approximately 30%. However, it was not related to subsequent morphology, developmental stage at the time of transfer, or to implantation (Table 15.2). The low number of embryos that effectively implanted may explain the fact that measurements of pyruvate uptake by human embryos during the first 2 days of development did not help to predict embryo viability post-transfer. Culture of isolated human blastomeres in vitro showed that those that underwent cell division had a significantly higher pyruvate uptake than those that did not cleave (Hardy, personal communication). Thus, pyruvate uptake appears to be related to cell division, and in the zygote it is possible that the cytoskeletal changes associated with expulsion of the second polar body and movement of pronuclei48,53 require energy, which is provided by pyruvate. In another study, although pyruvate uptake by human embryos cultured in sequential media was higher for embryos that developed up to the blastocyst stage, this could not be correlated to embryo morphology.54 Table 15.2 Mean pyruvate uptake (pmol/embryo/h) by normally fertilized embryos according to the stage reached at the time of transfer. Values are mean ⫾ SEM. From Devreker et al.47 Stage reached by day 2 One cell stage 2–3-cell stage 4-cell stage ⬎4-cell stage Highly fragmented embryos aSignificantly

Whitney).



n 13 76 106 38 23

Day 0–1 10.0 ⫾ 2.2 15.6 ⫾ 1.4 13.4 ⫾ 0.9 19.7 ⫾ 1.9a,b 15.1 ⫾ 2.2

Day 1–2 11.8 ⫾ 2.0 15.0 ⫾ 1.3 16.0 ⫾ 1.0 12.5 ⫾ 1.5 13.3 ⫾ 2.6

higher compared with the 1-cell stage; b *p ⫽ 0.010 (Mann–

GLUCOSE

In contrast to other cell lines, cleavage stage embryos have little capacity to metabolize glucose prior to the activation of the embryonic genome. Using radiolabeled [14C]-glucose30 or microfluorescence assays,2,35,36,47 several authors have demonstrated that glucose uptake by early human embryos is very low before the 4-cell stage, less than 10 pmol/l per embryo/h,30 and reached 20–30 pmol/ l per embryo/h at the blastocyst stage.10,14,30,35,49 A similar pattern for glucose uptake has been observed for other mammalian species, including mouse,14,31,33 pig,29 rabbit,26 and bovine.20,23 For example, the total glucose metabolism almost doubles between the morula and expanded blastocyst stages in the cow55 and increases linearly with the volume of horse blastocysts.42 Glucose uptake by mouse embryos is low before the 8-cell stage and increases sharply at the transition from morula to blastocyst stage.14,33 The consumption of glucose was later shown to help in predicting the resistance of bovine embryos to the freezing–thawing procedure.1 Glucose uptake by human embryos prior to activation of the embryonic genome was undetectable using fluorescence assays.47 Gardner et al1 reported that day 4 human embryos that developed up to the blastocyst stage in sequential media culture took up significantly more glucose during blastocyst formation. Glucose uptake could also be correlated to blastocyst morphology, being higher in those blastocysts of highest grade. The absence of detection of glucose uptake by early human embryos does not necessarily mean that these embryos do not metabolize small quantity of glucose. Low levels of glucose in culture media are probably necessary for embryo development. However, as glucose can easily enter the cell by diffusion, a high level of glucose in culture media for precompaction embryos can have deleterious effect on embryo viability. Indeed, part of the glucose can be stored under the form of glycogene. Too much intracellular glycogene can disturb the cytoplasmic architecture and would be toxic for the cell.

HPE_Chapter15.qxp

7/14/2007

5:54 PM

Page 185


LACTATE

Lactate production by preimplantation embryos is low before compaction and increases thereafter, coincident with the rise in glucose consumption. Lactate production has been observed even in the absence of exogenous substrate, suggesting that embryos use some endogenous source of energy.2,12,14,33,35 Radiolabeled experiments showed that embryos produce lactate from glucose or pyruvate.18,48 A high lactate production, however, could be a sign of impaired embryo viability. Some authors56 have reported that mouse embryos with lower viability produced a higher amount of lactate in vitro. Alteration of embryo viability is compromised as early as 6 h of culture in vitro, and is associated with a net increase in lactate production.1 This increase in lactate production can be suppressed if amino acids and vitamins are added to the culture medium. Moreover, mouse embryos grown in vitro produced twice as much lactate as those grown in vivo.57 Measurements of both glucose consumption and lactate production by mouse embryos allowed their glycolytic activity to be evaluated. It was shown that blastocysts with the lowest level of anaerobic glycolysis had the highest potential to develop post-transfer.1 Van den Bergh et al58 reported a similar observation for human embryos cultured in sequential media. However, to date, lactate production has not been used to predict or evaluate human embryo viability before transfer.

AMINO ACIDS

Amino acids are one of the key constituents of cellular metabolism, and embryos are equipped with an array of sodium-dependent and -independent transporter systems that regulate amino acid flux and availability.8 Amino acids not only form the basic components of proteins, but also fulfil a number of other functions. They contribute to ATP production via deamination reactions, producing intermediary metabolites that can enter the TCA cycle, and also stimulate activation of the embryonic

genome, as well as blastocyst formation and hatching. Amino acids have been shown to act as osmolytes, and help to maintain intracellular pH.6,22,56 Glutamine, taurine, and glycine have been reported to protect cleavage stage embryos against osmotic shock.59 Glycine maintains embryo development in the presence of high osmolarities,60 and amino acids are thought to have a significant role in signal transduction cascades.60 By providing substrates to the TCA cycle, in mouse and hamster embryos amino acids can protect against the deleterious effects of glucose by inhibiting glycolytic activity.22,56 The oxidation of amino acids causes allosteric inhibition of the glycolytic enzyme phosphofructokinase through an increase in ATP and/or citrate, or by the direct inhibition of pyruvate kinase by alanine.56 Amino acids are present in large amounts in female reproductive tract fluids, zygotes, and embryos, and are therefore likely to be important for preimplantation embryo development.22,57 For example, glycine content is the highest in rabbit oviductal fluids, along with taurine, and is cycle dependent.61,62 Alanine, threonine, glutamate, serine, taurine, and glycine represent 85% of the amino acid content of rabbit oviductal fluids.61,62 In the mouse, the most abundant amino acids in oviduct and uterine fluids are glutamine, taurine, glycine, and alanine, while alanine, aspartate, glutamate, taurine, and glycine are among the most abundant amino acids in oocytes and embryos.61 Human uterine fluids contain large amounts of taurine, glutamate, glycine, and alanine.22 These four amino acids are also predominant in bovine uterine fluids.22 Amino acids singly or in combination have been shown to promote the early development of embryos from various species, including mouse,63 cow,64 hamster,22 sheep,65 rabbit,66 rat,67 and human.51,56,68,69 However, they have differential effects on embryo development; some promote, while others can inhibit embryo development.22 Several authors therefore measured amino acid depletion or appearance in the culture medium during embryo development in vitro. The profile of amino acid uptake or secretion differs between species, and also at different embryo



HPE_Chapter15.qxp

7/14/2007

5:54 PM

Page 186


stages.38,64,70,71 Overall, amino acid depletion increases from early cleavage stage up to the blastocyst stage. In human embryos, the amino acid profile differed between embryos that reached the blastocyst stage compared with those that arrested.39,72 Developing embryos depleted amino acids from the culture medium to a lesser extent than did those that arrested, and the profile was also different, in that developing embryos took up significantly more alanine, glycine, and serine.39 However, the amino acid profile could not be related to embryo morphology. Appearance of alanine in the culture medium during early cleavage could be used to select embryos that will arrest their development. Houghton et al39 reported that a higher concentration of alanine in the culture medium was observed for arrested embryos, and suggested that alanine is used to metabolize ammonium production by embryos. The presence of high levels of alanine in the culture medium can therefore reflect the level of amino acid turnover and ammonium production. As developing embryos have a low turnover level, the amount of alanine in the culture medium can be related to embryo viability and therefore may help to select embryos before transfer.72 Leucine is another amino acid that could be useful to assay, as leucine is the only amino acid that is significantly depleted from the culture by early cleavage stage human embryos.39,72

OXYGEN CONSUMPTION

Oxygen is essential for the conversion of ADP to ATP in oxidative phosphorylation through its role as an electron acceptor in the mitochondrial electron transport chain. Compared with the extensive analysis of the effects of carbohydrates or amino acids on early embryo development, few data are available for oxygen consumption. However, oxygen consumption reflects embryo metabolism and viability more closely. Indeed, oxidative phosphorylation requires both oxygen and mitochondrial integrity. Oxygen uptake by mouse embryos is present from the 1-cell stage, and increases drastically at 

the blastocyst stage.72 A similar pattern is observed for bovine embryos, although the consumption of oxygen during oocyte maturation is as high as that during blastocyst formation.

OTHER MARKERS OF EMBRYO VIABILITY HUMAN CHORIONIC GONADOTROPIN

The culture of embryos up to the blastocyst stage is a means of improving embryo selection prior to transfer, as it is considered that grossly abnormal embryos will arrest their development before reaching the blastocyst stage. Blastocyst embryos are also more metabolically active. Dokras et al73 therefore tested the secretion of human chorionic gonadotropin (hCG) by individual postcompaction stage human embryos. Unfortunately, although secretion of hCG correlated well with blastocyst morphology, it could not be detected until day 8, and therefore is not detectable early enough to assist in the selection of embryos for transfer.73

CONCLUSIONS

Early embryo development is a complex process involving many different metabolic reactions that require the production of energy. Deficiencies in ATP production can be responsible for alterations in gene expression, chromosomal segregation, and signal transduction cascades. Alterations in the metabolic status of early embryos can therefore impair embryo viability. Embryo viability is dependent on both the oocyte from which it derives and in vitro culture conditions. In this respect intrafollicular and ooplasmic conditions may also have an important influence on the chromosomal status or cytoplasmic reserve necessary for the embryos to undergo the first cellular division and to survive in vitro conditions. Indeed, these conditions impose a great deal of cellular stress on preimplantation embryos. Although early cleavage stage embryos possess a great plasticity, their adaptation to in vitro conditions impairs their viability as reflected by

HPE_Chapter15.qxp

7/14/2007

5:54 PM

Page 187


lower cleavage rates, cleavage arrest, poor morphology, retarded genome activation, abnormal gene expression, or energy production. The more an embryo has to adapt, the less likely it is to be viable. Similar to the rat,74 in the mouse and human there is little vasculature in the vicinity of the implantation site for several hours. Glycolysis will therefore be the only available means of energy production for the blastocyst during this period.10 Analysis of variations in metabolic activity during the different stages of development could be of great help in the evaluation of embryo viability and its potential to form a neonate. In summary, different metabolic studies show that embryos are relatively metabolically quiescent prior to compaction, and oxidative phosphorylation is the major pathway for energy production. After activation of the embryonic genome, embryos are more active and a shift to energy production occurs with the suppression of phosphofructokinase inhibition, the enzyme that converts fructose 6-phosphate to fructose 1,6-diphosphate, or at the level of hexokinase, where it is converted into glucose 6-phosphate (Figure 15.1). This adaptation is thought to be necessary for embryos to survive in hypoxic conditions at the time of implantation, although oxidative phosphorylation is still active. The formation of the blastocoel cavity and the first two lineages are processes that require a large amount of energy. They are accompanied by a drastic increase in gene expression and protein synthesis.6,22,74 Pyruvate, glucose, and amino acid consumption and lactate production are all related to embryo viability. Oxygen, which is one of the most important substrates in energy production, has been relatively less studied. All experiments seem to agree that embryos with a relatively low metabolic activity, i.e. low level of lactate production and amino acid turnover, are those with the highest potential. Viable embryos that do not have to struggle to develop in vitro are probably those with a metabolic activity comparable with that of those developed in vivo.1,75 Indeed, transformation of high amounts of glucose into lactate is associated with a general suppression of respiration and oxidative phosphorylation (Crabtree-like effect)22

and modification of the redox status of the embryo. Glucose phosphorylation into glucose 6-phosphate by hexokinase depletes the stock of cellular and mitochondrial ATP. An excessive production of lactate decreases the level of NADH necessary for oxidative phosphorylation. Both mechanisms impair the function of the mitochondrial electron transport chain reactions, which in turn decreases the amount of energy available for biosynthetic processes.14,22 The ratio of ATP/ADP decreases, raising phosphofructokinase inhibition and resulting in the activation of the glycolytic pathway. This excessive production of lactate during the cleavage stage could therefore impair embryo viability, by altering mitochondrial energy production. However, the lack of prediction shown by metabolic measurements could be due to different factors. First, those performed during early cleavage stages represent oocyte competence more than that of the future embryos. Although correct oocyte maturation and preservation of the integrity of cellular components and metabolic pathways are important to sustain future development, it is probably not sufficient to predict viability of later stages. Second, carbohydrates and amino acids represent only a fraction of embryo metabolism. Little is known about lipid metabolism, which can be a valuable source of energy substrate. Third, the majority of nutrient uptake and production studies have been performed in suboptimal culture media. Further experiments should be performed in better adapted culture media, for example a sequential salt solution supplemented with human serum albumin, pyruvate, low level of glucose (0.5 mmol/l), so-called ‘non essential’ amino acids with glycine, taurine, and leucine for cleavage stage embryos and all amino acids with 1 mmol/l glucose for later embryo development. Such culture media should probably contain some growth factors, but this remains to be analyzed in detail. Fourth, human embryos display a large range of values for pyruvate, lactate, glucose, or amino acids, which renders it difficult to use a single value to predict viability. Finally, all the methods actually employed to measure metabolic activity – even if simplified – are too time consuming for routine use in an IVF program. 

HPE_Chapter15.qxp

7/14/2007

5:54 PM

Page 188


ACKNOWLEDGMENTS

I thank Professor Kate Hardy for welcoming me into her laboratory. With great patience and kindness, she trained me to handle preimplantation embryos in culture in vitro as well as in the different techniques used in this work. She was most generous with her time in teaching me indispensable analytical methods; her comments were always pertinent and most helpful. I am grateful to Professor Yvon Englert for enabling me to pursue my research successfully and his constant encouragement and many constructive discussions. The work was supported by the Belgian Funds for National Research. REFERENCES 1. Gardner DK, Sakkas D. Assessment of embryo viability: the ability to select a single embryo for transfer – a review. Placenta 2003; 24: S5–12. 2. Hardy K. Development of human blastocysts in vitro. In: Bavister BD, ed. Preimplantation Embryo Development. New York: Springer-Verlag, 1993: 184–99. 3. Khosla S, Dean W, Reik W et al. Culture of preimplantation emryos and its long-term effects on gene expression and phenotype. Hum Reprod Update 2001; 7: 419–27. 4. Pedersen ME, Ozdas OB, Farstad W et al. Effects of bovine oviduct epithelial cells, fetal calf serum and bovine serum albumin on gene expression in single bovine embryos produced in the synthetic oviduct fluid culture system. Reprod Fertil Dev 2005; 17: 251–7. 5. Lane M, Gardner DK. Selection of viable mouse blastocysts prior to transfer using a metabolic criterion. Hum Reprod 1996; 11: 1975–8. 6. Fleming TP, Kwong WY, Porter R et al. The embryo and its future. Biol Reprod 2004; 71: 1046–54. 7. Epstein C, Smith SA. Amino acid uptake and protein synthesis in preimplantation mouse embryos. Dev Biol 1973; 33: 171–84. 8. Houghton FD, Thompson JG, Kennedy CJ et al. Oxygen consumption and energy metabolism of the early mouse embryo. Mol Reprod Dev 1996; 44: 476–85. 9. Thompson JG, Bell ACS, Pugh PA et al. Metabolism of pyruvate by pre-elongation sheep embryos and effect of pyruvate and lactate concentrations during culture in vitro. Reprod Fertil Dev 1993; 5: 417–23. 10. Leese HJ. Metabolism of the preimplantation mammalian embryo. In: Oxford Review of Reproductive Biology, Milligan S.R. (Ed.). Oxford: Oxford University Press, 1991; 13: 35–72. 11. Goodall H, Johnson MH. The nature of intercellular coupling within the preimplantation mouse embryo. J Embryo Exp Morphol 1984; 79: 53–76. 12. Tarkowsk AK, Wroblewska J. Development of blastomeres of mouse eggs isolated at the 4- and 8-cell state. J Embryol Exp Morphol 1967; 18: 155–80. 13. Benos DJ, Balaban RS. Current topic: transport mechanisms in preimplantation mammalian embryos. Placenta 1990; 11: 373–80. 14. Leese HJ, Conaghan J, Martin KL et al. Early human embryo metabolism. Bioassays 1993; 15: 259–64. 15. Dale B, Menezo Y, Cohen J et al. Intracellular pH regulation in the human oocyte. Hum Reprod 1998; 13: 964–70.



16. Brinster RL. Carbon dioxide production from lactate and pyruvate by the preimplantation embryo. Exp Cell Res 1967; 47: 634–7. 17. Brinster RL. Carbon dioxide production from glucose by the preimplantation embryo. Exp Cell Res 1967; 47: 271–7. 18. Wales RG, Whittingham DG. The metabolism of specifically labelled lactate and pyruvate by two-cell mouse embryos. J Reprod Fertil 1973; 33: 207–22. 19. Biggers JD, Stern S. Metabolism of the preimplantation mammalian embryo. Adv Reprod Physiol 1973; 6: 1–59. 20. Leese HJ, Barton AM. Pyruvate and glucose uptake by mouse ova and preimplantation embryos. J Reprod Fertil 1984; 72: 9–13. 21. Gardner DK, Leese HJ. Assessment of embryo metabolism and viability. Handbook of In Vitro Fertilization. Boca Raton: CRC Press, 1993: 195–211. 22. Bavister BD. Culture of preimplantation embryos: facts and artifacts. Hum Reprod Update 1995; 1: 91–148. 23. Rieger D, Guay P. Measurement of the metabolism of energy substrates in individual bovine blastocysts. J Reprod Fertil 1988; 83: Z85–91. 24. Pinyopummintr T, Bavister BD. Energy substrate requirements for in vitro development of early cleavage-stage bovine embryos. Mol Reprod Dev 1996; 44: 193–9. 25. Fridhandler L, Wastila WB, Palmer WM. The role of glucose in metabolism of the developing mammalian preimplantation conceptus. Fertil Steril 1967; 18: 819–30. 26. Brinster RL. Radioactive carbon dioxide production from pyruvate and lactate by preimplantation rabbit embryo. Exp Cell Research 1969; 54: 205–9. 27. Gardner DK, Lane M, Batt P. Uptake and metabolism of pyruvate and glucose by sheep preattachment embryos developed in vivo. Mol Reprod Dev 1993; 36: 313–9. 28. Brison DR, Leese HJ. Energy metabolism in the late preimplantation rat embryo. J Reprod Fertil 1991; 93: 245–51. 29. Flood MR, Wiebold JL. Glucose metabolism by preimplantation pig embryos. J Reprod Fertil 1988; 84: 7–12. 30. Wales RG, Whittingham DG, Hardy K et al. Metabolism of glucose by human embryos. J Reprod F 1987; 79: 289–97. 31. Gardner DK, Leese HJ. The role of glucose and pyruvate transport in regulating nutrient utilization by preimplantation mouse embryos. Development 1988; 104: 423–9. 32. Hardy K, Handyside AH, Winston RML. The human blastocyst: cell number, death and allocation during late preimplantation development in vitro. Development 1989; 107: 597–604. 33. Gardner DK, Leese HJ. Non-invasive measurement of nutrient uptake by single cultured pre-implantation mouse embryos. Hum Reprod 1986; 1: 25–7. 34. Lowry OH, Passoneau JV. A flexible system of enzymatic analysis. New York: Academic Press, 1972. 35. Hardy K, Hooper MAK, Handyside AH et al. Non-invasive measurement of glucose and pyruvate uptake by individual human oocytes and preimplantation embryos. Hum Reprod 1989; 4: 188–91. 36. Leese HJ, Hooper MAK, Edwards RG et al. Uptake of pyruvate by early human embryos determined by a non-invasive technique. Hum Reprod 1986; 1: 181–2. 37. Martin KL, Hardy K, Winston RML et al. Activity of enzymes of energy metabolism in single human preimplantation embryos. J Reprod Fertil 1993; 99: 259–66. 38. Orsi NM, Leese HJ. Ammonium exposure and pyruvate affect the amino acid metabolism of bovine blastocysts in vitro. Reproduction 2004; 127: 131–40. 39. Houghton FD, Hawkhead JA, Humpherson PG et al. Non-invasive amino acid turnover predicts human embryo developmental capacity. Hum Reprod 2002; 17: 999–1005.

HPE_Chapter15.qxp

7/14/2007

5:54 PM

Page 189


40. Sturmey RG, Leese HJ. Energy metabolism in pig oocytes and early embryos. Reproduction 2003; 126: 197–204. 41. Pantaleon M, Kaye PL. Glucose transporters in preimplantation development. Rev Reprod 1998; 3: 77–81. 42. Rieger D, Loskutoff NM, Betteridge KJ. Developmentally related changes in the uptake and metabolism of glucose, glutamine and pyruvate by cattle embryos produced in vitro. Reprod Fertil Dev 1992; 4: 547–57. 43. McKiernan SH, Bavister BD, Tasca RJ. Energy substrate requirements for in-vitro development of hamster 1- and 2-cell embryos to the blastocyst stage. Hum Reprod 1991; 6: 64–75. 44. Cross PC, Brinster RL. The sensitivity of one-cell mouse embryos to pyruvate and lactate. Exp Cell Res 1973; 77: 57–62. 45. Biggers JD, Whittingham DG, Donahue RP. The pattern of energy metabolism in the mouse oocyte and zygote. Proc Natl Acad Sci USA 1967; 58: 560–7. 46. Conaghan J, Handyside AH, Winston RML et al. Effects of pyruvate and glucose on the development of human preimplantation embryos in vitro. J Reprod Fertil 1993; 99: 87–95. 47. Devreker F, Hardy K, Van den Bergh M et al. Non-invasive assessment of glucose and pyruvate uptake by human embryos after ICSI and during the formation of pronuclei. Fertil Steril 2000; 73: 947–54. 48. Roberts R, Franks S, Hardy K. Culture environment modulates maturation and metabolism of human oocytes. Hum Reprod 2002; 17: 2950–6. 49. Conaghan J, Hardy K, Handyside AH et al. Selection criteria for human embryo transfer: a comparison of pyruvate uptake and morphology. J Assist Reprod Genet 1993; 10: 21–30. 50. Hardy K, Robinson FM, Parascos T et al. Normal development and metabolic activity of preimplantation embryos in vitro from patients with polycystic ovaries. Hum Reprod 1995; 10: 2125–35. 51. Devreker F, Winston RML, Hardy K. Glutamine improves human preimplantation development in vitro. Fertil Steril 1998; 69: 293–9. 52. Turner K, Martin KL, Woodward BJ et al. Comparison of pyruvate uptake by embryos derived from conception and non-conception cycles. Hum Reprod 1994; 9: 2362–6. 53. Payne D, Flaherty SP, Barry MF et al. Preliminary observations on polar body extrusion and pronuclear formation in human oocytes using time-lapse video cinematography. Hum Reprod 1997; 12: Z32–41. 54. Gardner DK, Lane M, Stevens J et al. Noninvasive assessment of human embryo nutrient consumption as a measure of developmental potential. Fertil Steril 2001; 76: 1175–80. 55. Tiffin GJ, Rieger D, Betteridge KJ et al. Glucose and glutamine metabolism in pre-attachment cattle embryos in relation to sex and stage of development. J Reprod Fertil 1991; 93: 125–32. 56. Gardner DK, Lane M. Culture and selection of viable blastocyst: a feasible proposition for human IVF? Hum Reprod Update 1997; 3: 367–82. 57. Gardner DK, Leese HJ. Concentrations of nutrients in mouse oviduct fluid and their effects on embryo development and metabolism in vitro. J Reprod Fertil 1990; 88: 361–8.

58. Van den Bergh M, Devreker F, Emiliani S et al. Glycolytic activity: a possible tool for human blastocyst selection. Reproduction Biomedicine Online 2001; 3(Suppl1): 8. 59. Biggers JD. Reflections on the culture of the preimplantation embryo. Int J Dev Biol 1998; 42: 879–84. 60. Van Winkle LJ. Amino acid transport regulation and early embryo development. Biol Reprod 2001; 64: 1–12. 61. Miller JGO, Schultz GA. Amino acid content of preimplantation rabbit embryos and fluids of the reproductive tract. Biol Reprod 1987; 36: 125–9. 62. Leese HJ, Aldridge S, Jeffries KS. The movement of amino acids into rabbit fluid. J Reprod Fertil 1979; 56: 623–6. 63. Lane M, Gardner DK. Differential regulation of mouse embryo development and viability by amino acids. J Reprod Fertil Dev 1997; 109: 153–64. 64. Partridge RJ, Leese HJ. Consumption of amino acids by bovine preimplantation embryos. Reprod Fertil Dev 1996; 8: 945–50. 65. Gardner DK, Lane M, Spitzer A et al. Enhanced rates of cleavage and development for sheep zygotes cultured to the blastocyst stage in vitro in the absence of serum and somatic cells: amino acids, vitamins, and culturing embryos in groups stimulate development. Biol Reprod 1994; 50: 390–400. 66. Kane MT, Foote RH. Culture of two- and four-cell rabbit embryos to the expanding blastocyst stage in synthetic media. Proc Soc Exp Biol Med 1970; 133: 921–5. 67. Kishi J, Noda Y, Narimoto K et al. Block to development in cultured rat one-cell embryos is overcome using medium HECM-1. Hum Reprod 1991; 6: 1445–8. 68. Devreker F, Van den Bergh M, Biramane J et al. Effects of taurine on human embryo development in vitro. Hum Reprod 1999; 14: 2350–6. 69. Devreker F, Hardy K, Van den Bergh M et al. Amino acids decrease cell death in human embryos cultured in vitro. Hum Reprod 2001; 16: 749–56. 70. Donnay I, Partridge RJ, Leese HJ. Can embryo metabolism be used for selecting bovine embryos before transfer? Reprod Nutr Dev 1999; 39: 523–33. 71. Lamb VK, Leese HJ. Uptake of a mixture of amino acids by mouse blastocysts. J Reprod Fertil 1994; 102: 169–75. 72. Houghton FD, Leese HJ. Metabolism and developmental competence of the preimplantation embryo. Eur J Obstet Gynecol 2004; 115S: S92–6. 73. Dokras A, Sargentt IL, Ross C et al. The human blastocyst: morphology and human chorionic gonadotropin secretion in vitro. Hum Reprod 1991; 6: 1143–51. 74. Rogers PAW, Murphy CR, Rogers AW et al. Capillary patency and permeability in the endometrium surrounding the implanting rat blastocyst. Int J Microcirc Clin Exp 1983; 2: 241–9. 75. Harvey AJ, Kind KL, Thompson JG. REDOX regulation of early embryo development. Reproduction 2002; 123: 479–86.



HPE_Chapter15.qxp

7/14/2007

5:54 PM

Page 190

HPE_Chapter16.qxp

7/13/2007

5:26 PM

Page 191

16. Preimplantation embryo metabolism and embryo interaction with the in vitro environment Yves J R Ménézo and Pierre Guérin

INTRODUCTION

The goal of in vitro fertilization (IVF) and embryo culture is to provide high quality embryos capable of continued development and implantation, which will result in the birth of healthy babies. Considerable progress has been made in culturing preimplantation embryos since the initial studies were undertaken. We began to design and define new, more complex culture media in the early 1970s, which were based on the composition of genital tract secretions.1 During the initial stages of zygote formation and early cleavage divisions, the cells carry out only a minimal level of transcription, since early preimplantation development, i.e. up to the stage of maternal to zygotic transition (MZT), is maternally driven. A mature oocyte must contain a storage pool of proteins and/or mRNA transcripts in order to maintain its viability during these early stages: all of the enzymes required for metabolic pathways must be present and in harmony with the components of the culture medium. During and after the cycle of MZT, the longest cycle of preimplantation development, transcription of the new zygote genome then results in an increase in mRNA levels. The requirements of the embryo before and after MZT differ, and the environment, i.e. culture conditions, will have a direct impact on transcription and translation. Moreover, epigenetic reprogramming throughout early preimplantation development is also important, and this has generated concerns regarding the role of culture conditions in assisted reproductive technologies.2–6 In this chapter we describe the basis of embryo metabolism, and the impact of culture media composition on embryo quality and viability.

IN VITRO CULTURE CONDITIONS AND ENVIRONMENTAL FACTORS

The physical conditions used during in vitro culture differ significantly from conditions in vivo, in terms of light, variations in pH, pCO2/O2, temperature, static medium, etc. Physiological pH is regulated by a HCO3/CO2 buffer according to the equation pH = pKa ⫹ log (HCO3)/(CO2). This HendersonHasselbach equation allows pH to be calculated under a 5% CO2 atmosphere, in relation to the concentration of bicarbonate. The embryo has an alkaline pH (7.4), and is not able to deal with an acidic pH before the stage of MZT. The gas phase may be from 5% to 6.5% CO2 in air, or 5% CO2, 5% O2 and 90% N2. Based on the genital tract environment and an increased potential to decrease free radical formation, a reduced O2 atmosphere might appear to be more physiological. However, no clear-cut data support the superiority of reduced oxygen tension in human embryo culture in terms of ongoing pregnancy rates, particularly if the medium is well protected by the addition of agents that counteract the reactive oxygen species (ROS). Gas phase contaminants such as NO and CO are deleterious, as are volatile organic compounds (VOCs). The effect of ammonia is discussed in the paragraph describing amino acid metabolism. An osmolarity between 280 and 300 mosmol allows fertilization and early embryonic development. Redox potential is difficult to evaluate, and is therefore rarely considered. Based on oviduct and uterine secretions, redox potential should be calibrated to ⫺0.1 mV. In culture conditions, antioxidants can turn into pro-oxidants, this is the case for vitamin C, 

HPE_Chapter16.qxp

7/13/2007

5:26 PM

Page 192


especially in the presence of ferric ions. On the other hand, an excessively reducing redox potential may have harmful effects on protein tertiary structure. Glutathione is sometimes added to culture media as a reducing agent, but it should be noted that this cannot enter the embryo. The addition of serum destabilizes the culture media, by contributing enzymes as well as unknown metabolites and catabolites. Serum does not make a useful contribution, and may even be deleterious by potentially increasing pathologies, in particular those linked to imprinting.7,8

GLUCOSE AND ITS METABOLITES (FIGURE 16.1)

High levels of glucose have been considered to be a major factor in precipitating embryonic developmental arrest during in vitro culture. The present

Amino acids

G6PDH Glucose

trend towards removing glucose and phosphate from mammalian embryo culture media is not physiological, and merely replaces one artifact with another. The mouse model should be interpreted with care as glucose metabolism in rodents may be impaired due to a metabolic cul de sac that gives glucose poor entry into the TCA cycle, with a block at the level of glucose 6-phosphate isomerase that leads to a useless accumulation of glycogen. However, there is a species specific difference between rodents and humans, and the human zygote has a completely different feature in its enzyme activity: hexokinase activity is high, and glycogen synthase is low.9 The activity of the pentose phosphate pathway is equally high in human and in rodents. High levels of glucose also have a deleterious effect through an increase in free radical formation. This aspect is particularly obvious in diabetic mammals, and is more than probably the case in humans. Glucose itself is not toxic per se, and its metabolism is necessary for the synthesis of ATP.

Ribose Deoxyribose

Nucleic acids

Hexokinase Glycogen

Glucose 6-phosphate Gln

Glu

Glucosamine-6phosphate

Glycoproteins, glycolipids…

G6Pi Hexokinase Fructose 6-phosphate

Fructose

Amino acids

Fructose 1,6-diphosphate

Pyruvate

Amino acids

Tricarboxylic cycle

LDH

Lactate

CO2 + H2O + ATP

Figure 16.1 Glucose and fructose metabolism in the embryo. Glucose metabolism leads to the synthesis of nucleic acids and amino acids, and produces energy. G6Pi, glucose 6-phosphoisomerase; G6PDH, glucose 6-phosphate dehydrogenase; LDH, lactate dehydrogenase.



HPE_Chapter16.qxp

7/13/2007

5:26 PM

Page 193

EMBRYO METABOLISM AND INTERACTION WITH IN VITRO ENVIRONMENT

The problem arises when too much glucose is present in a medium that has an inadequate balance of substrates. This is the case for several metabolic pathways, including purine salvage in bovine (Figure 16.2), where excess of glucose also induces shifts in the sex ratio towards female10 as the female embryo has a more effective antiapoptotic mechanism. A relationship between glucose-related apoptosis and monozygotic twinning in humans has been proposed.11 Pyruvate, on the other hand, is an interesting compound, in that it acts not only as an energy source, but can also detoxify ammonia in the embryo, through transamination and export of alanine formed as a result (Figure 16.3). It also plays a role in protection against oxidative stress by preventing peroxide induced injury. Lactate is an end product of metabolism, and therefore adding it to culture media is of questionable value. In order to be re-introduced into metabolism, lactate must be converted (oxidized) to pyruvate, with generation of NADH. These short chain carboxylic acids require monocarboxylate transporters (MCTs) and their chaperone protein basigin to allow their entry into metabolic pathways; both of these are present in human oocytes and embryos.12

Glucose

HK

Glucose 6-phospate G6PDH G6PDH Pentose phosphate pathway

These C2/3 carbon chains are also used for the synthesis of lipids. Other glucose-derived metabolites (such as acetate, citrate, malate, etc.) are sometimes added to media, although no real impact on development has been established; they will be used for the synthesis of amino acids.

LIPID METABOLISM

The embryo needs to synthesize ‘sophisticated’ lipids at a very early stage. For this purpose, it is able to ‘pick up’ saturated and unsaturated fatty acids, free or bound to proteins, from its environment. The ratio of individual fatty acids is much more important than the presence of any single one. Phospholipid synthesis can be carried out with exogenous choline and glucose (metabolized toward fatty acid synthesis). In the mouse oocyte, cholesterol levels increase (three fold) from the 1-cell to the blastocyst stage.13 Cholesterol can be taken up from the environment, and it can be synthesized from the precursors mevalonate and lanosterol, as well as from acetate that is present as a result of pyruvate (and probably lactate) decarboxylation. Blocking the metabolic pathway for cholesterol synthesis with specific inhibitors such as compactine or diosgenine leads to developmental arrest before the blastocyst stage (Figure 16.4). Cholesterol and fatty acids bound to albumin can also be directly incorporated into the embryo.

1

AMINO ACIDS Purines 2

3 HPRT

Hypoxanthine

O2–

. Xanthine

Figure 16.2 Glucose and purine metabolism. Excess glucose will inhibit the purine salvage pathway and thus increase reactive oxygen species (ROS) formation. (1) Glucose inhibits hypoxanthine phosphoribosyl transferase (HPRT). (2) Purine catabolism. (3) Purine salvage. HK, hexokinase.

Amino acids are necessary very early after fertilization, and even for short-term embryo handling. Active amino acid synthesis is grafted onto the tricarboxylic acid (TCA) cycle. Amino acids are used for protein synthesis after translation of the mRNA stored during maturation, and again for messages that are translated after MZT. Moreover, there is an accelerated protein turnover under in vitro conditions. Microarray technology has demonstrated that expression of 114 genes is affected in culture without amino acids, versus 29 genes with affected expression in culture with amino acids.8 Glycine is present at a higher 

HPE_Chapter16.qxp

7/13/2007

5:26 PM

Page 194


Glucose glutamine Glutamine

Pyruvate pyruvate Oxaloacetate Acetyl CoA Malate

Lactate Glycine

Pyruvate

Glycine

Glutamate GPT

Glyoxylate

Alanine

TCA Cycle cycle NH3 Oxaloacetate GOT

Oxoglutarate

Aspartate

Alanine

Figure 16.3 Transamination reactions in the embryo. GPT, glutamate-pyruvate transaminase; GOT, glutamate-oxaloacetate transaminase.

Fatty acid synthesis

2 Acetyl CoA Acetoacetyl CoA β HMG-CoA NADPH2 β HMG-CoA reductase

Diosgenine CoA

NADP+

MEVALONATE CO2

NADPH2

CHOLESTEROL Figure 16.4 Cholesterol synthesis and its inhibition in the mouse embryo. Inhibition of this pathway will cause the embryo to arrest before blastocyst stage. HMG, hydroxymethyl-glutaryl.



concentration than any other amino acid in the female genital tract. It can reach 5 mmol, 10–50 times higher than the other amino acids that are present during the embryo’s transition through the tract. Glycine, glutamine, alanine, and taurine may act as organic osmolytes, allowing endogenous osmolarity and volume regulation, and preventing an eventual ‘salting out’ effect. Amino acid uptake takes place through active transport, and the overall scheme is complex. Different amino acids will compete for the same class of transporters, and individual affinities will allow a higher or lower incorporation. Moreover, some transporters are subject to a kinetic catabolism before genomic activation, while others appear after MZT. It is impossible to have a clear idea of the quantitative exchanges for each amino acid in the developing embryo, as the efficiency of uptake may differ widely, as is the case for glycine and methionine (Met). Glycine transport in the embryo is severely reduced in the presence of Met (Figure 16.5). It should also be noted that the Met concentration in vivo is much lower than the concentration of glycine. The concentrations of the different amino acids

HPE_Chapter16.qxp

7/13/2007

5:26 PM

Page 195


With methionine Without methionine Methionine

Glycine

Carrier Time (1) AA external concentration

Glycine uptake

AA uptake

250 µmol/l 100 µmol/l 50 µmol/l

Glycine Glycine free in proteins (2) AA competition

Amino acid pool

(3) Protein synthesis

(4) Protein catabolism

Export (ala)

Figure 16.5 Factors influencing the endogenous pool of amino acids (AA) in the embryo. (1) The external concentration of each amino acid affects its entry into the embryo. (2) Competition between amino acids for the same carriers will facilitate the entry of some and reduce uptake of others. (3) Protein synthesis will decrease the endogenous pool. (4) Protein turnover will partly increase the endogenous pool.

found in vivo are probably an important consideration for the composition of culture media. Several points requiring clarification are discussed below. ESSENTIAL AMINO ACIDS AND SULFUR AMINO ACIDS

It has been suggested that ‘essential amino acids’ may be toxic during preimplantation development before the stage of MZT.14 However, the distinction between essential and non-essential amino acids is somewhat unclear, as some amino acids can be produced from others. Phenylalanine, valine, tryptophan, threonine, isoleucine, methionine, histidine, arginine, lysine, leucine, cysteine, and tyrosine were described by Waymouth in a particular context (diploid cell culture).15 Leucine uptake is considered to be a good marker of embryo quality before genomic activation (MZT), indicating that this ‘essen-

tial amino acid’ may be important for human embryo construction.16 Methionine is classified as an essential amino acid and is thus omitted in the majority of first phase sequential media. This is a questionable concept, because in mouse embryos, silent paternal alleles of H19, Igf2, Grb10, and Grb7 are aberrantly expressed and hypomethylated in simple media, but not in media with amino acids. In reality, the balance between Met and the other amino acids is of major importance. Met has such a high affinity for the transporter molecules that if it is present in too high a concentration, it may prevent the uptake of other amino acids, thus disrupting the equilibrium so that the endogenous pool becomes unbalanced. Met is required for the initiation of all protein synthesis through Met-tRNA (Figure 16.6), and is incorporated by mouse, bovine, and human embryos. Equally important, Met is the fuel of methylation through the 

HPE_Chapter16.qxp

7/13/2007

5:26 PM

Page 196


Ammonia detoxification

4

1

Methionine

Methionyl tRNA (starts protein synthesis)

Trans-sulfuration pathway

Homocysteine

3

Cysteine

Glutathione

2

Oxydative stress ROS

DNA,

Lipid

Peroxides

Me Imprinting DNA methyltransferase

APOPTOSIS

SAM

Figure 16.6 Principal methionine metabolic pathways in the embryo. (1) Initiation of protein synthesis through methionyl tRNA. (2) Methylation pathway through the synthesis of S-adenosyl methionine (SAM). Homocysteine has an inhibitory effect as it prevents the entry of methionine and inhibits SAM synthesis.17 (3) Methionine can lead to cysteine formation and thus allow glutathione synthesis, preventing reactive oxygen species (ROS)-linked decays. (4) Methionine increases the release of alanine, thus facilitating ammonia detoxification.

synthesis of S-adenosyl methionine (SAM). SAM synthesis occurs before genomic activation in the mouse and the human embryo.17 SAM and other folic acid (the other methylation agent) substrates are the critical epigenetic regulators that can affect DNA methylation and imprinting. Methyltransferase activity will also affect gene silencing through histone methylation, with subsequent cytogenetic problems and altered DNA methylation, including CpG methylation associated with Prader-Willi syndrome (PWS). Moreover, it should be clearly noted that CpG methylation in this PWS-1C site occurs after fertilization in human embryos, whereas in the mouse PWS-1C methylation occurs during oogenesis.18 Once again it must be emphasized that the ‘mouse embryo model’ may be misleading for human embryo



culture. In human ART, the Beckwith-Wiedemann syndrome appears to be age dependent, related to culture in media without Met.5,19 ‘Superovulation’ very probably disturbs the endogenous pool of the oocytes, including the pool of amino acids such as Met. Homocysteine in the embryo’s environment is deleterious, as it induces efflux of Met from the embryo, with an impact on the process of methylation. Met can also be converted to cysteine via the trans-sulfuration pathway, and cysteine is then a ‘partner’ in the synthesis of glutathione and hypotaurine, two free radical scavengers of major importance in vivo. Met restriction induces apoptosis with formation of nucleosomal DNA fragments, and this is mitochondria dependent, as mitochondrial metabolism is the major source of free radical formation.

HPE_Chapter16.qxp

7/13/2007

5:26 PM

Page 197


In conclusion, Met is involved in numerous regulatory events with multiple complex interactions in the embryo, as early as the 1-cell stage. TOXICITY OF AMMONIA AND DEGRADATION OF GLUTAMINE

Ammonia is produced through amino acid catabolism, and also as a result of spontaneous degradation in vitro. In conditions of open in vitro culture, ammonia is mainly in the form of ammonium bicarbonate, which is highly unstable at a pH of 7.2–7.4; ammonia is then released into the CO2 atmosphere, and does not accumulate. For this reason, the toxicity of ammonia is highly controversial,20,21 if ammonia is indeed released from amino acids. Adding of NH4Cl to culture media has recently been proposed as a model to mimic the toxicity of ammonium ions, but NH4Cl is an acidic and highly stable salt; therefore its addition to culture media cannot correspond to the in vitro situation. Furthermore, the embryo can remove ammonia

through transamination, with the formation and release of alanine into the culture medium. Accumulation of ammonia and its related toxicity may occur in very specific and inappropriate culture conditions (poor regulation of pH, etc.). Finally, degradation of glutamine produces ammonia and glutamic acid, but this process may also form pyrrolidone carboxylic acid (PCA) or pyroglutamic acid. Normally, glutamine degradation is weak at the neutral/slightly alkaline pH of the culture media. It is not certain that replacing glutamine with alanylglutamine or glycyl-glutamine is an answer to this problem, as they may also easily form glyPCA or alaPCA, a common rearrangement for N-terminal glutamines (Figure 16.7). In addition, albumin binds numerous short chain peptides, which are probably used as a marginal source of amino nitrogen. In summary, the embryo appears to require all of the amino acids as early as the 1-cell stage, especially for ‘older’ female patients in whom the endogenous pool of mRNAs, proteins, and substrates may already be compromised. However, including all of

O

CH2

CH2

C

CH HN

O H2N

C

C

OH

NH3

+

O

Pyroglutamic acid (pyrrolidone carboxylic acid) CH2

CH2

H2N

CH

C

OH O

O Glutamine

HO

C

CH2

CH2

H2N

CH

C

OH

O Glutamic acid

Figure 16.7 Transamination reactions in the embryo. Ammonia is eliminated through transamination of pyruvate and export of alanine.



HPE_Chapter16.qxp

7/13/2007

5:26 PM

Page 198


them in culture media at the same concentration is a highly simplistic view, bearing in mind the complex interactions and potential pathways described above; a referral to the in vivo situation is useful.

Lipid soluble vitamins (A and E) are natural potent antioxidants, and introducing them into culture media is highly problematic unless their binding capacity to defatted albumin is used.

PRECURSORS OF RNA AND DNA

MACROMOLECULES INCLUDING GROWTH FACTORS

Transport systems for purine and pyrimidine bases and their de novo synthesis are active even before the stage of genomic activation, and the purine salvage pathway is active in early embryos. During the first few cell cycles, the total RNA content of the embryo does not increase; inhibiting mRNA synthesis with alpha amanitin leads to developmental arrest at the time of genomic activation, and inhibiting DNA synthesis also causes developmental arrest. An endogenous pool of A/T/G/C triphosphate is mandatory for DNA repair activity, which is of major importance during early embryogenesis. DNA adducts are mutagenic and clastogenic, causing damage to chromosomes. Unrepaired or incorrectly repaired DNA leads to compound damage, with errant transcripts, cell death and/or mutagenesis leading to malignant transformation. An adequate supply of ATP is also mandatory for mRNA polyadenylation, which is necessary for regulation of translation.

VITAMINS

Several animal experiments have suggested that the addition of water soluble vitamins to culture media may be beneficial, but a positive role has not been unequivocally demonstrated. Ascorbic acid is a potent antioxidant. It reduces the redox potential via the evolution of cysteine and Met, and the stable modification of lipoproteins by dehydroascorbate also increases resistance to divalent cation-induced oxidation. However, its addition should be handled with care, since vitamin C may also act as a pro-oxidant. Folic acid is another parameter to be considered. Reduced folic acid is a methyl donor (as Met through S-adenosyl methionine), and this involves the synthesis of thymidine in particular. There is no evidence that folic acid supplementation in vitro is necessary. 

Specific embryotrophic factors secreted by the genital tract have not been described. Serum is not necessary for IVF culture;22 its presence affects mRNA content, and although it has not been demonstrated in human embryo culture, in the bovine system there is evidence that serum is involved in producing the large calf syndrome. Serum albumin is not absolutely necessary for embryo culture, but it is always added to culture media since it was first observed that, a minima, it facilitates embryo handling. The beneficial role of albumin is not fully understood. It binds several compounds of various molecular weights, such as lipids, amino acids, peptides, catecholamines. It may be incorporated directly into the embryo and thus act as a carrier of nutrients. Addition of growth factors to embryo culture media is still a matter of controversy (Figure 16.8). Growth factors are present in the female genital tract, and the corresponding receptors are present on the embryos, sooner or later. They may act through a balance between stimulatory and inhibitory effects. Growth hormone is present in the female tract, and a receptor is present in mouse, bovine, and human embryos; therefore it seems to have a positive effect on embryonic development. In human embryos, granulocyte macrophage colony stimulating factor (GM-CSF) seems to have both a stimulatory and an epigenetic regulatory impact. Insulin is usually added for the second phase of culture, but whether or not it has the same effect as insulin growth factor 1 (IGF1) is questionable. Embryos do synthesize growth factors after genomic activation, but it is not clear whether they can have a type of ‘autostimulation’ effect through an internal loop. Mouse and bovine embryo development is significantly better when embryos are cultured in groups rather than alone.23,24 This suggests that diffusible paracrine/autocrine factors could be

HPE_Chapter16.qxp

7/13/2007

5:27 PM

Page 199


EGF

TNFa

TGFβ

+ –

8-cell

+

TGFβ*, bFGF, Insulin, GH

Compacted morula

+

Blastocyst

TGFβ*, bFGF, Insulin, EGF, PDGF (c-myc), GH, PAF

–

+ +

TGFβ*, LIF (?), Insulin, PDGF (c-myc)

Hatching EGF, LIF (?), TGFβ, PAF

Figure 16.8 Growth factors and preimplantation development. The impact of growth factors may differ from one species to another, depending upon the presence of the corresponding receptor. Growth hormone (GH) receptor is present in mouse, bovine, and human embryos. TNF␣, tumor necrosis factor ␣; EGF, epidermal growth factor; TGF␤, transforming growth factor ␤; bFGF, basic fibroblast growth factor; PDGF, platelet derived growth factor; PAF, platelet activating factor; LIF, leukemia inhibitory factor; *hyaluronic acid may act synergistically.

partly responsible for the regulation of early embryo development, at least in vitro. For the mouse embryo, growth factors of the transforming growth factor family seem to be involved;23 platelet activating factor (PAF) could be an intermediate mediator in bovine.24 The positive ‘group effect’ has not been confirmed in human systems.

CULTURE TECHNIQUES: SEQUENTIAL MEDIA, MICRODROPS, AND OIL

Culture in microdrops under a layer of mineral oil is often recommended, on the basis of mimicking the in vivo situation, and following the mouse model for embryo culture. Although the volume of liquid surrounding the embryo in vivo is minimal, or virtual, the embryo is in contact with an epithelial cell layer, with a film of liquid that is permanently renewed. This is not the case for microdrop culture, and there are three features to be considered: (1) Metabolic waste from the embryo may accumulate at a higher concentration, due to the suboptimal volume of liquid.

(2) The microdrop surface area allows a maximum exchange with oil, so that lipid soluble compounds in the medium can be absorbed at a maximal level by the oil, thus depriving the embryo of these compounds. (3) Similarly, any micropollution of the oil with compounds that are water soluble will pass into the culture medium with a higher efficacy in a microdrop system. Therefore, a volume of 150–350 microliters overlaid with oil seems a better compromise, as a means of maintaining the pH and temperature, reducing evaporation, and minimizing potential micriobial exposure during observation and handling. The merit of sequential media is that the environment follows the metabolic needs of the embryos before and after genomic activation, potentially creating a more ‘physiological’ situation. Clearly, before MZT the medium must be more protective against ROS as there is a decrease in internal protection. However, human embryos have been shown to have a significant ‘plasticity’ and ability to adapt in vitro, and culture from fertilization to blastocyst formation can still be achieved with single media. 

HPE_Chapter16.qxp

7/13/2007

5:27 PM

Page 200


DISCUSSION AND CONCLUSION 6.

The ‘quality’ of an embryo is basically determined by the gametes (quality of sperm, maternal age, etc.), and the process of fertilization in vitro does not have precise and specific culture media requirements. However, it is clear that embryo culture after fertilization does have more complex requirements. The idea that the in vivo environment is better than in vitro conditions does not always correspond to reality. To evaluate new techniques and/or new culture media, in vitro grown embryos must be transferred to the uterus in order to avoid misleading observations based upon morphology alone, which has been clearly demonstrated to have its limitations. The significance of specific animal models for human assisted reproductive technologies (ART) is questionable. Although the mouse model has been useful, due to significant and specific differences in centrosome transmission, speed of development, metabolic pathways, etc., it is not a good model for human culture. Prior to and during culture, it is important to be aware of, and to control the interactions of different compounds with each other and with the gas phase. In view of the fact that some embryos arrest early in development, embryo transfer at early stages is at times a ‘blind’ process. However, later transfer at the blastocyst stage is also faced with the problem of ensuring that the blastocysts are viable and healthy after their period of extended in vitro culture.

7.

8.

9. 10.

11. 12.

13. 14.

15.

16.

17. 18.

19.

20. REFERENCES 1. Ménézo Y. Milieu synthétique pour la survie et la maturation des gamètes et pour la culture de l’œuf fécondé. C R Acad Sci Paris Ser D 1976; 282: 1967–70. 2. Khosla S, Dean W, Reik W et al. Culture of preimplantation embryos and its long-term effects on gene expression and phenotype. Human Reprod Update 2001; 7: 419–27. 3. El-Maarri O, Buiting K, Peery EG et al. Maternal methylation imprints on human chromosome 15 are established during or after fertilization. Nat Genet 2001; 27: 341–4. 4. Niemitz EL, Feinberg AP. Epigenetics and assisted reproductive technology: a call for investigation. Am J Hum Genet 2004; 74: 599–609. 5. Gicquel C, Gaston V, Mandelbaum J et al. In vitro fertilization may increase the risk of Beckwith-Wiedman syndrome related to the



21.

22.

23.

24.

abnormal imprinting of KCN1OT gene. Am J Hum Genet 2003; 72: 1338–41. De Baun MR, Niemitz ML, Feinberg AP. Association of in vitro fertilisation with Beckwith-Wiedemann syndrome and alterations of LT1 and H19. Am J Hum Genet 2003; 72: 156–60. Fernandez-Gonzalez R, Moreira P, Bilbao A et al. Long-term effect of in vitro culture of mouse embryos with serum on mRNA expression of imprinting genes, development, and behavior. Proc Natl Acad Sci USA 2004; 101: 5880–5. Rinaudo P, Schultz RM. Effects of embryo culture on global pattern of gene expression in preimplantation mouse embryos. Reprod 2004; 128: 301–11. Chi MMY, Manchester JK, Yang V et al. Contrast in levels of metabolic enzymes in human and mouse ova. Biol Reprod 1988; 39: 295–307. Jimenez A, Madrid-Bury N, Fernandez R et al. Hyperglycemia-induced apoptosis affects sex ratio of bovine and murine preimplantation embryos. Mol Reprod Dev 2003; 65: 180–7. Menezo Y, Sakkas D. Monozygotic twinning: is it related to apoptosis in the embryo? Hum Reprod 2002; 17: 247–8. Herubel F, El Mouatassim S, Guérin P et al. Genetic expression of monocarboxylate transporters during human and murine oocyte maturation and early embryonic development. Mol Hum Reprod 2002; 10: 175–81. Pratt HPM. Preimplantation mouse embryo synthesize membrane sterols. Dev Biol 1982; 89: 101–10. Lane M, Hooper K, Gardner DK. Effect of essential amino acids on mouse embryo viability and ammonium production. J Assist Reprod Genet 2001; 18: 519–25. Waymouth C. Construction of tissue culture media. In: Rothblat GH, Cristofalo VJ, eds. Growth, Nutrition and Metabolism of Cells in Culture. New York: Academic Press, 1972; 1. Brison DR, Houghton FD, Falconer D et al. Identification of viable embryos in IVF by non-invasive measurement of amino acid turnover. Hum Reprod 2004; 19: 2319–24. Ménézo Y, Khatchadourian C, Gharib A et al. Regulation of S-adenosyl methionine synthesis in the mouse embryo. Life Sci 1989; 44: 1601–9. El-Maarri O, Buiting K, Peery EG et al. Maternal methylation imprints on human chromosome 15 are established during or after fertilization. Nat Genet 2001; 27: 341–4. Diaz-Meyer N, Day CD, Khatod K et al. Silencing of CDKN1C (p57KIP2) is associated with hypomethylation at KvDMR1 in BeckwithWiedemann syndrome. J Med Genet 2003; 40: 797–801. Lane M, Gardner DK. Ammonium induces aberrant blastocyst differentiation, metabolism, pH regulation, gene expression and subsequently alters fetal development in the mouse. Biol Reprod 2003; 69: 1109–17. Summers MC, Biggers JD. Chemically defined media and the culture of mammalian preimplantation embryos: historical perspective and current issues. Hum Reprod Update 2005; 9: 557–82. Ménézo Y, Testart J, Perone D. Serum is not necessary in human in vitro fertilization, early embryo culture, and transfer. Fertil Steril 1984; 42: 750–5. Paria BC, Dey SK. Preimplantation embryo development in vitro: cooperative interactions among embryos and the role of growth factors. Proc Natl Acad Sci USA 1990; 87: 4756–60. Gopichandran N, Leese HJ. The effect of paracrine/autocrine interactions on the in vitro culture of bovine preimplantation embryos. Reprod 2006; 131: 269–7.

HPE_Chapter17.qxp

7/13/2007

5:29 PM

Page 201

17. Polar body chromosome abnormalities and their consequences for human embryo development Anver Kuliev and Yury Verlinsky

INTRODUCTION

The fact that chromosomal abnormalities in the female gamete originate predominantly from errors in meiosis is well known. Studies using DNA polymorphism performed in families with aneuploid spontaneous abortions or liveborn babies with common chromosomal syndromes demonstrate that these abnormalities derive mainly from meiosis I.1–3 It has been suggested that the age-related increase in the incidence of common trisomies is probably determined by an age-related reduction in meiotic recombination, resulting in a premature separation of bivalents and chromosomal non-disjunction. Meiosis II errors may also be related to errors involving an increased rate of meiotic recombination during meiosis I, which may result in a failure of bivalents to separate.4 Fluorescence in situ hybridization (FISH) technology can be used to analyze biopsied first (PB1) and second polar bodies (PB2) in order to test the outcome of first and second meiotic divisions directly, as described elsewhere.5 PB1 represents a by-product of meiosis I, and is extruded following oocyte maturation; PB2 is a by-product of meiosis II, extruded following exposure of oocytes to sperm or ICSI. As described below, the frequency and types of chromosomal errors detected by this approach differ from those described in traditional studies of metaphase II (MII) oocyte meiotic chromosomes; the latter suggested that the majority of chromosomal anomalies in oocytes originate from errors in whole bivalents as a result of chromosomal nondisjunctions.6 In contrast, direct testing of meiotic outcome using PB1 and PB2 analysis showed not

only a higher prevalence of meiotic errors, but also that a significant proportion of errors are in chromatids, rather than chromosomes. This discrepancy may be due to the technical difficulties involved in obtaining good quality meiotic chromosome preparations for the earlier studies, and also to a failure to test the corresponding chromosome set extruded in PB1. The resulting oocyte karyotype cannot be reliably evaluated without testing the corresponding chromosome set, particularly in cases where chromosomes or chromatids are missing. This was demonstrated by simultaneously testing MII oocytes with their corresponding PB1, which showed that the normal chromosome pattern is represented by paired fluorescent signals for each chromosome; the absence, or addition of one or both signals in either the oocyte or PB1 reflects a pattern that is the exact opposite of that in the corresponding MII oocytes or PB1. This suggests that PB1 testing provides an accurate prediction of the oocyte genotype.7–11 Based on the above observations, PB1 testing was applied clinically demonstrating that this approach is of practical relevance for IVF patients of advanced reproductive age12–16 and provides a means of improving the efficiency of assisted reproduction technology (ART). Aneuploidy-free oocytes and embryos with the highest development potential can thus be preselected, as an alternative to preselection relying on morphological criteria that cannot exclude aneuploid embryos from transfer.17–21 The data also demonstrated that the genotype of the resulting zygotes could not be accurately predicted without information about the outcome of the second meiotic division, which may be inferred from PB2 

HPE_Chapter17.qxp

7/13/2007

5:29 PM

Page 202


testing. Approximately 15 000 oocytes have so far been analyzed by FISH, and the accuracy of evaluating the oocyte karyotype by testing PB1 and PB2 has been demonstrated. This strategy also provides an attractive approach for studying the origin of human aneuploidies.

FREQUENCY AND TYPES OF CHROMOSOMAL ABNORMALITIES IN POLAR BODIES

PB1 and PB2 analysis allows errors in meiosis to be directly tested. Following fertilization, the two polar bodies are removed simultaneously, fixed on the same slide, and analyzed by FISH.5 Both polar bodies are extruded from oocytes as a normal process during maturation and fertilization, and therefore their removal is not expected to have any biological effect on the development of the embryo; this has been confirmed by the fact that hundreds of normal pregnancies have resulted following PB1 and PB2 biopsy.22 Biopsied and fixed PB1 and PB2 were studied using fluorescent probes specific for chromosomes 13, 16, 18, 21, and 22 (Abbott Laboratories, Downers Grove, IL). Analysis of 8382 oocytes obtained in 1297 IVF cycles indicated that a total of 3509 (52.1%) of 6733 oocytes with the results were aneuploid, of which 965 (27.5%) had errors in both PB1 and PB2, 1467 (41.8%) in PB1 only, and 1077 (30.7%) in PB2 only.23 Errors were found in 2432 out of 5831 (41.7%) PB1 with results, compared with 2042 errors detected in 5808 (35.2%) PB2. The average maternal age of IVF patients from whom the oocytes were obtained was 38.5 years, and therefore these figures may represent an overestimate within the overall IVF patient population. Of the oocytes analyzed, 3630 were tested using tri-color probes specific for chromosomes 13, 18, and 22. Five-color probes specific for chromosomes 13, 16, 18, 21 and 22 were used to test 3103 oocytes, of which 1974 were detected as aneuploid (63.6%). Missing signals (nullisomy) was detected at a frequency at least three times higher than extra signals (disomy) in PB1; PB2 analysis showed a comparable 

distribution of missing and extra signals. PB1 data also showed a 63.5% chromatid error rate (48.1% missing and 15.4% extra chromatids), compared with a 6.4% chromosome error rate (5.9% missing and 0.5% extra chromosomes). Therefore, as with chromatid errors, missing chromosomes were more frequent than extra chromosomes (8.3% and 0.7%, respectively). This suggests that MII oocytes might retain extra chromatid or chromosome material, which is consistent with data obtained from postimplantation spontaneous abortions, in which a higher frequency of trisomies than monosomies is observed. Only 15.9% of PB1 abnormalities were disomies, compared with 54% nullisomies – the remainder were of complex origin. Although the observed excess of missing signals in PB1 may also be attributable to technical errors such as hybridization failure, it is also possible that there is a mechanism in meiosis I that prevents extrusion of extra chromosome material into PB1 if meiotic errors occur during the process of oocyte maturation. A significant proportion of abnormalities in PB1 and PB2 (30.1% and 32.1%, respectively) were of complex origin, with different types of errors, or errors in different chromosomes. Of 1582 complex abnormalities, 1242 (78.5%) involved two or more chromosomes simultaneously, and 340 (21.5%) involved the same chromosome(s) in both PB1 and PB2. Both polar bodies were found to be abnormal in 14.3% oocytes, of which approximately half have different and half the same chromosomes involved. Although one third of zygotes resulting from the oocytes with the same chromosomes aneuploidy in the first and second polar body appeared to be balanced, which may be similar to the phenomenon of aneuploidy rescue, incidental abnormalities in other chromosomes cannot be excluded in these cases, and therefore the preselection and transfer of the embryos resulting from such oocytes has yet to be justified. The chromosomes involved most frequently in meiotic errors were chromosomes 21 and 22 (10.9% and 11.8%, respectively), deriving comparably from meiosis I and meiosis II. Chromosomes 13, 16, and 18 were less frequently involved in meiotic errors (6.0%, 6.4%, and 6.8%, respectively), and their

HPE_Chapter17.qxp

7/13/2007

5:29 PM

Page 203

POLAR BODY CHROMOSOME ABNORMALITIES AND EMBRYO DEVELOPMENT

error patterns were not similar. Errors in chromosome 16 originated predominantly in meiosis II (52.4% meiosis II vs 26.2% meiosis I), in contrast to chromosome 13 and 18 errors, which originated more frequently from meiosis I (46.6% and 60.2% vs 36.2% and 29.7% in meiosis II, respectively). The data show that testing for only five chromosomes reveals an aneuploidy rate as high as 52%. Although limitations of the FISH technique may result in an overestimate, the majority of PB1 and PB2 abnormalities were confirmed to be true errors by follow-up studies of the embryos resulting from oocytes with errors in meiosis I and meiosis II.13 In contrast to the well-established concept that chromosome abnormalities originate from female meiosis I, our results show that the errors observed originate from both meiosis I and II, in accordance with the expected patterns of segregation.24 Only half of the abnormalities that are derived from the second meiotic division may be detected by PB1 analysis as complex errors, and therefore testing of both polar bodies is still required in order to avoid the transfer of embryos resulting from aneuploid oocytes. Since, as mentioned previously, PB1 and PB2 are extruded as a normal process during oocyte maturation and fertilization, and have no biological significance in pre- and postimplantation development, their removal and testing may become a useful tool for identifying aneuploidy-free oocytes in assisted reproduction practice. As mentioned, this should help in the preselection of oocytes and embryos with the highest potential for establishing a viable pregnancy, significantly improving the efficiency of IVF. Previous studies suggest that the majority of abnormalities in meiosis I are due to chromatid errors, and not chromosomal non-disjunction.25 However, not all MII oocyte abnormalities are due to chromatid errors; chromosomal errors are still observed in 6.4% of oocytes.23 It therefore seems likely that errors in both chromatids and chromosomes are involved in producing MII abnormalities, although the frequency of chromatid errors is much higher than the freuquency of chromosomal errors. Certainly both types of errors in meiosis I can lead to aneuploidy in the resulting embryos; this has been

confirmed by follow-up studies of non-transferred embryos resulting from these oocytes. However, whether the types of errors have different effects on pre- and postimplantation development is still not known. The data suggest that PB1 testing will detect approximately 41.8% of abnormal oocytes deriving from meiosis I. This may also allow a significant proportion of oocytes with second meiosis division errors to be predicted, based on the fact that almost half of the oocytes with meiosis II errors also had PB1 aneuploidies. The other half could not have been predicted by PB1 analysis, as they became abnormal only following the second meiotic division. Therefore, in order to identify all embryos originating from oocytes with chromosomal abnormalities, the outcome of both the first and second meiotic divisions should be studied, by analyzing both polar bodies. Although testing for additional chromosomes might be expected to increase the rate of aneuploidy observed, available data indicate that this strategy results in an increased detection of complex abnormalities, rather than an increase in the overall aneuploidy rate.15 Studies in XO female mice demonstrated that a meiotic error in one chromosome may affect the segregation of other chromosomes,26 and this was also observed in our follow-up of meiosis I errors through meiosis II and cleavage of the resulting embryos, as described below. The results are consistent with the mouse data, in that chromatid errors represent the majority of aneuploidies. Even allowing for the fact that some of the errors are attributable to technical factors, a more than twofold difference was observed between missing and extra chromatids in the resulting MII oocytes. This may indicate that extra chromosome (chromatid) material is maintained in the oocyte when the meiosis I errors originate, rather than being extruded. In contrast, there was no difference in missing or extra chromatid error rates following meiosis II (14.9% and 14.2%, respectively), suggesting that different mechanisms are involved in generating errors during the two meiotic processes. It is of interest to note that more than one-third of oocytes with meiosis I errors have sequential 

HPE_Chapter17.qxp

7/13/2007

5:29 PM

Page 204


errors in meiosis II, representing approximately half of meiosis II errors overall. Follow-up of the outcome of meiosis I errors through meiosis II showed that almost one-third (32.5%) of meiosis I errors appeared to result in apparently euploid zygotes, following a sequential error in meiosis II. Although the mechanism whereby such self-corrected (balanced) zygotes may be formed is not yet understood, it may be similar to the well-known phenomenon of ‘trisomy rescue’ in postzygotic embryo development, a mechanism that may result in uniparental disomy and imprinting disorders. As will be described below, the observed aneuploidy rescue mechanism in female meiosis cannot ensure that the resulting embryos will be chromosomally normal and useful for embryo transfer.

RELATIONSHIP BETWEEN CHROMOSOMAL ABNORMALITIES IN POLAR BODIES AND EMBRYOS

As described above, approximately half of meiosis II errors are observed in oocytes that also had prior errors in meiosis I. If preceding errors in meiosis I and meiosis II have no effect on subsequent mitotic divisions in the resulting zygote, sequential errors may result in almost one-third of these zygotes being considered as normal (euploid). These apparently euploid zygotes may develop into chromosomally normal embryos. A series of 100 embryos deriving from meiosis II-rescued zygotes were followed up at the cleavage stage, and this showed that only 18% were euploid for the five chromosomes analyzed; the remaining 82% had chromosomal abnormalities.28–30 All of the embryos that were euploid for all five chromosomes tested apparently resulted from zygotes with only one chromosome error rescued, and none resulted from zygotes that were balanced for two or more chromosomes. This might suggest that the observed sequential errors in female meiosis may be due to an abnormality of the meiotic apparatus overall, rather than to a single chromosome segregation defect leading to a general defect of the mitotic apparatus in resulting embryos. This concurs with 

the types of aneuploidies detected in the resulting embryos, which in the majority of cases are represented by complex errors, including mosaicism, known to be common in cleavage stage embryos.17,18 The average reproductive age of these PB biopsy patients was approximately 38.5 years, and therefore the observed genomic instability in mitotic divisions of apparently balanced zygotes following meiosis II rescue may also be age related. Although the mechanisms that might be affected by increasing age are not known, the underlying mechanisms of the aging process do involve increasing errors in the mitotic machinery of dividing cells, with resulting chromosomal abnormalities. It has also been suggested that deviations in cytoplasmic organization, such as mitochondrial distribution, may reduce the meiotic competence of oocytes and predispose the embryos to abnormalities during cleavage.31–33 Prospective analysis of pronuclear zygote morphology in relation to chromosomal abnormalities detected in preimplantation genetic diagnosis (PGD) cycles carried out for poor prognosis IVF patients suggests that the relationship between these cytoplasmic changes and nuclear organization during maturation and fertilization of oocytes may determine the course of potentially abnormal development due to mitotic errors at the cleavage stage.34 Data gathered from PGD cycles carried out to diagnose aneuploidies at the cleavage stage indicate that at least 60% of embryos tested have chromosomal abnormalities.17,18,35 Although the types of aneuploidies reported may differ in different studies, there seems to be no doubt that mosaicism is responsible for approximately half of these abnormalities. Information about the initial set of chromosomes in zygotes that led to mosaic embryos was not available in any of these studies, and therefore despite its high prevalence and potential clinical relevance, the nature of mosaicism in preimplantation embryos is not known. Indirect observations suggest that different types of mosaicism may be observed at the cleavage stage. Some types of mosaicism may increase with maternal age,36 probably stemming from errors in female meiosis, and others are possibly attributable to immature centrosome structures in sperm. An active centrosome is required from the first mitotic

HPE_Chapter17.qxp

7/13/2007

5:29 PM

Page 205


divisions of the zygote, and evidence from patients who require testicular biopsy for sperm extraction suggests that centrosomal function may be inadequate in some cases.37 A significant proportion of mosaic embryos may also originate from oocytes that are aneuploid from the onset, through a process of trisomy ‘rescue’. Oocytes with complex aneuploidies may lead to a high rate of further mitotic errors in the cleaving embryos, and this may also explain the phenomenon of chaotic embryos, which represent almost half of mosaic embryos. The majority of embryo chromosome abnormalities, including mosaicism, are probably prezygotic in origin; this supports the fact that the level of aneuploidy found in oocytes and embryos is comparable. Comparison of chromosome-specific aneuploidy rates in oocytes and embryos may also help in understanding the relationship between oocyte and embryo abnormalities. Preliminary data suggest that oocytes have higher error rates for each chromosome than do embryos, apart from chromosome 16. This may indicate that some of the aneuploidies might be corrected through the chance mechanism of trisomy rescue, which may lead to mosaicism in a certain proportion of embryos after the first three cleavage divisions.30 Precise data concerning the rate of mosaicism in preimplantation development is not known, because only a limited number of preimplantation embryos have been fully studied. The majority of these embryos were available from single blastomere biopsies in PGD cycles for aneuploidies, which may not be representative of the whole embryo. Although it is possible that postzygotic mitotic errors may occur in cleavage stage embryos that were originally euploid, the proportion of aneuploidy/mosaicism that might be due to such errors, and the impact they might have on pre- and postimplantation embryo development is not known. Different types of aneuploidies have been detected by PB and blastomere testing. As mentioned previously, PB testing predicted a predominance of disomy over nullisomy in oocytes following meiosis I, which is in agreement with the predominance of trisomies over monosomies in spontaneous abortions.

With the sole exception of monosomy 21, autosomal monosomies are not compatible with postimplantation development and have never been detected in recognized pregnancies or at birth. On the other hand, a significantly higher prevalence of autosomal monosomies over trisomies has recently been reported in cleaving embryos,30 suggesting that these might originate from postzygotic errors through mitotic non-disjunction or anaphase lag in the first cleavage divisions. Discordance in the trisomy/monosomy ratio detected in oocytes and embryos may also be explained by the fact that a certain proportion of monosomies may not be true monosomies, but represent mosaic embryos that will actually form euploid embryos in the process of pre- or postimplantation development. Follow-up of monosomies that had been detected pre- and postzygotically to the blastocyst stage showed that 56.6% of postzygotic monosomies developed to blastocysts, compared with 80.4% blastocyst development from prezygotic monosomies. Out of a total of 134 prezygotic monosomic embryos, 88.1% were monosomic at the blastocyst stage, which is a significantly higher proportion than was obtained from postzygotic monosomies, 59.6%. The remainder of these postzygotic monosomies were found to be mosaic, or had a normal karyotype.30 It appears that autosomal monosomies are compatible with preimplantation development, although the proportion of monosomies that reach the blastocyst stage depends on the origin of the monosomy. These autosomal monosomies are probably lost during the process of implantation. The above data suggest that the most accurate prediction of embryo development may be achieved by sequential PB1, PB2, and blastomere sampling, which yields sequential testing of errors in meiosis I, meiosis II, and mitosis. This may avoid the transfer of embryos with prezygotic chromosomal errors; these errors seem to be the major source of chromosomal abnormalities in the embryo. The proportion of mitotic errors in embryos that originate from euploid zygotes is not known, and these can also be detected using this approach. Data accumulated from this sequential sampling will help to evaluate 

HPE_Chapter17.qxp

7/13/2007

5:29 PM

Page 206


possible differences in viability of embryos with chromosomal abnormalities of meiotic and mitotic origin. Testing for errors in meiosis may also be useful for predicting developmental abnormalities in embryos that originate from uniparental disomies. Recent reviews suggest that uniparental disomies of chromosomes 6, 7, 11, 15, 19, and 20 can potentially lead to imprinting disorders.38 Sufficient data on the prevalence of uniparental disomies in preimplantation development is not available, because FISH techniques in current use cannot distinguish between uniparental disomies and normal sets of chromosomes. Uniparental disomy may be suspected in euploid embryos that were predicted as trisomic by PB1 and PB2 testing, because one-third of the embryos originating from trisomic zygotes may become uniparentally disomic as a result of trisomy rescue. Uniparental disomy may also be accurately detected by DNA fingerprinting.30 More than half of IVF patients are 35 years and older, and more than half of their oocytes may be aneuploid. Therefore, avoiding the transfer of embryos resulting from oocytes diagnosed as aneuploid through PGD should be clinically useful, by potentially improving implantation and pregnancy rates and avoiding uniparental disomies. Although the biological significance of uniparental disomies in preimplantation development is not known, it is possible that the detection and avoidance of uniparental disomies may also contribute to an improvement in implantation and pregnancy rates.

CONCLUSION

The above data indicate that more than half of the oocytes or embryos that were tested are aneuploid, and PGD is therefore of practical relevance for poor prognosis patients. Aneuploidy clearly can affect the embryo’s developmental competence and potential to implant following transfer. In contrast to the data obtained from traditional studies of meiosis, direct testing of meiotic outcomes in patients of advanced reproductive age shows that chromosomal abnormalities originate to a similar extent from meiosis I 

and meiosis II, and are predominantly of chromatid origin. Although isolated errors in meiosis I and meiosis II were also observed, overall 42.7% of oocytes with meiosis I errors also had sequential errors in meiosis II, resulting in apparently balanced zygotes in 32.5% of cases. This may represent a phenomenon of aneuploidy rescue in female meiosis. However, the embryos resulting from such apparently balanced zygotes were predominantly aneuploid, suggesting that these zygotes may be inherently predisposed to postzygotic chromosomal errors following sequential errors in meiosis I and meiosis II. Patterns of errors in meiosis I and meiosis II differed according to the chromosome tested, and these patterns were not in agreement with previously reported data based on DNA polymorphism in liveborn trisomies or spontaneous abortions. Comparing the types of chromosomal aneuploidies and the prevalence of each chromosome specific error detected by FISH and DNA fingerprinting in oocytes and embryos suggests that the majority of chromosomal aneuploidies in embryos originate from female meiosis. This error can predispose the oocyte to further sequential postzygotic errors, and this may explain the high rate of mosaicism in preimplantation embryos. PB and blastomere testing has also demonstrated that uniparental disomies can occur, and the possible impact of this phenomenon is still to be documented. REFERENCES 1. Sherman SL, Peterson MB, Freeman SB et al. Nondisjunction of chromosome 21 in maternal meiosis I: evidence for a maternal agedependent mechanism involving reduced recombination. Hum Mol Genet 1994; 3: 1529–35. 2. Hassold T, Merril M, Adkins K, Freemen S, Sherman S. Recombination and maternal age-dependent nondisjunction: molecular studies of trisomy 16. Am J Hum Genet 1995; 57: 867–74. 3. Peterson MB, Mikkelsen M. Nondisjunction in trisomy 21: origin and mechanisms. Cytogenet Cell Genet 2000; 91: 199–203. 4. Lamb NE, Freeman S, Savage-Austin A et al. Susceptible chiasmate configurations of chromosome 21 predispose to nondisjunction in both maternal meiosis I, and meiosis II. Nat Genet 1996; 14: 400–5. 5. Verlinsky Y, Kuliev A. Atlas of Preimplantation Genetic Diagnosis. London: Parthenon, 2000. 6. Pellestor F, Andreo B, Armal F, Humeau C, Demaille J. Mechanisms of non-disjunction in human female meiosis: the co-existence of two modes of malsegregation evidenced by the karyotyping of 1397 invitro unfertilized oocytes. Hum Reprod 2002; 17: 2134–45.

HPE_Chapter17.qxp

7/13/2007

5:29 PM

Page 207


7. Dyban A, Fredine M, Severova E et al. Detection of aneuploidy in human oocytes and corresponding first polar bodies using FISH. Presented at 7th International Conference on Early Prenatal Diagnosis, Jerusalem, Israel, 22–27 May, 1994: Abstract 97. 8. Verlinsky Y, Cieslak J, Freidin M et al. Pregnancies following preconception diagnosis of common aneuploidies by fluorescent in-situ hybridization. Hum Reprod 1995; 10: 1923–7. 9. Munné S, Daily T, Sultan KM, Grifo J, Cohen J. The use of first polar bodies for preimplantation diagnosis of aneuploidy. Hum Reprod 1995; 10: 1014–120. 10. Dyban A, Fredine M, Severova E et al. Detection of aneuploidy in human oocytes and corresponding first polar bodies by FISH. J Assist Reprod Genet 1996; 13: 72–7. 11. Pujol A, Boiso I, Benet J et al. Analysis of nine chromosome probes in first polar bodies and metaphase II oocytes for the detection of aneuploidies. Eur J Hum Genet 2003; 11: 325–36. 12. Verlinsky Y, Cieslak J, Ivakhnenko V et al. Birth of healthy children after preimplantation diagnosis of common aneuploidies by polar body FISH analysis. Fertil Steril 1996; 66: 126–9. 13. Verlinsky Y, Cieslak J, Ivakhnenko V et al. Preimplantation diagnosis of common aneuploidies by the first and second polar body FISH analysis. J Assist Reprod Genet 1998; 15: 285–9. 14. Verlinsky Y, Cieslak J, Ivakhnenko V et al. Prepregnancy genetic testing for common age-related aneuploidies by polar body analysis. Genet Testing 1998; 1: 231–5. 15. Verlinsky Y, Cieslak J, Ivakhnenko V et al. Prevention of age-related aneuploidies by polar body testing of oocytes. J Assist Reprod Genet 1999; 16: 165–9. 16. Verlinsky Y, Cieslak J, Ivakhnenko V et al. Chromosomal abnormalities in the first and second polar body. Mol Cell Endocrinol 2001; 183: S47–9. 17. Gianaroli L, Magli MC, Ferraretti AP. The in vivo and in vitro efficiency and efficacy of PGD for aneuploidy. Mol Cell Endocrinol 2001; 183: S13–18. 18. Munne S. Preimplantation genetic diagnosis of numerical and structural chromosome abnormalities. Reprod BioMed Online 2002; 4: 183–96. 19. Kuliev A, Verlinsky Y. Current feature of preimplantation genetic diagnosis. Reprod BioMed Online 2002; 5: 296–301. 20. Kahraman S, Bahce M, Samli H et al. Healthy births and ongoing pregnancies obtained by preimplantation genetic diagnosis in patients with advanced maternal age and recurrent implantation failures. Hum Reprod 2000; 15: 2003–7. 21. De Boer KA, Catt JW, Jansen RPC et al. Moving to blastocyst biopsy for preimplantation genetic diagnosis and single embryo transfer at Sydney IVF. Fertil Steril 2004; 82: 295–8. 22. Verlinsky Y, Munne S, Cohen J et al. Over a decade of preimplantation genetic diagnosis experience – a multi-center report. Fertil Steril 2004; 82: 292–4.

23. Kuliev A, Cieslak J, Illkewitch Y, Verlinsky Y. Chromosomal abnormalities in a series of 6733 human oocytes in preimplantation diagnosis of age-related aneuploidies. Reprod BioMed Online 2003; 6: 54–9. 24. Kuliev A, Verlinsky Y. Meiotic and mitotic nondisjunction: lessons from preimplantation genetic diagnosis. Hum Reprod Update 2004; 10: 401–7. 25. Angel R. First meiotic division nondisjunction in human oocytes. Am J Hum Genet 1997; 65: 23–32. 26. Hunt P, LeMaraire R, Embury P, Sheean L, Mroz K. Analysis of chromosome behaviour in intact mammalian oocytes: monitoring the segregation of a univalent chromosome during female meiosis. Hum Mol Genet 1995; 4: 2007–12. 27. Fisher JM, Harvey JF, Morton NE, Jacobs PA. Trisomy 18: studies of the parent and cell division of origin and effect of aberrant recombination on nondisjunction. Am J Hum Genet 1996; 56: 669–675. 28. Kuliev A, Cieslak J, Zlatopolsky Z et al. Origin of aneuploidies in preimplantation embryos. 2003 Fifth International Symposium on Preimplantation Genetics, 5–7 June, Antalya, Turkey. P. 16–17. 29. Kuliev A, Cieslak J, Zlatopolsky Z et al. Aneuploidy rescue after female meiosis I and follow up analysis of its outcome in resulting preimplantation embryos. Am J Hum Genet 2003; 73 (Suppl): 189. 30. Verlinsky Y, Kuliev A. Practical Preimplantation Genetic Diagnosis. Springer, London: New York, 2006. 31. Kim NH, Chung HM, Cha KY, Chung KS. Microtubule and microfilament organization in maturing human oocytes. Hum Reprod 1998; 13: 2217–22. 32. Barrit J, Brenner C, Cohen J, Matt D. Mitochondrial DNA rearrangement in human oocytes and embryos. Mol Hum Reprod 1999; 5: 927–33. 33. Perez G, Flaherty S, Barry M, Matthews C. Preliminary observations of polar body extrusion and pronuclear formation in human oocytes using timelapse video cinematography. Hum Reprod 1997; 12: 532–41. 34. Gianaroli L, Magli MC, Ferraretti AP et al. Pronuclear morphology and chromosomal abnormalities as scoring criteria for embryo selection. Fertil Steril 2003; 80: 837–44. 35. Munne S, Bahce M, Sandalinas M. Differences in chromosome susceptibility to aneuploidy and survival to first trimester. Reprod BioMed Online 2004; 8: 81–90. 36. Munne S, Sandalinas M, Escudero T et al. Some mosaic types increase with maternal age. Reprod BioMed Online 2002; 4: 223–32. 37. Silber S, Sadowy S, Lehahan K et al. High rate of chromosome mosaicism but not aneuploidy in embryos from karyotypically normal men requiring TESE. Reprod BioMed Online 2002; (Suppl 2), 20. 38. Lucifero D, Chaillet JR, Trasler M. Potential significance of genomic imprinting defects for reproduction and assisted reproductive technology. Hum Reprod Update 2004; 10: 3–18.



HPE_Chapter17.qxp

7/13/2007

5:29 PM

Page 208

HPE_Chapter18.qxp

7/13/2007

5:29 PM

Page 209

18. Chromosomal status of human embryos Santiago Munné and Luca Gianaroli

INTRODUCTION

Data from SART and ASRM1 indicate that only 38.1% of cycles in women younger than 34 years reach delivery of a healthy child after assisted reproduction. In the age group 35–37 years, this figure is 32%, 23% in ages 38–40 years, and 11% in women older than 40 years. Assuming an average of three embryos transferred in women 35 and older, these figures indicate that very few of their embryos implant. Irrespective of whether this is a result of chromosome abnormalities or other factors, the high rate of chromosome abnormalities in cleavage stage embryos itself can largely explain the high degree of implantation failure observed in assisted reproductive technology (ART), since 60–80% of human embryos are chromosomally abnormal, depending on maternal age (see Table 18.1).2–4 The chromosome abnormalities seen in human IVF embryos create three immediate broad questions that are addressed in this chapter: (1) What causes these abnormalities, and can they be prevented? (2) How reliable is morphological selection in selecting against chromosome abnormalities? (3) How effective is preimplantation genetic diagnosis (PGD) as a selection tool to improve ART outcome? Before answering these questions, it is necessary to understand what type of data we are working with, and how it was obtained. Several requirements must be met in order to study numerical chromosome abnormalities in preimplantation human embryos. First, cleavage arrested embryos must be evaluated by interphase analysis, because their cells are seldom at metaphase stage when fixed. Second, distinguishing differences in aneuploidy rates among different

chromosomes requires analysis of individual chromosomes. Third, all blastomeres of non-transferred embryos should be analyzed to distinguish mosaicism from other abnormalities. To date, fluorescence in situ hybridization (FISH) is the most efficient method used to fulfill these requirements. FISH can be used to study the chromosome constitution of cleavage stage blastomeres in interphase, with efficiencies of over 90% per cell.5,6 Simultaneous analysis of more than three chromosome pairs permits differentiation of polyploidy, haploidy, and aneuploidy in arrested human embryos.7 When most or all cells of an embryo are analyzed, mosaicism can be differentiated from FISH failure, as well as from aneuploidy, and their mechanism of formation ascertained.7 However, FISH has its disadvantages: FISH supplies information only on the chromosomes for which specific probes are used, and only five fluorochromes of the visible spectrum can be used simultaneously. Thus, sequential FISH must be applied, and even then only a maximum of three rounds of FISH can be used reliably; this reduces the potential number of chromosomes that can be studied to 15, so far. Other techniques such as spectral karyotyping (SKY),8 or comparative

Table 18.1 Frequency (%) of chromosome abnormalities in cleavage stage human embryos Age (years)

Aneuploidy (9 chromosomes) Other aneuploidy (by CGH)a Postmeiotic abnormalitiesb Total abnormal

20–34

35–39

40–47

24

27

39

p 0.001

5

6

8

Unknown

35

36

35

NS

64

69

82

p 0.001

CGH, comparative genomic hybridization; NS, not significant. From Munné et al,2 Marquez et al,3 and aGutierrez-Mateo et al.4 bMosaics, polyploid, haploid, chaotic.



HPE_Chapter18.qxp

7/13/2007

5:29 PM

Page 210


genome hybridization (CGH)9–11 have also been used. However, studies using CGH10 indicate that the nine-chromosome probe FISH test currently used in PGD12 does detect about 75–85% of chromosome abnormalities that can be detected by CGH, both because the nine chromosomes are the ones most frequently involved in aneuploidy, and also because the other chromosomal aneuploidies tend to occur simultaneously with those involving the nine chromosomes tested.10,13 FISH therefore remains the most suitable method for preimplantation genetic analysis of human embryos.

CAUSES OF CHROMOSOME ABNORMALITIES IN ART EMBRYOS HORMONAL STIMULATION

Aneuploidy rates in infertile women undergoing ovarian stimulation with gonadotropins have been uniformly reported as high. Whether these rates are ‘natural’, or are the consequence of gonadotropin stimulation is, however, unknown. There are virtually no reports comparing aneuploidy rates in natural and stimulated cycles. In a study that compared the rate of spontaneous abortions between IVF patients and patients conceiving naturally, it was found that IVF patients had a 1.2 greater risk of miscarriage than did the control group.14 However, there is mounting evidence to suggest that different types of hormonal stimulation may have a different impact on the frequency of euploid embryos. Tarin et al15 reported that cycles generating slow-cleaving embryos occurred in women who were treated with gonadotropin releasing hormone agonist (GnRH-a) before ovarian stimulation for a shorter period of time than in women whose cycles generated fast-cleaving embryos. Slowly developing embryos have a higher frequency of chromosomal abnormalities than do those that develop normally.2 Another study demonstrated that embryos obtained from several laboratories that used different stimulation protocols had very diverse rates of mosaicism (p 0.001).16 Specifically, cycles resulting



from clomiphene citrate stimulation showed 40% mosaicism, compared with 25% mosaicism in embryos from down regulated cycles (p 0.05). The chromosome abnormality rates in the two different hormonal regimens are probably higher, since that study was performed using FISH with only five probes. The fact that embryos with multinucleated blastomeres (MNBs) have higher rates of chromosome abnormalities than non-MNB embryos is well documented,17 and it also seems that the incidence of embryos with MNBs is related to hormonal stimulation response. Jackson et al18 also found that in cycles containing MNBs, the level of estradiol on the day of human chorionic gonadotropin (hCG) was doubled (2401 vs 1270 pg/ml, p 0.001), twice as many oocytes were collected (22 vs 10, p 0.001), and fewer ampoules of gonadotropins (p 0.001) were used than in cycles without MNBs. Follicular underoxygenation has also been correlated with multinucleation seen at the 2-cell stage of development.19 In addition, cycles with large numbers of oocytes retrieved have significantly more MNB embryos,18,20 and lower implantation rates.21 Another study demonstrated that cycles containing MNB embryos have double the level of estradiol on the day of hCG administration, and need fewer ampoules of gonadotropins than cycles without MNBs.18 However, this was not confirmed by other studies (see Chapter 5).20 The duration of hormonal stimulation has also been associated with multinucleation; MNB occur more commonly in embryos of patients who required fewer days of gonadotropins before hCG (Chapter 5).20 Patients with low (1 mIU/ml) levels of follicle stimulating hormone (FSH) on day 3 tend to have a higher incidence of multinucleation (Chapter 5). According to Van Royen et al20 the higher rates of multinucleation observed in patients that require a higher dose of gonadotropins, as well as in patients with shorter cycles, suggests that the incidence of MNB might be related to a high number of immature follicles at the time of ovulation induction. Immature follicles may reach metaphase II, but either their nucleus or cytoplasm might be inadequately mature, so that they are unable to achieve correct cleavage.

HPE_Chapter18.qxp

7/13/2007

5:29 PM

Page 211

CHROMOSOMAL STATUS OF HUMAN EMBRYOS

Nogueira et al22 found that embryos developing from in vitro matured oocytes were mostly multinucleated, and Van Royen et al.20 therefore suggested that the frequency of MNB embryos might be minimized by modifying hormonal stimulation. Recently, Weghofer et al23 evaluated the impact of controlled ovarian hyperstimulation with gonadotropins on euploidy rates in human cleavage stage embryos. Three different protocols of ovarian hyperstimulation with gonadotropins were compared: (1) long agonist stimulation; (2) short agonist stimulation; and (3) the use of a GnRH antagonist. Their preliminary results suggested that higher gonadotropin dosages result in more euploid embryos for transfer, but the percentage of euploid embryos is lower overall. Therefore, high dose gonadotropin stimulation, resulting in large numbers of embryos, may represent a potential indication for PGD. Rates of euploidy appear to be improved by exogenous luteinizing hormone (LH) supplementation in long agonist protocols, especially in younger women. Euploidy rates after ovarian stimulation thus appear to be affected both quantitatively and qualitatively by stimulation protocols and doses of medication. Differing responses to hormonal stimulation may also have an impact on the frequency of euploid embryos per cycle. For example, patients with reduced ovarian reserve and a lower total number of follicles in their ovaries have an increased risk of trisomy.24,25 Indeed, Gianaroli et al26 described a high rate of chromosome abnormalities in embryos of women who had a poor response to hormonal stimulation. High ovarian response is considered to be detrimental to implantation, because this has a negative effect on the physiological relationship between estradiol levels and endometrial receptivity, reducing implantation potential.27,28 Reis Soares et al29 also observed that patients showing high ovarian response produce more chromosomally abnormal embryos. Oocyte donors were compared with patients undergoing PGD for X-linked diseases, and these two groups were found to have a significant difference in oocytes retrieved (25 and 15, respectively), in chromosome abnormalities (56% and 37%, respectively) (p 0.01), and in implantation rate (25%

and 36.4%, respectively). High ovarian response has been associated with multinucleation,18 and uneven pronuclei,30 both features that correlate with chromosome abnormalities.17 Patients with polycystic ovarian syndrome tend to have HOR and higher miscarriage rates,31 but they do not have a higher proportion of chromosomally abnormal embryos.23 OTHER FACTORS

Although controlled ovarian hyperstimulation may be the principal factor responsible for the high rate of chromosome abnormalities in cleavage stage embryos created by ART, other factors may also have a detrimental effect, such as exposure of gametes and embryos to microscope or room light, changes in temperature when opening and closing incubators or while handling dishes, air impurities, and unsuitable culture media. Spindle microtubules are thermosensitive, and even a small change in temperature can deeply disturb oocyte spindle structure.32,33 Pickering et al32 found that after 10 minutes at room temperature, human oocytes have a disassembled spindle in 50% of cases, and 100% have a disassembled spindle after 30 minutes at room temperature, accompanied by chromosome malsegregation. Cohen et al34 showed that volatile compounds negatively affect mouse embryo development, and also decrease human pregnancy and implantation rates in IVF programs. More recently, Boone et al35 demonstrated that fertilization rates, embryo quality, and pregnancy rates increased significantly after the installation of air filters in IVF laboratories. The effect of volatile organic compounds on chromosome abnormalities has not been studied. Identifying the factors that produce an increase in chromosomally abnormal embryos will lead to better ART methodology, with the creation of a higher proportion of euploid embryos. However, the current reality is that more than 50% of embryos are chromosomally abnormal. Can these chromosomally abnormal embryos be selected morphologically or developmentally?



HPE_Chapter18.qxp

7/13/2007

5:29 PM

Page 212


MORPHOLOGICAL SELECTION AGAINST CHROMOSOME ABNORMALITIES DYSMORPHISM AND CHROMOSOME ABNORMALITIES

Morphological and developmental variants in cleavage stage embryos have recently been reviewed in relation to chromosome abnormalities.17 Several types of dysmorphic embryos were compared with non-dysmorphic embryos, with respect to chromosome abnormalities. Some of the features associated with an increased incidence of chromosomal abnormality were fragmentation, multinucleation, and asymmetric blastomeres. However, decreasing morphological quality was not related to a higher rate of aneuploidy, but instead to an increase in postmeiotic abnormalities, such as mosaicism, monospermic polyploidy, haploidy, and chaotic embryos. In contrast, aneuploidy increases only with maternal age and is not related to morphology, at least up to day 3 of development. Most dysmorphisms (fragmentation, MNB, asymmetry, etc.) tend to occur in the same embryos. For example, Hardarson et al36 found that asymmetric embryos have higher rates of multinucleation, and Van Royen et al20 found that highly fragmented embryos also have more multinucleated blastomeres. In terms of cleavage rate, embryos with 4 cells on day 2 and 8 cells on day 3 had a lesser chance of being multinucleated than those at other stages.20 FRAGMENTATION

It is well known that fragmented embryos have a lower implantation potential than those that are not fragmented.37–39 Fragmentation rate has also been associated with chromosome abnormalities.40–43 Chromosome abnormalities increase from 50–60% in non-fragmented embryos to 70–90% in embryos with 35% fragmentation, but while fragmentation is strongly correlated with mosaicism and other postzygotic abnormalities, aneuploidy does not appear to increase with fragmentation.41,42 

MULTINUCLEATION

The frequency of multinucleated embryos per patient ranges from 15% to 33.6% (Chapter 5).18,20,44 The different rates reported by different centers are significant, and could be attributed to differences in hormonal stimulation and culture conditions.16 The effect of multinucleation on implantation is very clear. Alikani et al37 reported that only 16% of embryos with one or more multinucleated cells on day 2 or day 3 reached blastocyst stage, compared with non-multinucleated embryos (32%, p 0.001). Similar low rates of implantation for embryos with multinucleation have been reported by others (Chapter 5).18,20,44 As with fragmented embryos, the implantation potential of MNB embryos was higher if they were transferred on day 3 (13%) than on day 5 (7%), compared with non-dysmorphic and normally developing embryos (40.1% and 49%, respectively) (Chapter 5). FISH studies on MNBs showed that the chromosomal content of each MNB nucleus was not always the same as the chromosomal content in nuclei of sibling blastomere MNBs.41,45 Several studies have analyzed MNBs observed at the 2-cell stage, and all have detected high rates of abnormalities, ranging from 55 to 100%; the differences between studies depend mostly on the number of chromosomes analyzed by FISH (Chapter 5).42,46–48 EMBRYOS WITH ASYMMETRIC BLASTOMERES

Blastomere asymmetry has been linked to reduced embryo competence.36,38,49 Compared with embryos that have symmetric blastomeres, embryos with asymmetric blastomeres had more chromosome abnormalities occurring postmeiotically (mosaicism, polyploidy, and haploidy) (35% vs 21%) (p 0.001), fewer normal embryos (32.5% vs 40%, p 0.05), but similar rates of aneuploidy.50 GIANT OOCYTES

Giant oocytes and embryos have an average diameter of 200 m, including the zona pellucida, and occur at a frequency of 0.3%.51,52 Embryos developing from giant oocytes were found to be invariably triploid or

HPE_Chapter18.qxp

7/13/2007

5:29 PM

Page 213


triploid mosaics, with XXX or XXY gonosome constitutions, which suggests a higher contribution from the maternal genome (Munné et al53 (n 6), Balakier et al51 (n 4)). Similar oocytes karyotyped at MII by Rosenbusch et al52 (n 6) and Balakier et al51 (n 9) were also found to be triploid. According to Balakier et al51 embryos developing from giant eggs can reach blastocyst stage, indicating that they can be a source of digynic triploid fetuses. EMBRYOS WITH A DOMINANT BLASTOMERE

Embryos with only one large cell surrounded by smaller blastomere-sized extracellular fragments were found to be invariably polyploid, and frequently polyploid mosaics.42,53 EMBRYOS WITH CYTOPLASMIC IRREGULARITIES

According to Magli et al,42 embryos with cytoplasmic aggregates had 86% chromosome abnormalities compared with 63% in embryos with regular cytoplasm (p 0.001). EMBRYOS WITH ELONGATED SHAPE

Magli et al42 analyzed the chromosome constitution of 18 elongated embryos, and found that they have the same proportion of chromosome abnormalities as non-elongated embryos. These embryos have been known to produce babies.54 Table 18.2 summarizes the relationships found between embryo morphologies and chromosome abnormalities. PRONUCLEAR MORPHOLOGY AND CHROMOSOMAL ABNORMALITIES

Previous studies have demonstrated that some morphological characteristics of pronuclear zygote morphology are an expression of events occurring in the oocyte after fertilization, strictly related to the design of the embryonic axis.57,58 Any alteration in these events may cause abnormalities in embryo cleavage which are often related to chromosomal abnormalities (see next section). Based on these considerations, an association was postulated between pronuclear morphology and the chromosomal

Table 18.2 Summary of morphological abnormalities and their relation to chromosomal abnormalities Embryo morphology

FISH analysis

Reference

Normal morphology 20–34 years old 35–39 years old 40–45 years old

16% abnormal 37% abnormal 53% abnormal

2 2 2

Dysmorphic 2pn embryos: Uneven 2PN Abnormal NPB distribution Giant embryos (220 mm) Dominant single blastomere 35% fragments Multinucleated embryos Asymmetric blastomeres

73–87% abnormal 71–81% abnormal Triploid Polyploid 70–90% abnormal 74–100% abnormal 67%

30,55 55,56 7,51 7 42,45 42,45 45

PN, pronuclei.

condition of preimplantation embryos.59 According to these results, the combination of patterns related to pronuclear morphology indicates that some configurations are associated with a higher proportion of euploid embryos, while the opposite is true for other configurations.59–61 These results are in full agreement with those derived from the insemination of euploid oocytes, as defined by first polar body testing.62 The current observations confirm that some patterns of pronuclear morphology are associated with a higher proportion of euploidy, not only reaffirming the relevance of oocyte quality in determining its fate, but also suggesting that this scoring system may be beneficial for the prediction of zygote viability. However, it was recently shown that cohorts of zygotes with poor pronuclear score have similar outcomes to cohorts of zygotes with very poor scores.63 The similarity in outcomes was measured from cohorts that were identical in scoring leaving little doubt that chromosomal anomalies are not absolute even in the poorest zygote cohorts. COMMENT CLEAVAGE STAGE PATTERNS AND CHROMOSOME ABNORMALITIES

In addition to the dysmorphisms mentioned above, the cleavage patterns of embryos from day 1–3 are at 

HPE_Chapter18.qxp

7/13/2007

5:29 PM

Page 214


least as important in the selection of embryos with the highest implantation potential. Developmental day 3 embryos can be classified into four large groups according to their rate of cleavage: ‘arrested’, ‘slow’, ‘normal’, or ‘fast’ developing embryos. Arrested embryos are those that have not cleaved during a 24hour period. Slow embryos have not reached the 7-cell stage on day 3 of development but have cleaved during a 24-hour period. Normal embryos reach 7–9 cells on day 3, with 15% fragmentation and without multinucleation, and have cleaved in the preceding 24 hours (‘normally developing’ embryos with abnormal morphology are here grouped with slow embryos). Accelerated embryos have 9 cells by day 3. Several articles have reported a poor clinical outcome and/or development for embryos with slow or accelerated cleavage rates.37,38,49,64 In contrast to Alikani et al,37 other studies have found that the major and almost sole indicator for an embryo to result in offspring was blastocyst morphology, and not day 3 morphology.38,65,66 Three studies now report the analysis of more than 500 embryos per study.42,67 Study I analyzed 524 embryos,67 study II analyzed 721 embryos,3 and study III analyzed a single cell from each of 1596 embryos.42 Study III analyzed chromosome abnormalities in relation to cell number, and found that normal day 3 embryos with 7–8 cells showed the lowest rate of abnormalities (55%), while slow embryos with 4 cells or less had 74% (p 0.001); accelerated embryos with 9 or more cells were similar, with 79% abnormal (p 0.005).42 Studies I and II3,67 classified the embryos in three maternal age groups: 20–34; 35–39; and 40–47 years old. Pooled results from a total of 1255 embryos from these two studies demonstrated a highly significant relationship between maternal age and aneuploidy (p 0.001). Individual chromosomes involved in non-disjunction were also analyzed using pooled results from the two studies, and chromosomes 16, 18, and 21 showed significantly higher frequencies of aneuploidy with increasing maternal age (p 0.05, p 0.05, and p 0.01, respectively).50 The chromosomes most frequently involved in aneuploidy at the cleavage stage were found to be, in order of frequency, chromosomes 22, 16, 15, and 21 (Table 18.3). 

Table 18.3 Specific chromosome aneuploidy rates. Double aneuploidies counted twice, once for each chromosome. Tetrasomies and nulisomies were counted as two trisomies and two monosomies, respectively Number of aneuploid embryos analyzed Chromosome

Number of embryos

n

%

XY 1 2 3 4 6 7 11 13 14 15 16 17 18 21 22

1741 559 426 426 753 194 244 426 2227 280 1492 2091 1035 2484 2437 1700

21 14 11 11 17 3 7 9 76 3 91 134 33 69 132 118

1.2 2.5 2.6 2.6 2.3 1.5 2.9 2.1 3.4 1.1 6.1 6.4 3.2 2.8 5.4 6.9

Total

2484

750

From Munne et al50 and Abdelhadi et al.13

A recent study analyzing 4665 embryos confirmed that the incidence of chromosomal abnormalities is significantly higher in arrested or slow-cleaving embryos as well as in accelerated embryos, compared to embryos with 8 cells at 62 hours postinsemination.68 This study also reported that the presence of an uneven number of blastomeres or fragments scattered in the perivitelline space was associated with an increased incidence of chromosomal abnormalities. Following these considerations, the study by Magli et al68 focused on the question of whether it is preferable to transfer a slow-cleaving embryo with no fragmentation, or a regularly cleaving embryo, i.e. with 7 or 8 cells on day 3, with a variable proportion and type of fragmentation. According to the reported data, when considering embryos with no fragments or with a concentrated pattern of fragments, the incidence of chromosomal abnormalities is dependent on the cellular stage, irrespective of the percentage of fragmentation. Conversely, in the presence of scattered fragmentation, the incidence

HPE_Chapter18.qxp

7/13/2007

5:29 PM

Page 215


of chromosomal abnormalities is significantly higher in 7- or 8-cell embryos compared with an embryo with fragments concentrated in one area. For slow-cleaving embryos (4–6 cells on day 3) there is apparently no correlation between fragmentation and the total amount of chromosomal abnormalities. Altogether, these observations suggest that 8-cell embryos on day 3 are always preferable for transfer, even in the presence of fragmentation, especially if fragments are concentrated in one area of the perivitelline space. However, the large difference in the number of aneuploid embryos detected at cleavage stage compared with prenatal data from spontaneous abortions69–71 indicates a strong selection against aneuploid embryos before or shortly after implantation. Aneuploidy does not necessarily lead to developmental arrest in the first 3 days of culture, since the embryonic genome is not fully active until day 2 or 3 of development.72,73 Indeed, aneuploidy does not increase with decreasing developmental potential and cannot be reduced by selecting day 3 embryos of normal appearance.3,42,67 In contrast, postmeiotic abnormalities do increase with decreasing embryonic competence and are more common in arrested, slow, and dysmorphic embryos than in those that are normally developing.3,42,67 These postmeiotic chromosome abnormalities account for more than half of the abnormalities detected in cleavage stage embryos. Although postmeiotic abnormalities significantly decrease with embryonic competence (polyploidy p 0.001, extensive diploid mosaicism p 0.01), they are not affected by maternal age.3,67 Polyploidy is found mostly in arrested embryos, and increases with decreasing embryo competence.3,67,74,75 As previously noted, it is unlikely that this represents polyspermy, since all of these embryos were derived from dipronucleated zygotes with two polar bodies.45 The most likely explanation is that their DNA synthesis continued in the absence of cellular division.76,77 Excluding the 40 years age group, extensive (3/8 abnormal) diploid mosaicism is the major chromosome abnormality in IVF-generated human embryos. For the age group 35–39, mosaicism was found in 23.3% of the embryos,3 followed by

polyploidy (21.8%), aneuploidy (10.2%), and haploidy (3.6%). Even in the group of normally developing embryos, which are closer in quality to embryos that are transferred, aneuploidy (19.3%) contributes to less than half of the chromosome abnormalities detected, with extensive diploid mosaicism (14.7%), polyploidy (4.5%), and haploidy (4%) together contributing to a greater extent than aneuploidy. Similarly, another study by Bielanska et al75 detected 2N/mosaicism (extensive and limited) and chaotic mosaics in 55% of surplus embryos, followed by 30% normal embryos, and the remainder were aneuploid, polyploidy, and haploid. Because the embryonic genome is not fully active until day 3 of development,72,73 mosaicism, polyploidy, and haploidy cannot produce dysmorphism originating in the first and second meiotic divisions. However, cytoplasmic impairment could produce both mosaicism and polyploidy, through cytoskeletal and spindle malfunction leading to a block in cell division, or other mechanisms. For example, abnormalities of the centriole in a fertilizing spermatozoon may produce mosaicism or other chromosome abnormalities in the resulting zygote.78–80 In addition, intrinsically low mitochondrial activity has been associated with chaotic mosaicism, but not with other types of chromosome abnormalities.81 The majority of chromosome studies on human embryos have focused on meiotic irregularities as the principal source of chromosome abnormalities. However, the reviewed data17 indicate that other sources of chromosome abnormalities are more important, and further investigation of other factors such as culture conditions, hormonal stimulation, centriole abnormalities, and cytoplasmic factors is justified. BLASTOCYST FORMATION AND CHROMOSOME ABNORMALITIES

STUDIES IN UNSELECTED BLASTOCYSTS

Many studies have assessed the chromosome composition of surplus blastocysts (Table 18.4).76,82–89 The rate of mosaicism detected by FISH is as high as 

HPE_Chapter18.qxp

7/13/2007

5:29 PM

Page 216


Table 18.4 Chromosome abnormalities in unselected blastocyst studies Number of blastocysts analyzed

Normal

2N/polyploid

2N/aneuploid (and/or polyploid)

2N/chaotic or chaotic

Polyploid

Haploid

Aneuploid

83 84 85 86 89a 89b 88 87

73 8 19 33 304 182 58 91

43 1 1 3 207 162 30 6

17 6 6 24 87 0 10 60

4 1 10 (6) 5 0 12 15 (5) 16

0 0 2 1 0 0 3 3

7 0 0 0 10 2 0 3

0 0 0 0 0 0 0 0

2 0 0 0 0 6 0 3

aPloidy

analysis only; bkaryotype analysis.

Reference

90%,75,85,87,89 but the percentage of abnormal cells was low compared with cleavage stage mosaic embryos, being no higher than 30% on average. In addition, the majority of abnormal cells found in mosaics were tetraploid, 2N/4N mosaics being very common (23–86% of all blastocysts), but triploid, haploid, aneuploid, and chaotic cells were also described. Although aneuploid cells seem to be detrimental to embryo development, high levels of mosaicism and chaotic embryos can still be detected at the blastocyst stage.90,91 Compared with lower rates of 2N/4N mosaics on day 3 of development,41 this indicates that the majority of 2N/4N mosaics arise at morula or later stages. In studies comparing chromosome abnormalities in blastocysts that developed from good or poor morphology day 3 embryos, higher rates of abnormal cells per blastocyst were found in those developing from poor morphology day 3 embryos.86,88 A study evaluating chromosome abnormalities in relation to blastocyst morphology found that 65% of mosaic blastocysts had good morphology.87 Thus, by itself, morphology is generally not an appropriate tool to screen for chromosome abnormalities. SURVIVAL OF CHROMOSOMALLY ABNORMAL EMBRYOS TO BLASTOCYST STAGE

The survival of chromosome abnormalities to the blastocyst stage has been analyzed.3,75,90,91 Several authors have suggested that the fact that many embryos arrest during the morula stage may act as a 

selection against chromosome abnormalities during extended culture.82,92 Magli et al90 reported that only 22% of chromosomally abnormal embryos reached blastocyst stage compared with 34% of euploid embryos (p 0.001). Of the embryos surviving to blastocyst stage, the majority had a mosaic inner cell mass, with 2–16 different cell lines. Marquez et al3 compared day 3 and day 4 embryos, and found that embryos analyzed on day 4 had much higher rates of polyploidy than those analyzed on day 3, suggesting that embryos arresting on day 3 become polyploid by day 4. MOSAIC EMBRYOS

Bielanska et al86 found chaotic mosaics to be more common in arrested day 3 and day 4 embryos than in blastocyst stage embryos. They also found that diploid/polyploid mosaics increased with developmental competence, with an overall decrease in the number of abnormal cells per mosaic embryo from cleavage stage to blastocyst stage. Sandalinas et al91 observed that embryos with a high frequency of mosaicism can occasionally develop to blastocyst stage, although they seldom had more than 60 cells, compared with an average of 114 for normal blastocysts. TRISOMIES

Of trisomies 37%91 and 34%93 reached blastocyst stage compared with 66%91 and 61%93 of normal

HPE_Chapter18.qxp

7/13/2007

5:29 PM

Page 217


embryos. The difference was statistically significant in both studies. Magli et al90 also found that abnormal embryos reached blastocysts less often than did those that were normal (22% vs 34%). MONOSOMIES AND HAPLOIDY

No haploid embryos, and only 9% monosomies, restricted to monosomy X and 21, reached blastocyst stage in the Sandalinas study.91 Magli et al90 and Rubio et al93 also detected a strong selection against haploidy and autosomal monosomy, but while some monosomies reached blastocyst stage they did not re-analyze the whole embryo to confirm the abnormality; some of the embryos developing to blastocyst might have been normal embryos that had been misdiagnosed93 or had differences between inner cell mass and trophectoderm.90 Rubio et al93 found that 54% of embryos with monosomy X survived to blastocyst. The fact that only monosomy X and 21 were found in blastocysts91 agrees with prenatal diagnosis data, where no other monosomies are detected in first trimester abortions.94

with apparently normal morphology in women 35–39 years old. This rate increases to about 60% in women aged 40 years. These frequencies are for six chromosomes,2,3 and could be higher with more chromosomes analyzed. Assuming that competent IVF centers already practice appropriate embryo selection on the basis of morphology and development, a fraction of chromosomally abnormal embryos have been selected, but it remains impossible to select against the majority of aneuploidies and some postmeiotic abnormalities. As a result, embryos with good morphology and development still show low implantation rates. Culture to blastocyst stage can further select against some chromosome abnormalities, but a number of slow embryos that are chromosomally normal and could potentially implant will not survive extended culture.

PREIMPLANTATION GENETIC DIAGNOSIS AS A SELECTION METHOD RATIONALE FOR PGD IN INFERTILITY

POLYPLOIDY

Polyploid embryos clearly reach blastocyst stage, as polyploid pregnancies do reach first trimester and beyond. Sandalinas et al91 found that 21% of polyploid embryos developed into blastocysts; this has been confirmed by others.93 EFFECTIVE BUT LIMITED IMPACT OF MORPHOLOGICAL AND DEVELOPMENTAL SELECTION

Embryo selection is critical to the success of IVF. Careful evaluation of embryo morphology performed under powerful inverted microscopes will detect many abnormalities such as multinucleation and fragmentation. Dysmorphic and arrested embryos, 50% of which are chromosomally abnormal, should not be transferred if better embryos are available. However, this evaluation does not allow selection against many chromosome abnormalities that occur with a frequency of 30% in embryos

More than 50% of IVF embryos are chromosomally abnormal, and this frequency is much higher than that reported in spontaneous abortions. A sizable proportion of chromosomally abnormal embryos are probably eliminated before they are recognized clinically.95 This loss of embryos could account for the low implantation potential of ART embryos. We hypothesized that PGD selection against chromosomally abnormal embryos predestined to fail implantation could reverse this trend,6 if a sufficient number of embryos are available. Avoiding the replacement of embryos with chromosome abnormalities should significantly increase implantation, and reduce both chromosomally abnormal conceptions and spontaneous abortions. This should result in higher take-home baby rates. APPROPRIATE METHODOLOGY FOR PGD

In general, PGD for numerical chromosome abnormalities is indicated for patients aged 35 and older. 

HPE_Chapter18.qxp

7/13/2007

5:29 PM

Page 218


Table 18.5 Estimate of PGD for infertility procedures performed worldwide to June 2006 US

Total

Reprogenetics RGI Integramed PGDbyART Others

6638 5010 2000c 1533 (until merged with Reprogenetics) 1500c

Europea IVI, Spain Centers in ESHRE consortium SISMER, Bologna, Italy Reprogenetics, Spain

4300 1880b 1710 560

Asia, Oceania, South America Farah Hospital, Jordan Abdelmassih, Sao Paulo, Brazil Sidney IVF, Australia Others Asia GEN-LAB, Ankara, Turkey Other Turkey Other South America Reprogenetics (Japan)

1629 1000c 1000c 800c 505 500c 500c 125

Total

It is also offered to younger patients who are oocyte donors, have recurrent miscarriages, or a history of failed implantation. So far, close to 29 000 PGD cases have been performed by either embryo biopsy on day 3 of development or polar body biopsy (Table 18.5). Several studies have shown an increase in implantation, pregnancy, and take-home baby rate with a decrease in spontaneous abortion.95–100 However, others have found little evidence of similar success.101 Technical differences can explain these results: in general, PGD results can be affected by several parameters (Table 18.6). CELL TYPE ANALYZED AND NUMBER OF CELLS ANALYZED

30 665

aMost

centers in Europe report to the European Society of Human Reproduction and Embryology (ESHRE) consortium. The largest centers are reported separatedly. b1788 cases from IVI, and 1320 from SISMER, although reported to ESHRE consortium, are not included here but in IVI and SISMER. Data to 2004. cApproximate estimate.

PGD for infertility can be performed after polar body biopsy (PBs),104 blastomere biopsy of day-3 cleavage stage,6 or blastocyst biopsies.105 The majority of PGD cases have been carried out after PB or cleavage stage biopsy. Both procedures have their advantages and disadvantages. FISH analysis of first PBs, sometimes in combination with second PB analysis, was first attempted by Verlinsky et al104 and Munné et al.67 Since autosomal aneuploidy

Table 18.6 Summary of studies comparing PGD and control ART outcome Reference Types of mosaics Cells biopsied Fixation used Chromosomes analyzed Type of study Cycles control Cycles PGD Average number of embryos replaced control Average number of embryos replaced PGD Implantation rate control (%) Implantation rate PGD (%) Pregnancy loss rate control (%) Pregnancy loss rate PGD (%) Ongoing implant rate* control (%) Ongoing implant rate* PGD (%) Pregnancy rate control (%) Pregnancy rate PGD (%)

96

99

97

102

101

98

1 Carnoid 4–8** comp 117 117 NA

1 Carnoid 8 comp 127 135 3.0a

1 Carnoid 8 comp 138 138 3.7a

1 Twin 20 8 randzd 28 29 NA

2 Several 6 randzd 141 148 2.8a

1 Carnoid 8 comp 8706*** 562*** NA

NA 13.7 17.6 33.8c 15.0c 10.6c 15.9c 29.9 35.9

1.8a

2.0a

NA

2.0a

NA

12.4a 24.2a 20.6 5.4 10.2a 22.5a 25.1 29.1

10.6c 17.6c NA NA NA NA NA NA

NA NA NA NA NA NA 20.7 43.0

11.5 17.1 25.6 25.0 10.4b 16.5b 27.7 19.6

NA NA 21.5a 16.7a NA NA NA NA

comp, prospective non-randomized comparative study; randzd, prospective randomized study; *fetus ongoing 12 weeks/embryos replaced; **36 cycles with four probes, 50 cycles with five probes, and 31 cycles with eight probes; ***pregnant cycles; ap 0.001; bp 0.06; cp 0.05. Adapted from Cohen et al.103



HPE_Chapter18.qxp

7/13/2007

5:29 PM

Page 219


occurs predominantly during maternal meiosis I,70,106,107 analyzing the first PB for aneuploidy can detect the majority of autosomal aneuploidies identified in blastomeres. PB biopsy has three advantages over embryo biopsy. First, in the case of a diagnosis error, chromosomally normal oocytes are discarded rather than embryos, but only if the diagnosis is known before fertilization is complete. For example, using a shortened FISH analysis, intracytoplasmic sperm injection (ICSI) can still be performed within 6 hours after retrieval, so that only oocytes diagnosed as normal are selected for injection.68,108 This is a demanding procedure for the IVF clinic, and very tight time constraints require the simultaneous collaboration of gynecologists, embryologists, and biologists from the cytogenetic laboratory. In countries with ultraconservative rulings regarding IVF (i.e. Chile and Italy), this is the only form of PGD that is possible.109 In other countries such as Switzerland and Germany, it is legal to inseminate many oocytes and freeze zygotes. The second PB can be diagnosed followed by zygote freezing, and only those that are normal are allowed to proceed to syngamy. Second, the first and second PBs are not involved in embryo development, and since their removal does not decrease rates of fertilization, cleavage, and blastocyst formation,10 theoretically there should be no negative effect on implantation rates. Third, embryo biopsy has the disadvantage that about 30% of embryos are mosaic, and this feature may result in misdiagnosis in about 5% of cells analyzed (see below). On the other hand, PB analysis cannot detect mosaic embryos, whereas single cell embryo biopsy can screen for a sizable fraction of mosaics. PGD using PBs also has many disadvantages, such as difficulties associated with the interpretation of FISH signals. Univalent chromosomes appear as double-dotted signals, with one dot per chromatid. Since predivision chromatids do frequently occur, single overlapping signals can allow chromatids to be confused with full chromosomes. In addition, predivision appears to increase as an artifact of increasing time in culture, both in first PBs and in oocytes.67 This increase is apparent as early as 6 hours after egg retrieval. The error rate for embryo

biopsy data has been published repeatedly, and in our hands is manageable, whereas the error rate for PB biopsy, measured as the full re-analysis of embryos after PB analysis has not been clearly reported. PB misdiagnosis can also be due to other artifacts, such as loss of chromosomes during fixation or the lack of probe penetration in some forms of chromatin. For example, hybridization errors or loss of chromosomes during fixation can lead to an excess of missing chromatids in the PB, resulting in an excess of 231/2 oocyte diagnoses52,111,112 (Table 18.7). Preinsemination diagnosis of aneuploidy also has the disadavantage that paternally inherited aneuploidies and chromosome abnormalities that occur postzygotically such as polyploidy, haploidy, and some mosaics cannot be detected. These account for about 30% of chromosome abnormalities, and they can be assessed by analyzing blastomeres instead of PBs. On the other hand, embryo biopsy may jeopardize the implantation potential of the embryo. However, to date PGD has been shown to increase implantation rates and decrease spontaneous abortion rates in women of advanced maternal age,96,99 although the increase in implantation obtained in these studies is less significant than was originally thought. One reason could be the effect of embryo biopsy itself, and this has been the focus of recent investigation. Studies have shown that the biopsy of 1 cell at the 8-cell stage of human embryos is not detrimental

Table 18.7 Excess of missing chromatids after polar body (PB) FISH analysis Non-disjunction 231

Predivision 231/2

221/2

37 (29%)

26 (21%)

39 (31%)

Abnormal oocyte karyotype deduced from first PBb 105 (11%) 21 (2%)

617 (65%)

208 (22%)

Abnormal oocyte 25 (20%)

231 karyoypesa

aMamiguchi et al, 1993; Nishino et al, 1994; Dailey et al, 1996; Angell, 1997;

Boiso et al, 1997; Nakaoka et al, 1998; Marquez et al, 1998; Eckel et al, 2000; Mahmood et al, 2000; Rosenbusch et al, in press. bMunné et al, 1995, 2000; Verlinsky et al, 1996, 1999; Dyban et al, 1996. Table adapted from Rosenbusch et al.52



HPE_Chapter18.qxp

7/13/2007

5:29 PM

Page 220


for embryo development to blastocyst stage,113 but the removal of 1/4 of a 4-cell embryo seems to reduce the ratio of inner cell mass : trophectoderm.15 The interpretation of these experiments15,113 is not consistent, and biomass reduction is often discussed in the context of a superfluous cell number and totipotency. Controlled experiments with ongoing pregnancy as an endpoint have not been carried out, so that the effects of biopsy on all stages of development, not just those with the most rapid rate of development, can be ascertained. The question of whether biopsy has a significant effect on the embryo remains unresolved. Cohen et al103 postulated that embryo biopsy affects embryo development prior to transfer, and may have lasting effects after transfer to the uterus. This suggestion is based on the assumption that cell loss from biopsy can be compared with cell loss, after thaw of frozen cleaved embryos. After cryopreservation, individual cells are frequently lost, while remaining cells have the potential to develop into a viable blastocyst, depending on the quality of the embryo and the proportion of cells damaged at thaw. The implantation potential of thawed embryos that are fully intact and have no degenerate cells may resemble that of embryos transferred without cryopreservation.114,115 Using this analogy, implantation potential becomes a function of cell loss, the quality of the original embryo, and the number of cells present at the time of biopsy. According to our EggCyte database, based on data from 75 000 embryos, the average cell number on day 3 is not 8, but 6.7 cells. Therefore another factor for consideration is that the proportional effect of cell loss on implantation potential might be higher than expected. Taking the analogy with cryopreservation further, the removal of a single cell from an 8-cell embryo is expected to reduce implantation potential by 12.5%.114 For the purpose of this discussion, if the expected implantation potential in a hypothetical 8-cell embryo is set at 20%, a single cell biopsy is predicted to lower its implantation probability to 17.5%. Consequently, for preimplantation diagnosis to succeed in increasing implantation rates, it must more than compensate for this initial setback. 

The challenge for PGD as a tool in infertility treatment becomes considerably more difficult to overcome when 2 cells are removed. The loss of 2 cells from an embryo as described above would reduce its implantation potential by 25%, resulting in a 15% chance of implanting versus an initial probability of 20%. In this case, in order for PGD to improve embryo selection, it must bridge a larger deficit if it is to provide any advantage in terms of embryo implantation. On the basis that the average cell number at the time of biopsy is only 6.7, rather than 8.0, an even more pronounced decline in implantation potential might be expected. With a theoretical implantation potential of 20% without biopsy, the implantation potential after a 1- or 2-cell biopsy would diminish to 17.0% and 14.0%, respectively. It is clear from this analysis that a 2-cell biopsy significantly impedes embryo development and is inadvisable in cases where PGD is employed for the purpose of increasing IVF success rates. If a PGD study fails to detect a difference in implantation rate when comparing routine IVF cycles with cycles that employ 2-cell biopsy, this is not an indication that chromosome screening has had no effect. On the contrary, it shows that PGD has succeeded in compensating for the significant reduction in implantation caused by the biopsy of 2 cells. This vital distinction has sometimes been overlooked,101 with resulting misinterpretations of data in subsequent reviews.116,117 A recent study by the same group has shown no difference in the error rate of PGD between 1- and 2-cell biopsy.118 One can conclude from this work that there is little benefit in biopsy of 2 cells. In theory, trophectoderm biopsy at the blastocyst stage should not affect the inner cell mass, and should be less detrimental than the biopsy of 2 cells in cleavage stage embryos; it should provide all the information that cleavage stage biopsy can provide, and more information than PB analysis. In addition, the results should not be affected by mosaicism, since five or more cells are simultaneously analyzed. So far, very few centers have implemented this technique clinically,105 and more data are needed to evaluate whether these advantages can be confirmed. In addition, blastocyst culture is not yet able to sustain

HPE_Chapter18.qxp

7/13/2007

5:29 PM

Page 221


some embryos that might result in live births if replaced on day 3.37,119 BIOPSY TECHNIQUE

Day 3 embryo biopsy of a single cell is our current recommendation, since this is the only method that has consistently shown an improvement in ART outcome. The technique used for biopsy can be mechanical; chemical, using acidified Tyrode’s solution; or physical, by non-contact infrared laser. The mechanical method was adopted by only a few groups, while the laser method, being very quick, precise, and easy to use, has become increasingly popular and is rapidly replacing the chemical method, which requires more experience and skill.120 However, recent studies have demonstrated that a safe working distance for the laser is crucial in order to prevent adverse immediate or long term effects on the development of laser biopsied human embryos.121 Very few studies in the literature report data that compare these methods. A recent study comparing acidified Tyrode’s solution and laser biopsy demonstrated identical pregnancy rates, but a slight increase in the number of damaged cells after the acid technique compared with after the laser.122 These observations suggest that accurate and highly specialized training is mandatory for the correct application of such an invasive procedure, regardless of the method used. One of the primary reasons for non-satisfactory results after PGD for aneuploidy obtained in some centers could be due to inaccurate performance of the biopsy in laboratories with a micromanipulation station that is equipped with a laser.123 FIXATION METHOD

The cell must be fixed after biopsy, in a manner that facilitates proper FISH processing and minimizes errors. The classical method of cell fixation involves acetic acid/methanol or Carnoid;124 the dynamics of nuclear fixation with these classical techniques have been studied mostly to improve metaphase chromosome spreading, and not for the purpose of

interphase FISH analysis. However, the principles are similar, as indicated by Spurbeck et al.125 ‘as the fixative evaporates, the surface tension of the fixative on the cell makes the cell thinner from top to bottom and wider from side to side. The cell continues to get thinner and wider as the fixative evaporates. As the cell widens and reaches its widest diameter potential and then dries, it results in a suitable spread. If the drying rate is too fast, the cell dries before it has a chance to reach the optimum diameter. This results in a tight, compact metaphase where many chromosomes are overlapped. On the other hand, if the drying rate is too slow the cell does not dry and currents in the fixative solution may move the cell and it eventually fixes in an uncontrolled manner.’ These authors found that the proper spreading of metaphase chromosomes was largely dependent on humidity and temperature, and they recommend 50% humidity at 25 C. In our experience with interphase blastomere nuclei, 40–50% humidity and 20 C produce optimal blastomere nuclear spreads. We observed that the ideal diameter is on average 60 m. In nuclei larger than 80 m in diameter (which usually happens in an atmosphere where the humidity is too high) the chromatin is excessively decondensed, with signals that are more widely spread and weaker than those found in regular size nuclei; the signals in these large nuclei are sometimes imperceptible, leading to misdiagnosis. Three methods of fixation are currently in use: method 1: acetic acid/methanol;124,126 method 2: Tween 20/HCl;127 and method 3: Tween 20/HCl and acetic acid/methanol.128 Method 1 can be optimized using the observations of Spurbeck et al.125 Method 2 involves dissolving the cytoplasmic membrane without fixative, and results in tight and condensed nuclei. Method 3 is a combination of methods 1 and 2. Nuclear diameter was previously demonstrated to be inversely correlated with chromosome overlaps and FISH misdiagnosis.129 Velilla et al130 compared the three methods, and found that method 1 produced the largest diameters, and as a result a minimal number of signal overlaps and misdiagnoses, 

HPE_Chapter18.qxp

7/13/2007

5:29 PM

Page 222


while method 2 produced the smallest diameters and the largest number of overlaps and misdiagnoses (p 0.005). However, Carnoy’s method (method 1) can be more difficult to master, and in some hands produces a larger proportion of lost cells,131 but at the same time a larger number of cells that can be analyzed.130,131 In conclusion, Carnoy’s method is more difficult to master, and more cells may initially be lost by novices with the technique, but once it is mastered it produces larger nuclear diameters, which translate into fewer signal overlaps and fewer FISH errors.130 Not surprisingly, PGD laboratories using method 2 usually report high error rates, unmanageable rates of discordance between cells, and no improvement of ART outcome after PGD of infertility.101,132 PROBES

In order to study as many chromosomes as possible, several types of FISH protocols have been used to maximize the use of a limited number of fluorochromes. One approach used ratios of fluorochromes, but this has the disadvantage that overlapping signals from two different chromosomes sharing one or more colors may produce a misdiagnosis. Therefore, single colors are favored, but there are only five colors in the visible spectrum that can be analyzed simultaneously. Thus, in order to study more than five chromosomes, the re-analysis of the same nucleus with different probes was implemented.133 The second set of probes works with high efficiency (95%). Coupled with fast protocols this allows ten chromosomes to be analyzed simultaneously in a single interphase nucleus, within a time frame that is compatible with routine IVF.99,134,135 We have improved the use of a third consecutive hybridization, which is capable of simultaneously analyzing up to 15 chromosomes, with only a 12% error rate.13 The question of which chromosomes should be studied, and whether a minimum number of chromosomes analyzed is needed in order to improve ART outcome needs to be addressed. Analysis of thousands of cleavage stage embryos for 16 different chromosomes (X, Y, 1, 2, 3, 4, 6, 7, 11, 13, 14, 15, 16, 

17, 18, 21, 22), demonstrates that the chromosomes most commonly found in aneuploidies are chromosomes 22 (6.6%), 16 (5.2%), 15 (4.7%), and 21 (4.7%)13,53 (Table 18.3). Thus, the chromosomes with the highest proportion of trisomies or risk of reaching term should be detected by the analysis. The standard panel of probes that has produced an improvement in ART results97,99 contains nine probes, for chromosomes X, Y, 13, 15, 16, 18, 21, 22 plus one more. Studies with fewer probes have not been able to show an improvement in implantation rates, although a reduction in spontaneous abortions has been found.96 MINIMIZING SOURCES OF FISH ERRORS

The use of PGD in infertility has been criticized on the basis that determining the chromosome complement is unpredictable, due to high levels of mosaicism in the early human embryo. Whereas such reservations are sometimes valid, they are not always based on biological phenomena, and technology may play a role in artificially amplifying the true rates of mosaicism. This section deals with the sources of errors that may be caused by technical and biological problems during the analysis of single blastomeres. Polar body and blastocyst biopsy errors are not evaluated here. When a single cell is analyzed, several types of technical problems may result in a misdiagnosis of that cell. These problems include unsuitable probe hybridization, loss of DNA during fixation, signal overlaps, stretched signals, and double chromatids giving the appearance of two close signals. In addition, biological phenomena such as micronucleation and mosaicism may render the cell not representative of the rest of the embryo. Criteria for differentiating between mosaicism and false positives and negatives have been previously described.45 These criteria apply only when all or most of the cells of an embryo are analyzed, and when all of the remaining embryo cells are fixed on day 3, or at the latest, early on day 4. This is due to the fact that some abnormal embryos arrest and degenerate between day 4 and 5, so that the remainder are enriched with normal embryos, and at the same time there is an increase in

HPE_Chapter18.qxp

7/13/2007

5:29 PM

Page 223


polyploidy and production of 2N/4N mosaics between day 3 and day 5.3 Re-analysis performed on day 5 or later usually results in artificially high rates of misdiagnosis.136 In order to keep the proportion of FISH errors under strict control, every center offering PGD for aneuploidy should systematically control their results by re-analyzing the non-transferred embryos, as recommended by the guidelines on good PGD practice.137 Unfortunately, these data are reported in the literature in only a limited number of publications.12,42,45,97,99,101,129,138–140 TECHNICAL SOURCES OF ERRORS

Unsuitable probe hybridization Reduced hybridization because of poor probe penetration or insufficient probe binding is usually due to suboptimal fixation, specifically if cytoplasm remains or the cell did not burst during fixation. This should not occur with proper protocols and fixation techniques, and can be foreseen after fixation by simple phase-contrast observation. Signal overlaps between homologs A false diagnosis of monosomy can result from overlap of chromosome-specific signals for the same chromosome. As reviewed previously, we found that poor spread of the nucleus during fixation and the DNA content in the nucleus increased signal overlap,129 and that some fixation methods produce more overlaps and misdiagnoses than others.130 In a study by Staessen et al,101 using the error-prone Tween 20/HCl method, 19/24 detected errors were false monosomies, even though they analyzed two cells per embryo. Signal overlaps between non-homologs when using color ratios Misdiagnosis may also occur through overlap of different chromosomes labeled with mixtures that share one color. For example, a chromosome labeled in orange could overlap with another labeled in orange and aqua, with the result that the first chromosome is masked by the second. The use of a single color per chromosome analyzed135 results in a

significant reduction in the error rate of false normal and abnormal PGD results, from 8% to 4%. Loss of micronuclei during fixation Although with correct technique DNA is not lost from the main nucleus, micronuclei may be lost during fixation. For example, the FISH error is higher in multinucleated blastomeres (MNBs) (11.5%, 13/113) than in those that are mononucleate (3.1%, 13/415).45,53 This is probably due to the fact that many MNBs very often contain micronuclei, and during fixation, some of them can be more easily lost than full nuclei, producing false negative FISH errors. We found a strong correlation between type of fixation and loss of chromosomes.126 During fixation with Carnoid, drops added before the cell breaks allow the cytoplasm to expand, with increasing expansion with more drops, while adding a drop postlysis removes cytoplasmic debris, and probably some anuclear DNA. It is therefore possible that the loss of DNA is higher after adding a drop postlysis, when the cell is more expanded (2 drops prelysis instead of one) as was observed in this study. Currently we recommend two drops prelysis, followed by no drops postlysis; with appropriate humidity conditions, this allow a good spread, minimal cytoplasmic debris, and minimal loss of DNA. Stretched signals and double chromatids giving the appearance of two close signals Excessive stretching of the DNA during fixation may cause a signal to split, giving a false positive result. This occurs more often with some probes than with others. A false positive result can also occur if replication is non-synchronous during an S-phase, so that one chromosome shows a single signal for its sole chromatid and the other chromosome sends two close signals, one for each chromatid. When the signals are two or more domains (individual signals) apart, we score two close signals as two separate chromosomes: this produced fewer misdiagnoses in our hands.141 RESCUE OF TECHNICAL ERRORS

Technical errors can be avoided by using appropriate fixation protocols. If there is doubt about a 

HPE_Chapter18.qxp

7/13/2007

5:29 PM

Page 224


possible artifact of the technique, this can be ascertained by analyzing the same chromosome twice at two different loci, which will then confirm the correct result.41 In a recent study by Colls et al140 cells with dubious or no results after two rounds of FISH with nine probes were re-analyzed in a third hybridization, using probes that bound to different loci from the original probes used. After analysis of 34 831 blastomeres, 7.5% showed inconclusive results. Re-analysis with additional probes reduced the number of cells with inconclusive results to 3.1% (p 0.001); FISH errors, measured as discrepancies between the PGD diagnosis and the analysis of the non-transferred embryo, decreased from 13.6% to 4.7% (p 0.001). Thus the use of rescue to solve dubious signals before transfer proved to be a powerful tool in reducing the error rate and the frequency of inconclusive results after PGD. Alternatively, biopsied embryos with dubious results may be transferred on day 4,142,143 allowing an extra day for the re-biopsy of another cell from these embryos. An embryo that has been diagnosed as monosomic for chromosomes other than X and 21 that reaches blastocyst stage contrary to expectation,91 could be biopsied as a blastocyst for a second PGD analysis. MOSAICISM AND PGD MISDIAGNOSIS

Published rates of mosaicism vary widely in the literature, due to several factors. (1) Mosaicism error rates are center dependent, and may be influenced by differing intrafollicular and laboratory conditions.16 (2) The frequency of mosaicism is related to the developmental stage of the embryo. Thus, the calculation of PGD errors related to mosaicism should be re-analyzed on the basis of rates in cleavage stage embryos, not blastocysts.136 (3) As discussed above, mosaicism errors must be distinguished from technical errors. Large studies conducted on cleavage stage embryos indicate that rates of mosaicism are in the region of 30%.3,75,144 Munné et al143 reported that 29% of 1903 

discarded embryos analyzed for at least five chromosomes were mosaic. The most common types of mosaicism were chaotic (48%), diploid/polyploid (26%), and those caused by mitotic nondisjunction (25%). The number of abnormal cells per embryo ranged from an average of 44% in diploid/polyploid to 84% in chaotic mosaics. Mosaics have been classified as benign if 1/2 cells from an 8-cell embryo are abnormal, or as detrimental if 3 cells are abnormal.144 This classification is based on the reduction in implantation rate after freeze–thaw cell loss following embryo cryopreservation.145 It is also based on the observation that mosaic embryos with a sizable number of abnormal cells do not develop to blastocyst stage,91 and that the majority of monosomic, haploid, and dinucleated cells arrest in culture.91,146 Thus two different types of PGD misdiagnosis can be due to mosaicism, depending on its extent. The first type of misdiagnosis occurs when a detrimental mosaic is classified as normal. This is estimated to occur in 4.3% of diagnoses,144 and is mainly due to the chaotic and mitotic aneuploid mosaics. The second type of misdiagnosis occurs when a benign mosaic is classified as abnormal, which is estimated to happen in 1.3% of diagnoses; this is mainly due to erroneous classification of a 2N/Pol benign mosaic as normal. With the implementation of correct methodology to eliminate technical errors, and with retesting to correctly interpret dubious signals to reduce no results, the overall PGD error rate when results are obtained, and that can be attributed to mosaicism, ranges from 5%140 to 5.6%144 (Table 16.8). EMBRYO TRANSFER TECHNIQUE

The protocol used for embryo transfer introduces another variable to the survival of the biopsied embryo. Very few reports have explored this issue, and although the majority of the data are anecdotal, they indicate that damage can be done very easily by unsuitable embryo transfer.147 Numerous studies have shown that embryo transfer technique is one of the important factors in ART success, together with maternal age, egg quality, and quality control in the laboratory.148

HPE_Chapter18.qxp

7/13/2007

5:29 PM

Page 225


Table 16.8 Estimated risk of PGD misdiagnosis due to mosaicism. Embryos were fully trisomic or monosomic for a specific chromosome and in addition mosaic for the same chromosome Overall frequency (A)

% Normal cells (B)

Risk of misdiagnosis (A B)

Risk of classifying an abnormal embryo as normal 2N/POL (detrimental) 3.7% (70/1903) Chaotic (detrimental) 12.7% (242/1903) Split (detrimental) 0.3% (5/1903) Mitotic aneuploid (all) 6.6% (126/1903) Meiotic and mitotic aneuploid (all)a 0.6% (12/1903) Total 23.9% (455/1903)

34.8% 9.8% 29.8% 24.2% 12.2% 18.0%

1.3% 1.2% 0.1% 1.6% 0.1% 4.3%

Risk of classifying a mostly normal embryo as abnormal 2N/POL (benign) 3.9% (74/1903) Chaotic (benign) 1.3% (24/1903) Split (benign) 0.2% (3/1903) Total 5.3% (101/1903)

23.1% 24.9% 26.7% 23.6%

0.9% 0.3% 0.1% 1.3%

Total misdiagnosis rate due to mosaicism Data from Munné et

al.144 The

5.6%

risk of misdiagnosis was calculated by multiplying column A by column B.

In addition to technical difficulties during transfer, other factors have been analyzed, including (in order of importance) removal of hydrosalpinges, absence of blood in the mucus, type of catheter (soft better than stiff), no touching of the uterine fundus, use of a tenaculum, removal of all mucus, ultrasonography of the uterus before and during the procedure, leaving the catheter in place for 1 minute, trial transfer, and the administration of prostaglandins to prevent uterine contractions. The presence of pathogenic organisms was also reported to be detrimental, as well as transferring the embryo in a large volume of media.149,150 Significant differences in pregnancy rates have also been reported to depend on the physician performing the transfer (reviewed by Schoolcraft et al149). One center reporting optimal results recommends using a continuous fluid column of 30 mL of transfer media, in a Wallace catheter, attached to a 1 mL airtight syringe, with the embryos loaded towards the tip of the catheter.149 Some of these factors are even more important when the embryo has been biopsied, such as the presence of mucus or blood that may occlude the catheter lumen and squeeze the embryo so that cells are lost.149 In conclusion, in order to maximize results and minimize errors, a single cell should be biopsied on

day 3 of development, fixed using appropriate techniques such as the modified Tarkowsky method,130 with analysis of, at least chromosomes X, Y, 13, 15, 16, 18, 21, and 22. The signals should be read and scored by very well qualified personnel, with each cell analyzed by two people independently; dubious signals or no results should be re-tested with a third set of probes binding to a different locus,140 and the embryo transfer procedure should be performed by doctors who are trained in the transfer of very fragile embryos, such as those with a large hole in the zona due to recent biopsy. Failure to clearly follow these guidelines invariably produces poor PGD results. The use of appropriate techniques as suggested above can improve several aspects of ART results. REDUCTION IN THE INCIDENCE OF TRISOMIC OFFSPRING

Misdiagnoses have been reported in 2/91138 and in 5/434 fetuses98 after PGD. In all cases, re-analysis of the misdiagnosed cells with probes binding to a different locus confirmed prior results, indicating that the errors were probably due to mosaicism, with the exception of one misdiagnosis of trisomy 15; this was attributed to the overlap between a Y 

HPE_Chapter18.qxp

7/13/2007

5:29 PM

Page 226


chromosome labeled with a mixture of aqua and orange, and a chromosome 15 labeled with orange. In spite of these misdiagnoses, the rate of trisomic offspring detected after PGD is lower than expected. For example, only five (1.15%) of 434 fetuses were found to have aneuploidies (trisomy 21, monosomy 21, monosomy X, trisomy 13, trisomy 15),98 whereas the incidence expected in a population of the same maternal age range is 4.7%.151 This represents a four-fold decrease in aneuploid conceptions after PGD for that maternal age group (p 0.001).98 DECREASE IN THE FREQUENCY OF SPONTANEOUS ABORTIONS

The majority of studies assessing risk factors related to pregnancy loss agree that maternal age, followed by previous miscarriage are the major factors involved.152–154 Chromosome abnormalities found in sporadic miscarriages in the general population range from 39% to 76%, depending on the study.71,155–159 Conventional karyotyping requires tissue culture, which is prone to maternal contamination and fails in up to 25% of cases, and more often if the embryo is chromosomally abnormal.158 For these reasons, euploid karyotypes may have been overestimated, particularly in earlier studies. Chromosome studies in spontaneous abortions of ART patients also indicate a high rate of chromosome abnormalities,160 with 71% abnormalities found in spontaneous abortions, and an increase in chromosome abnormalities with maternal age, from 65% in women 39 years and younger to 82% in women 40 years and older. Because of the high rate of chromosome abnormalities in spontaneous abortions, PGD should substantially reduce the rate of miscarriage in infertile patients undergoing ART. Indeed, Jobanputra et al161 reported that FISH with probes for chromosomes 13, 15, 16, 18, 21, 22, X, and Y can detect 83% of chromosomally abnormal fetuses routinely detected by prenatal diagnostic karyotyping. Since this combination of probes is our current standard, PGD should also eliminate close to 80% of all chromosomally abnormal embryos at risk of causing a miscarriage. 

A multicenter study compared controls (of similar age, number of embryos, hormonal response, previous number of cycles) with a test group undergoing embryo biopsy and PGD for aneuploidy of chromosomes X, Y, 13, 18, and 21.96 The results revealed a significant reduction in spontaneous abortions (measuring pregnancy as the presence of a heart beat) from 23% in the controls to 9% in the PGD group (p 0.05). Another study reported a spontaneous abortion rate of only 9% after PGD of aneuploidy for 343 cases in women 36 years,26,138 instead of an expected 16% for this age group. Our most current data98 using PGD for nine chromosomes showed that for age 35–40, the rate of spontaneous abortion was reduced from 19.4% to 14.1% (p 0.025) after PGD, and for patients over 40, from 40.6% to 22.2% (p 0.001). IMPROVED IMPLANTATION, PREGNANCY, AND TAKE-HOME BABY RATES

Three types of comparative trials have investigated PGD for aneuploidy. (1) Trials that used prospective datasets comparing patients who agreed to PGD, with patients who declined PGD.96,99,162 (2) Trials that used prospective data in a randomized fashion.101,132 (3) Trials that used retrospective analyses.95,98 Investigations in the first two categories are considered here. A literature review reveals two contrasting conclusions: one group of investigators supports the hypothesis that PGD for infertility improves implantation and reduces miscarriage rates,96,99,162 resulting in higher take-home-baby rates, while a second group was not able to demonstrate any significant differences between control and PGD patients.101,132 As discussed previously, the lack of positive results results in group 2 may have been due to inappropriate methodology. Two cells were biopsied instead of one, an insufficient number of chromosomes was tested, and fixation methods were used that are prone to error, resulting in poor embryo selection. One of the studies in which only one cell

HPE_Chapter18.qxp

7/13/2007

5:29 PM

Page 227


was biopsied also used a restrictive set of chromosomes,96 and in this case there was no increase in implantation rates, but the incidence of miscarriage was diminished with a subsequent increase in takehome-baby rate. Studies that used a single cell biopsy technique and nine chromosome probes all demonstrated an acceptable efficiency of FISH for detecting aneuploidy, with improved implantation rates.97,99 Sample size was not appropriate to determine an improvement in pregnancy rate. PREGNANCY RATE AND NUMBER OF EMBRYOS AVAILABLE FOR BIOPSY

The importance of the number of embryos available for biopsy has been studied in relation to the effect of PGD on implantation and pregnancy rates.97 Irrespective of the number of embryos available for biopsy, the same doubling of implantation rate was obtained after PGD. However, when patients had fewer than eight embryos available for biopsy, pregnancy rates did not improve (or decrease) after PGD. This is due to the fact that when there is no limit to the number of embryos transferred (such as in the US), if the total number of embryos available is four or less, the majority of the embryos were transferred in the control group, so that the total number of normal embryos transferred was similar in both control and PGD groups. In ART, PGD is a selection tool to improve the outcome; if there are insufficient embryos for selection, pregnancy rates cannot improve. However, even with a limited number of embryos, PGD can be used to select an appropriate number of normal embryos in order to minimize multiple conceptions, and to avoid the transfer of embryos resulting in spontaneous abortions and aneuploid conceptions. Provided that suitable methodology is used, in countries where embryo transfer is restricted by law to one or two embryos, PGD should be even more effective than has been reported from its use in the US. OTHER INDICATIONS FOR PGD OF INFERTILITY

In addition to maternal age of 35 and above as an indication, PGD has been used for other subgroups

of infertile patients, and for patients who suffer recurrent miscarriages. Patients with recurrent pregnancy loss (RPL) have been shown to have a higher chance of becoming pregnant after PGD (64% vs 38%),102 with a higher implantation rate (38% vs 31% in 35, and 26% vs 14% in 35 years old, p 0.01).95 The most important improvement for these patients was a reduction in spontaneous abortions.95 Previously, RPL patients were reported to have lost 87% of their previous pregnancies. After PGD, 16.7% pregnancies were lost (p 0.001). Based on the data of Brigham et al,163 in the RPL group the a priori expected losses were 36.5% compared with the observed loss rate of 16.7% (p 0.028). In the 35 years subgroup, the expected loss in the next pregnancy was 44.5%, whereas the observed loss following PGD was 12% (p 0.007). Our findings indicate that PGD can be recommended to RPL patients who are 35 and older who show no clear etiology for their recurrent miscarriages. Repeated implantation failure (RIF) comprises a poorly defined group of patients. If a hypothetical center had a 50% pregnancy rate in patients 35 years old, by pure chance, 12.5% of their patients would suffer RIF, with no defined criteria to distinguish them from other patients. The incidence of RIF is greater in IVF centers with suboptimal overall pregnancy rates. Thus, the fact that all studies to date reporting that PGD does not significantly improve pregnancy rates in patients 35 years with RIF is not surprising.26,97,99,103,164 Other potential indications have been suggested, based on increased rates of chromosome abnormalities in these subgroups of patients, but a sufficient number of cases to assess whether there is an improvement in ART outcome have not been performed. (1) Patients 35 years old with one previous trisomic conception.165 Although the literature based on prenatal data suggest that this is a sporadic phenomenon, these patients had almost double the rate of aneuploid embryos than the control group. (2) Patients with non-obstructive azoospermia had significantly higher rates of chaotic and mosaic 

HPE_Chapter18.qxp

7/13/2007

5:29 PM

Page 228


embryos than the control group.59,80 These findings are confirmed by FISH studies on sperm, which report that testicular spermatozoa are significantly more prone to aneuploidy than ejaculated spermatozoa.166 Consequently, the possibility exists that for patients with nonobstructive azoospermia, the paternal contribution to aneuploidy in the resulting embryos could be more significant than expected. Preliminary data have indeed indicated a correlation between sperm aneuploidy and the resulting embryos.166 If these data are confirmed on a larger scale, this condition could represent a novel indication for PGD for aneuploidy. (3) The SISMeR group26,139 have studied patients with chromosome mosaicism detected in peripheral blood, under the assumption that they may also be germline mosaics. They found an incidence of 63% abnormal embryos in these patients, which is similar to that found in older patients (71% abnormal). They suggested that mosaicism in peripheral blood may indicate a predisposition to mosaicism in embryos. (4) Counter intuitively, PGD is being offered to egg donors for different reasons: a subset of egg donors have been found to have high rates of embryos with chromosome abnormalities, probably because of the different protocols that are used for ovarian stimulation.167

OTHER CLINICAL IMPACTS OF PGD FOR ANEUPLOIDY

The rationale behind the application of PGD for aneuploidy is also confirmed by the following considerations. REPRODUCTIVE HISTORY OF PGD COUPLES

The majority of couples undergoing PGD for aneuploidy have experienced recurrent pregnancy losses and repeated implantation failures throughout their reproductive history. In a recent study, the clinical impact of PGD for aneuploidy was evaluated in 193 patients who subsequently achieved 208 clinical 

pregnancies in relation to their reproductive history.166 Without PGD, 61 of the 193 patients had previously experienced 112 pregnancies with 105 abortions and seven deliveries, corresponding to 3.6% take-home-baby rate and 10.9% implantation rate. In 68% of the oocyte retrievals, euploid embryos that had been analyzed for five to nine chromosomes were transferred, yielding 171 term pregnancies with 210 infants born; 34 aborted spontaneously and three were ectopic. The resulting take-home-baby rate per pregnant patient was 88.6% and the ongoing implantation rate per pregnant patient was 53.2%. These data suggest that the selection made against chromosomal abnormalities in preimplantation embryos is associated with a significantly higher take-home-baby rate when compared with the previous history experienced by the patients themselves. PROGNOSTIC ROLE OF PGD FOR ANEUPLOIDY IN SUBSEQUENT ART CYCLES

After a first unsuccessful PGD cycle, especially when no euploid embryos were available for transfer, the couple’s first concern is the chance for a pregnancy in a subsequent attempt. In other words, is there a tendency to repeat the same pattern of chromosomal abnormalities, or is it cycle dependent? This query was addressed in a study that included 141 couples with a previous unsuccessful PGD cycle, whose performance was analyzed in 175 subsequent PGD cycles.168 According to the results, patients with no euploid embryos during the first PGD cycle underwent significantly fewer transfers (45%) as compared with patients with at least one euploid embryo in the first PGD cycle (69%, p 0.05), and compared with patients with at least two normal embryos in the first PGD cycle (85%, p 0.001). The pregnancy rate per transfer and the live birth rate per patient was significantly higher in the latter group compared with the others, suggesting that the outcome of the first PGD cycle may have a predictive role on subsequent attempts. This information is important in predicting the outcome of subsequent attempts and in motivating couples with a positive prediction to persist in their quest for a

HPE_Chapter18.qxp

7/13/2007

5:29 PM

Page 229


child. Conversely, patients with a negative prognosis could be assisted in the difficult decision of refraining from further attempts or considering the possibility of resorting to gamete donation.

●

●

CONCLUSION

In conclusion, morphology and developmental selection of embryos can select against some chromosome abnormalities, but even following this selection, more than 50% of cleavage stage and 40% of blastocyst stage embryos with apparently normal development are chromosomally abnormal. PGD could therefore be an added selection tool, provided that it is properly performed. In order to obtain the best results, appropriate fixation techniques should be used, with one-cell biopsy, analysis of eight or more chromosomes, re-testing of dubious results, and suitable culture and embryo transfer techniques. When these criteria are applied, PGD significantly reduces trisomic offspring and spontaneous abortions, and improves implantation rates, with an increase in take-home-baby rates. Further aspects regarding the morphological and genetic selection of embryos are yet to be investigated, including: ●

●

●

●

●

●

Assessment of the rate of chromosome abnormalities in natural cycles Establishment of optimal stimulation protocols that maximize the total number and frequency of euploid embryos Analysis of the impact of atmospheric contamination on rates of euploid embryos Accumulation of CGH, SKY, and other full karyotype data on whole embryos to determine which are the chromosomes most frequently involved in aneuploidy at cleavage stages Re-evaluation of embryonic development in relation to chromosome abnormalities, using eight or more chromosome probes and thousands of embryos Assessment of the error rate after PB analysis, measured as the comparison of abnormalities detected by PB analysis and full re-analysis of non-transferred embryos

●

●

●

Elucidation of whether blastocyst biopsy methods provide a higher implantation potential than PB or embryo biopsy, provided that transfer is carried out on the same day Development of criteria to classify an embryo as normal or abnormal based on blastocyst biopsy Establishment of the optimal transfer technique for biopsied embryos Evaluation of the possible relationship between sperm abnormalities and embryo abnormalities in male factor patients Adapting microarray technology to single cells in order to analyze all chromosomes simultaneously, and in several locus at the same time to provide redundancy, and minimize errors.

REFERENCES 1. Society for Assisted Reproduction and Technology and American Society for Reproductive Medicine Assisted Reproductive Technology in the United States: 2000 results generated from the American Society for Reproductive Medicine/Society for Assisted Reproduction and Technology Registry. Fertil Steril 2004; 81: 1207–20. 2. Munné S, Alikani M, Tomkin G, Grifo J, Cohen J. Embryo morphology, developmental rates and maternal age are correlated with chromosome abnormalities. Fertil Steril 1995; 64: 382–91. 3. Márquez C, Sandalinas M, Bahçe M, Alikani M, Munné S. Chromosome abnormalities in 1255 cleavage-stage human embryos. Reprod Biomed Online 2000; 1: 17–27. 4. Gutierrez-Mateo C, Benet J, Wells D et al. Aneuploidy study of human oocytes first polar body comparative genomic hybridization and metaphase II fluorescence in situ hybridization analysis. Hum Reprod 2004; 19: 2859–68. 5. Griffin DK, Handyside AH, Penketh RJA, Winston RML, Delhanty JDA. Fluorescent in situ hybridization to interphase nuclei of human preimplantation embryos with X and Y chromosome specific probes. Hum Reprod 1991; 6: 101–5. 6. Munné S, Weier HUG, Stein J, Grifo J, Cohen J. A fast and efficient method for simultaneous X and Y in situ hybridization of human blastomeres. J Assist Reprod Genet 1993; 10: 82–90. 7. Munné S, Weier HUG, Grifo J, Cohen J. Chromosome mosaicism in human embryos. Biol Reprod 1994; 51: 373–9. 8. Márquez C, Cohen J, Munné S. Chromosome identification on human oocytes and polar bodies by spectral karyotyping. Cytogenet Cell Genet 1998; 81: 254–8. 9. Wells, D, Delhanty, JDA. Comprehensive chromosomal analysis of human preimplantation embryos using whole genome amplification and single cell comparative genomic hybridization. Mol Hum Reprod 2000; 6: 1055–62. 10. Voullaire L, Wilton L, McBain J, Callaghan T, Williamson R. Chromosome abnormalities identified by comparative genomic hybridization in embryos from women with repeated implantation failure. Mol Hum Reprod 2002; 8: 1035–41. 11. Wilton L, Voullaire L, Sargeant P, Williamson R, McBain J. Preimplantation aneuploidy screening using comparative genomic hybridization or fluorescence in situ hybridization of embryos from



HPE_Chapter18.qxp

7/13/2007

5:30 PM

Page 230


12.

13.

14.

15.

16. 17.

18.

19.

20. 21.

22.

23.

24.

25. 26. 27.

28. 29.

30.

31.



patients with recurrent implantation failure. Fertil Steril 2003; 80: 860–8. Munné S, Márquez C, Magli C, Morton P, Mitual. Scoring criteria for preimplantation genetic diagnosis of numerical abnormalities for chromosomes X,Y, 13, 16, 18 and 21. Mol Hum Reprod 1998; 4: 863–70. Abdelhadi I, Colls P, Sandalinas M, Escudero T, Munné S. Preimplantation genetic diagnosis of numerical abnormalities for 13 chromosomes. Reprod Biomed Online 2003; 6: 226–31. Wang JX, Norman RJ, Wilcox AJ. Incidence of spontaneous abortion among pregnancies produced by assisted reproductive technology. Hum Reprod 2004; 19: 272. Tarin JJ, Conaghan J, Winston RML, Handyside AH. Human embryo biopsy on the 2nd day after insemination for preimplantation diagnosis: removal of a quarter of embryo retards cleavage. Fertil Steril 1992; 58: 970–6. Munné S, Magli C, Adler A et al. Treatment-related chromosome abnormalities in human embryos. Hum Reprod 1997; 12: 780–4. Munné S. Chromosome abnormalities and their relationship to morphology and development of human embryos. Reprod Biomed Online 2006; 12: 234–53. Jackson KV, Ginsburg ES, Hornstein D, Rein MS, Clarke RN. Multinucleation in normally fertilized embryos is associated with an accelerated ovulation induction response and lower implantation and pregnancy rates in in vitro fertilization-embryo transfer cycles. Fertil Steril 1998; 70: 60–6. Van Blerkom J, Antezak J, Schrader R. The developmental potential of the human oocyte is related to the dissolved oxygen content of follicular fluid: association with vascular endothelial growth factor levels and perifollicular blood flow characteristics. Hum Reprod 1997; 12: 1047–55. Van Royen E, Mangelschots K, Vercruyssen M et al. Multinucleation in cleavage stage embryos. Hum Reprod 2003; 18: 1062–9. Pellicer A, Ruiz A, Castellvi RM et al. Is the retrieval of high numbers of oocytes desirable in patients treated with gonadotropin-releasing hormone analogues (GnRHa) and gonadotropins? Hum Reprod 1989; 4: 536–40. Nogueira D, Staessen C, Van de Velde H, Van Steirteghem A. Nuclear status and cytogenetics of embryos derived from in vitro matured oocytes. Fertil Steril 2000; 74: 295–8. Weghofer A, Munné S, Chen S et al. Lack of association between polycystic ovary syndrome and embryonic aneuploidy. Fertil Steril 2007; in press. Freeman SB, Yang Q, Allran K, Taft LF, Sherman SL. Women with a reduced ovarian complement may have an increased risk for a child with Down syndrome. Am J Hum Genet 2000; 66: 1680–3. Kline J, Kinney A, Levin B, Warburton D. Trisomic pregnancy and early age of menopause. Am J Hum Genet 2000; 67: 395–404. Gianaroli L, Magli MC, Ferraretti AP et al. The role of preimplantation diagnosis for aneuploidy. Reprod Biomed Online 2001; 4: 31–6. Simon C, Cano F, Valbuena D, Remohi J, Pellicer A. Clinical evidence for a detrimental effect on uterine receptivity of high estradiol concentrations in high and normal responder patients. Hum Reprod 1995; 10: 2432–7. Ng et al. 2000. Reis Soares S, Rubio C, Rodrigo L et al. High frequency of chromosomal abnormalities in embryos obtained from oocyte donation cycles. Fertil Steril 2003; 80: 656–7. Sadowy S, Tomkin G, Munné S, Ferrara-Congedo T, Cohen J. Impaired development of zygotes with uneven pronuclear size. Zygote 1998; 6: 137–41. Homburg R, Jacobs HS. Etiology of miscarriage in polycystic ovarian syndrome. Fertil Steril 1989; 51: 196–8.

32. Pickering SJ, Braude PR, Johnson MH, Cant A, Currie J. Transient cooling to room temperature can cause irreversible disruption of the meiotic spindle in the human oocyte. Fertil Steril 1990; 54: 102–8. 33. Aman RR, Parks JE. Effects of cooling and rewarming on the meiotic spindle and chromosomes of in vitro-matured bovine oocytes. Biol Reprod 1994; 50: 103–10. 34. Cohen J, Gilligan A, Esposito W, Schimmel T, Dale B. Ambient air and its potential effects on conception. Hum Reprod 1997; 12: 1742–9. 35. Boone WR, Johnson JE, Locke AJ, Crane MM, Price TM. Control of air quality in an assisted reproductive technology laboratory. Fertil Steril 1999; 71: 150–4. 36. Hardarson T, Hanson C, Sjogren A, Lundin K. Human embryos with uneven sized blastomeres have lower pregnancy and implantation rates: indications for aneuploidy and multinucleation. Hum Reprod 2001; 16: 313–8. 37. Alikani M, Calderon G, Tomkin G et al. Cleavage anomalies in human embryos and survival after prolonged culture in vitro. Hum Reprod 2000; 15: 2634–43. 38. Racowsky C, Combelles CMH, Nureddin A et al. Day-3 and day-5 morphological predictors of embryo viability. Reprod Biomed Online 2003; 6: 323–31. 39. Terriou P, Sapin C, Giorgietti C et al. Embryo score is a better predictor of pregnancy than the number of transferred embryos or female age. Fertil Steril 2001; 75: 525–31. 40. Plachot M, Junca AM, Mandelbaum J et al. Chromosome investigations in early life. II. Human preimplantation embryos. Hum Reprod 1987; 1: 29–35. 41. Munné S and Cohen J. Chromosome abnormalities in human embryos. Hum Reprod Update 1998; 4: 842–55. 42. Magli MC, Sandalinas M, Escudero T et al. Double locus analysis of chromosome 21 for preimplantation genetic diagnosis of aneuploidy. Prenat Diagn 2001; 21: 1080–5. 43. Munné S, Sandalinas M, Cohen J. Chromosome abnormalities in human embryos. In: Textbook of Assisted Reproductive Techniques. Laboratory and Clinical Perspectives. Gardner DK, Weissman A, Howles CM, Shoham Z (eds), Martin Dunitz Ltd, London. Pp 297–318. 44. Meriano J, Clark C, Cadesky K, Laskin CA. Binucleated and micronucleated blastomeres in embryos derived from human assisted reproduction cycles. Reprod Biomed Online 2004; 9: 511–20. 45. Munné S, Grifo J, Cohen J, Weier HUG. Chromosome abnormalities in arrested human preimplantation embryos: a multiple probe fluorescence in situ hybridization (FISH) study. Am J Hum Genet 1994; 55, 1: 150–9. 46. Kligman I, Benadiva C, Alikani M, Munné S. The presence of multinucleated blastomeres in human embryos correlates with chromosomal abnormalities. Hum Reprod 1996; 11: 1492–8. 47. Laverge H, De Sutter P, Verschraegen-Spae MR, De Paepe A, Dhont M. Triple colour fluorescent in situ hybridization for chromosomes X, Y and 1 on spare human embryos. Hum Reprod 1997; 12: 809–14. 48. Staessen C, Van Steirteghem AC. The genetic constitution of multinucleated blastomeres and their derivative daughter blastomeres. Hum Reprod 1998; 13: 1625–31. 49. Giorgietti C, Terriou P, Auquier P et al. Embryo score to predict implantation after in vitro fertilization: based on 957 single embryo transfers. Hum Reprod 1995; 10: 2427–31. 50. Munné S, Sandalinas M, Alikani M, Cohen J. Chromosome abnormalities in human embryos. Gardner, Weissman, Howles, and Shoham (eds), Textbook of ART, 2nd edn. London: Martin Dunitz, 2004: 355–77. 51. Balakier H, Bouman D, Sojecki A, Librach C, Squire JA. Morphological and cytogenetic analysis of human giant oocytes and giant embryos. Hum Reprod 2002; 17: 2394–401.

HPE_Chapter18.qxp

7/13/2007

5:30 PM

Page 231


52. Rosenbusch B, Schneider M, Glaeser B, Brucker C. Cytogenetic analysis of giant oocytes and zygotes to assess their relevance for the development of digynic triploidy. Hum Reprod 2002; 17: 2388–93. 53. Munné S, Alikani M, Grifo J, Cohen J. Monospermic polyploidy and atypical embryo morphology. Hum Reprod 1994; 9: 506–10. 54. Edwards et al. 1984. 55. Gamiz et al. 2003. 56. Coskun et al. 2003. 57. Scott L, Alvero R, Leondires M, Bradley M. The morphology of human pronuclear embryos is positively related to blastocyst development and implantation. Hum Reprod 2000; 15: 2394–403. 58. Wittemer C, Bettahar-Lebugle K, Ohl J et al. Zygote evaluation: an efficient tool for embryo selection. Hum Reprod 2000; 15: 2591–7. 59. Gianaroli L, Magli MC, Ferraretti AP, Fortini D, Grieco N. Pronuclear morphology and chromosomal abnormalities as scoring criteria for embryo selection. Fertil Steril 2003; 80: 341–9. 60. Kahraman S, Kumtepe Y, Sertyel S et al. Pronuclear morphology scoring and chromosomal status of embryos in severe male infertility. Hum Reprod 2002; 17: 3193–200. 61. Balaban B, Yakin K, Urman B, Isiklar A, Tesarik J. Pronuclear morphology predicts embryo development and chromosome constitution. Reprod Biomed Online 2004; 8: 695–700. 62. Gianaroli L, Magli Mc, Ferraretti AP et al. Oocyte euploidy, pronuclear zygote morphology and embryo chromosomal complement. Hum Reprod 2007; 22: 241–9. 63. James AN, Hennessy S, Reggio B et al. The limited importance of pronuclear scoring of human zygotes. Hum Reprod 2006; 21: 1599–604. 64. Ziebe S, Petersen K, Lindenberg S et al. Embryo morphology or cleavage stage: how to select the best embryos for transfer after in vitro fertilization. Hum Reprod 1997; 12: 1545–9. 65. Graham et al. 2000. 66. Shapiro et al. 2000. 67. Munné S, Dailey T, Sultan KM, Grifo J, Cohen J. The use of first polar bodies for preimpantation diagnosis of aneuploidy. Hum Reprod 1995; 10: 1015–21. 68. Magli MC, Gianaroil L, Ferraetti AP et al. Embryo morphology and development is dependent on the chromosomal complement. Fertil Steril 2006; 86: 629–35. 69. Hassold T, Chen N, Funkhouser J et al. Jacobs PA. A cytogenetic study of 1000 spontaneous abortuses. Ann Hum Genet 1980; 44: 151–78. 70. Hassold T, Chiu D. Maternal age-specific rates of numerical chromosome abnormalities with special reference to trisomy. Hum Genet 1985; 70: 11–7. 71. Warburton D, Kline J, Stein Z, Strobino B. Cytogenetic abnormalities in spontaneous abortions of recognized conceptions. In: Porter IH, Willey A, eds. Perinatal genetics: diagnosis and treatment. New York: Academic Press, 1986: 133–48. 72. Braude P, Bolton V, Moore S. Human gene expression first occurs between the four- and eight-cell stages of preimplantation development. Nat 1988; 333: 459–61. 73. Tesarik J, Kopecny V, Plachot M, Mandelbaum J. Early morphological signs of embryonic genome expression in human preimplantation development as revealed by quantitative electron microscopy. Dev Biol 1988; 128: 15–20. 74. Laverge H,Van der Elst J, De Sutter P et al. Fluorescent in situ hybridization on human embryos showing cleavage arrest after freezing and thawing. Hum Reprod 1998; 13: 425–9. 75. Bielanska M, Tan SL, Ao A. Chromosomal mosaicism throughout human preimplantation development in vitro: incidence, type, and relevance to embryo outcome. Hum Reprod 2002; 17: 413–19. 76. Artley JK, Braude PR, Johnson MH. Gene activity and cleavage arrest in human pre-embryos. Hum Reprod 1992; 7: 1014–21.

77. Winston N, Johnson M, Pickering S, Braude P. Parthenogenetic activation and development of fresh and aged human oocytes. Fertil Steril 1991; 56: 904–12. 78. Palermo G, Munné S, Cohen J. The human zygote inherits its mitotic potential from the male gamete. Hum Reprod 1994; 9: 1220–5. 79. Hewitson L, Simerly C, Schatten G. Inheritance defects of the sperm centrosome in humans and its possible role in male infertility. Int J Androl 1997; 20 (Suppl 3): 35-43. 80. Silber S, Escudero T, Lenahan K et al. Chromosomal abnormalities in embryos derived from TESE. Fertil Steril 2003; 79: 30–8. 81. Wilding M, De Placido G, De Matteo L et al. Chaotic mosaicism in human preimplantation embryos is correlated with a low mitochondrial membrane potential. Fertil Steril 2003; 79: 340–6. 82. Evsikov S, Verlinsky Y. Mosaicism in the inner cell mass of human blastocysts. Hum Reprod 1998; 11: 3151–5. 83. Clouston HJ, Fenwick J, Webb AL et al. Detection of mosaic and nonmosaic chromosome abnormalities in 6- to 8-day old human blastocysts. Hum Genet 1997; 101: 30–6. 84. Veiga A, Gil Y, Boada M et al. Confirmation of diagnosis in preimplantation genetic diagnosis (PGD) through blastocyst culture: preliminary experience. Prenat Diagn 1999; 19: 1242–7. 85. Ruangvutilert P, Delhanty JDA, Serhal P et al. FISH analysis on day-5 post-insemination of human arrested and blastocyst stage embryos. Prenat Diagn 2000; 20: 552–60. 86. Bielanska M, Tan SL, Ao A. High rate of mixoploidy among human blastocysts cultured in vitro. Fertil Steril 2002; 78: 1248. 87. Bielanska M, Jin S, Bernier M, Tan SL, Ao A. Diploid-aneuploid mosaicism in human embryos cultured to the blastocyst stage. Fertil Steril 2005; 84: 336–42. 88. Hardarson T, Caisander G, Sjogren A et al. A morphological and chromosomal study of blastocysts developing from morphologically suboptimal human pre-embryos compared with control blastocysts. Hum Reprod 2003; 18: 399–407. 89. Clouston HJ, Herbert M, Fenwick J, Murdoch AP, Wolstenholme J. Cytogenetic analysis of human blastocysts. Prenat Diagn 2002; 22: 1143–52. 90. Magli MC, Jones GM, Gras L et al. Chromosome mosaicism in day-3 aneuploid embryos that develop to morphologically normal blastocysts in vitro. Hum Reprod 2000; 15: 1781–6. 91. Sandalinas M, Sadowy S, Alikani M et al. Developmental ability of chromosomally abnormal human embryos to develop to the blastocyst stage. Hum Reprod 2001; 16: 1954–8. 92. Janny L, Menezo YJR. Maternal age effect on early human embryonic development and blastocyst formation. Mol Reprod Devel 1996; 45: 31–7. 93. Rubio C, Simon C, Vidal F et al. Chromosomal abnormalities and embryo development in recurrent miscarriage couples. Hum Reprod 2003; 18: 182–8. 94. Eiben B, Bartels I, Bahr-Porsch S et al. Cytogenetic analysis of 750 spontaneous abortions with the direct-preparation method of chorionic villi and its implications for studying genetic causes of pregnancy wastage. Am J Hum Genet 1990; 47: 656–63. 95. Munné S, Chen S, Fischer J et al. Preimplantation genetic diagnosis reduces pregnancy loss in women 35 and older with a history of recurrent miscarriages. Fertil Steril 2005; 84: 331–5. 96. Munné S, Magli C, Cohen J et al. Positive outcome after preimplantation diagnosis of aneuploidy in human embryos. Hum Reprod 1999: 14: 2191–9. 97. Munné S, Sandalinas M, Escudero T et al. Improved implantation after preimplantation genetic diagnosis of aneuploidy. Reprod Biomed Online, 2003; 7: 91–7.



HPE_Chapter18.qxp

7/13/2007

5:30 PM

Page 232


98. Munné S, Fischer J, Warner A, Chen S, Zouves C, Cohen J, and referring centers PGD group. Preimplantation genetic diagnosis significantly reduces pregnancy loss in infertile couples: a multi-center study. Fertil Steril 2006; 85: 326–32. 99. Gianaroli L, Magli C, Ferraretti AP, Munné S. Preimplantation diagnosis for aneuploidies in patients undergoing in vitro fertilization with a poor prognosis: identification of the categories for which it should be proposed. Fertil Steril 1999; 72: 837–44. 100. Verlinsky Y, Tur-Kaspa I, Cieslak J et al. Preimplantation testing for chromosomal disorders improves reproductive outcome of poor prognosis patients. Reprod BioMed Online 2005; 11: 219–5. 101. Staessen C, Platteau P, Van Assche E et al. Comparison of blastocyst transfer with or without preimplantation genetic diagnosis for aneuploidy screening in couples with advanced maternal age: a prospective randomized controlled trial. Hum Reprod 2004; 19: 2849–58. 102. Werlin L, Rodi I, DeCherney A et al. Preimplantation genetic diagnosis (PGD) as both a therapeutic and diagnostic tool in assisted reproductive technology. Fertil Steril 2003; 80: 467–8. 103. Cohen J, Wells D, Munné S. Removal of two cells from cleavage stage embryos is likely to reduce the efficacy of chromosomal tests employed to enhance implantation rates. Fertil Steril 2007; in press. 104. Verlinsky Y, Cieslak J, Frieidine M et al. Pregnancies following preconception diagnosis of common aneuploidies by fluorescence in situ hybridization. Hum Reprod 1995; 10: 1923–7. 105. De Boer KA, Catt JW, Jansen RPS, Leigh D, McArthur S. Moving to blastocyst biopsy for preimplantation genetic diagnosis and single embryo transfer at Sydney IVF. Fertil Steril 2004; 82: 295–8. 106. Hassold TJ, Takaesu N. Analysis of nondisjunction in human trisomic spontaneous abortions. In: Hassold TJ, Epstein CJ, eds. Molecular and Cytogenetic Studies of Nondisjunction. New York: Alan Liss Inc, 1989: 115–34. 107. Antonorakis SE, Lewis JG, Adelsberg PA et al. Parental origin of the extra chromosome in trisomy 21 revisited: DNA polymorphism analysis suggests maternal origin in 95% of cases. N Engl J Med 1991; 324: 872–6. 108. Munné S, Sepulveda S, Balmaceda J, Fernandez E, Fabres C, Mackenna A, Lopez T, Crosby JA, Zegers-Hochschild F. Selection of the most common chromosome abnormalities in oocytes prior to ICSI. Prenat Diagn 2000; 20: 582–6. 109. Benagiano G, Gianaroli L. The new Italian IVF legislation. Reprod Biomed Online 2004; 9: 117–25. 110. Verlinsky Y, Kuliev AM, eds. Preimplantation Diagnosis of Genetic Diseases: A New Technique in Assisted Reproduction. New York: Wiley-Liss, 1993. 111. Verlinsky Y, Kuliev A. Preimplantation diagnosis of common aneuploidies in fertile couples of advanced maternal age. Hum Reprod 1996; 11: 2076–7. 112. Verlinsky Y, Cieslkak J, Ivanhnenko V et al. Prevention of age-related aneuploidies by polar body testing of oocytes. J Assist Reprod Genet 1999; 16: 165–9. 113. Hardy K, Martin KL, Leese HJ, Winston RML, Handyside AH. Human preimplantation development in vitro is not adversely affected by biopsy at the 8-cell stage. Hum Reprod 1990; 5, 6: 708–14. 114. Edgar DH, Bourne H, Jericho H, McBain JC. The developmental potential of cryopreserved human embryos. Mol Cell Endocrinol 2000; 169: 69–72. 115. Edgar DH, Bourne H, Speirs AL, McBain JC. A quantitative analysis of the impact of cryopreservation on the implantation potential of human early cleavage stage embryos. Hum Reprod 2000; 15: 175–9. 116. Shahine LK, Cedars MI. Preimplantation genetic diagnosis does not increase pregnancy rates in patients at risk for aneuploidy. Fertil Steril 2006; 85: 51–6.



117. Twisk M, Mastenbroek S, van Wely M, Heineman M, Van der Veen F, Repping S. Preimplantation genetic screening for abnormal number of chromosomes (aneuploidies) in in vitro fertilisation or intracytoplasmic sperm injection. Cochrane Database Syst Rev 2006; 25: CD005291. 118. Michiels A, Van Assche E, Liebaers I, Van Steirteghem A, Staessen C. The analysis of one or two blastomeres for PGD using fluorescence in-situ hybridization. Hum Reprod 2006; 21: 2396–402. 119. Racowski et al. 2000. 120. Harper JC, Boelaert K, Geraedts J et al. ESHRE PGD Consortium data collection V: Cycles from January to December 2002 with pregnancy follow-up to October 2003. Hum Reprod 2006; 21: 3–21. 121. Chatzimeletiou K, Morrison EE, Panagiotidis Y et al. Comparison of the effects of zona drilling by non-contact infrared laser or acid Tyrode’s on the development of human biopsied embryos as revealed by blastomere viability, cytoskeletal analysis and molecular cytogenetics. Reprod BioMed Online 2005; 11: 697–710. 122. Joris H, De Vos A, Janssens R, Devroey P, Liebaers L, Van Steirteghem A. Comparison of the results of human embryo biopsy and outcome of PGD after zona drilling using acid tyrode medium or laser. Hum Reprod 2003; 18: 1896–902. 123. Malter et al. 2001. 124. Tarkowski AK. An air drying method for chromosome preparations from mouse eggs. Cytogenetics 1966; 5: 394–400. 125. Spurbeck JL, Zinsmeister AR, Meyer KJ, Jalal SM. Dynamics of chromosome spreading. Am J Med Genet 1996; 61: 387–93. 126. Munne et al. 1998f. 127. Coonen E, Harper JC, Ramaekers FCS et al. Presence of chromosomal mosaicism in abnormal preimplantation embryos detected by fluorescence in situ hybridization. Hum Genet 1994; 94: 609–15. 128. Dozortsev DI, McGinnis KT. An improved fixation technique for fluorescence in situ hybridization for preimplantation genetic diagnosis. Fertil Steril 2001; 76(1): 186–8. 129. Munné S, Dailey T, Finkelstein M, Weier HUG. Reduction in signal overlap results in increased fish efficiency: implications for preimplantation genetic diagnosis. J Assist Reprod Genet 1996; 13: 149–56. 130. Velilla E, Escudero T, Munné S. Blastomere fixation techniques and risk of misdiagnosis for PGD of aneuploidy. Reprod Biomed Online 2002; 4: 210–7. 131. Xu K, Huang T, Tiezheng L, Zhongmig S, Rosenwaks Z. Improving the fixation method for preimplantation genetic diagnosis for fluorescent in-situ hydridization. J Assist Reprod Genet 1998; 15: 570–4. 132. Platteau P, Staessen C, Michiels A, Van Steirteghem A, Liebaers I, Devroey P. Preimplantation genetic diagnosis for aneuploidy screening in patients with unexplained recurrent miscarriages. Fertil Steril 2005; 83: 393–7. 133. Benadiva CA, Kligman I, Munné S. Aneuploidy 16 in human embryos increases significantly with maternal age. Fertil Steril 1996; 666: 248–55. 134. Munne et al. 1998c. 135. Bahçe M, Escudero T, Sandalinas M. Improvements of preimplantation diagnosis of aneuploidy by using microwave-hybridization, cell recycling and monocolor labeling of probes. Mol Hum Reprod 2000; 9: 849–54. 136. Li M, DeUgarte CM, Surrey M, Danzer H, DeCherney A, Hill DL. Fluorescence in situ hybridization reanalysis of day-6 human blastocysts diagnosed with aneuploidy on day 3. Fertil Steril 2005; 84: 1395–400. 137. The Preimplantation Genetic Diagnosis International Society (PGDIS) Guidelines for good practice in PGD. Reprod Biomed Online 2004; 9: 430–4. 138. Gianaroli L, Magli MC, Ferraretti AP. The in vivo and in vitro efficiency and efficacy of PGD for aneuploidy. Mol Cell Endocrinol 2001; 183: S13–18.

HPE_Chapter18.qxp

7/13/2007

5:30 PM

Page 233


139. Magli MC, Gianaroli L, Ferraretti AP, Gordts S, Feliciani E. Impact of parental gonosomal mosaicism detected in peripheral blood on preimplantation embryos. Reprod Biomed Online 2002; 5: 306–12. 140. Colls P, Escudero T, Zheng X et al. Increased efficiency of preimplantation genetic diagnosis (PGD) for infertility through reanalysis of dubious signals. Fertil Steril 2007; in press. 141. Munné S, Weier, U. Simultaneous enumeration of chromosomes 13, 18, 21, X and Y in interphase cells for preimplantation genetic diagnosis of aneuploidy. Cytogenet Cell Genet 1996; 75: 263–70. 142. Grifo JA, Giatras K, Tang YX, Krey LC. Successful outcome with day 4 embryo transfer after preimplantation diagnosis for genetically transmitted diseases. Hum Reprod 1998; 13: 1656–9. 143. Gianaroli L, Magli MC, Munné S, Fortini D, Ferraretti AP. Advantages of day 4 embryo transfer in patients undergoing preimplantation genetic diagnosis of aneuploidy. J Assist Reprod Genet 1999; 16: 170–5. 144. Munne et al. 2002a. 145. El-Toukhy T, Khalaf Y, Al-Darazi K et al. Effect of blastomere loss on the outcome of frozen embryo replacement cycles. Fertil Steril 2003; 79: 1106–11. 146. Hardy K, Winston RML, Handyside AH. Binucleate blastomeres in preimplantation human embryos in vitro: failure of cytokinesis during early cleavage. J Reprod Fertil 1993; 98: 549–8. 147. Cohen J. Zona pellucida micromanipulation and consequences for embryonic development and implantation. In: Cohen J, Malter HE, Talansky BE, Grifo J eds. Micromanipulation of Human Gametes and Embryos. New York: Raven Press, 1992; 191–222. 148. Krey LC, Grifo JA. Poor embryo quality: the answer lies (mostly) in the egg. Fertil Steril 2001; 57: 156–62. 149. Schoolcraft WB, Surrey ES, Gardner DK. Embryo transfer: techniques and variables affecting success. Fertil Steril 2001; 863–70. 150. Levi Setti PE, Albani E, Cavagna M, Bulletti C, Colombo GV, Negri L. The impact of embryo transfer on implantation – a review. Placenta 2003; 24: S20–6. 151. Eiben B, Goebel R, Hansen S, Hammans W. Early amniocentesis. A cytogenetic evaluation of over 1500 cases. Prenat Diagn 1994; 14: 497–501. 152. Regan L, Braude PR, Trembath PL. Influence of past reproductive performance on risk of spontaneous abortion. Br Med J 1989; 299: 541–5. 153. La Rochebrochard E, Thonneau P. Paternal age and maternal age are risk factors for miscarriage; results of a multicentre European study. Hum Reprod 2000; 17: 1649–56. 154. Kupka MS, Dorn C, Montag M, Felberbaum RE, Van der Ven H, Kulczycki A, Friese K. Previous miscarriages influence IVF and intra-

155.

156.

157.

158. 159.

160.

161.

162.

163.

164.

165.

166.

167.

168.

cytoplasmic sperm injection pregnancy outcome. Reprod Biomed Online 2004; 8: 349–57. Jacobs PA, Hassold TJ. Chromosome abnormalities: origin and etiology in abortions and live births. In Vogal F, and Sperling K (eds) Human Genetics, Springer-Verlag, Berlin, 1987; pp 233–44. Stern JJ, Dorfman AD, Gutierrez-Najar MD. Frequency of abnormal karyotype among abortuses from women with and without a history of recurrent spontaneous abortions. Fertil Steril 1996; 65: 250–3. Ogasawara M, Aoki K, Okada S, Suzumori K. Embryonic karyotype of abortuses in relation to the number of previous miscarriages. Fertil Steril 2000; 73: 300–4. Carp H, Toder V, Aviram A et al. Karyotype of the abortuse in recurrent miscarriages. Fertil Steril 2001; 75: 678–82. Stephenson MD, Awartani KA, Robinson WP. Cytogenetic analysis of miscarriages from couples with recurrent miscarriage: a case-control study. Hum Reprod 2002; 17: 446–51. Spandorfer SD, Davis OK, Barmat LI, Chung PH, Rosenwaks Z. Relationship between maternal age and aneuploidy in in-vitro fertilization pregnancy loss. Fertil Steril 2004; 81: 1265–9. Jobanputra V, Sobrino A, Kinney A, Kline J, Warburton D. Multiplex interphase FISH as a screen for common aneuploidies in spontaneous abortions. Hum Reprod 17: 1166–70. Munné S, Sandalinas M, Escudero T, Marquez C, Cohen J. Chromosome mosaicism in cleavage stage human embryos: evidence of a maternal age effect. Reprod Biomed Online 2002; 4: 223–32. Brigham SA, Colon C, Farquharson RG. A longitudinal study of pregnancy outcome following idiopathic recurrent miscarriage. Hum Reprod 1999; 14: 2868–71. Pehlivan T, Rubio C, Rodrigo L, Romero J, Remohi J, Simon C, Pellicer A. Impact of preimplantation genetic diagnosis on IVF outcome in implantation failure patients. Reprod Biomed Online 2002; 6: 232–7. Munné S, Sandalinas M, Magli C, Gianaroli L, Cohen J, Warburton D. Increased rate of aneuploid embryos in young women with previous aneuploid conceptions. Prenat Diagn 2004; 24: 638–47. Gianaroli L, Magli MC, Cavallini G et al. Frequency of aneuploidy in spermatozoa from patients with extremely severe male factor infertility. Hum Reprod 2005; 20: 2140–52. Munné S, Ary J, Zouves C, Escudero T, Barnes F, Cinioglu C, Ary B, Cohen J. Wide range of chromosome abnormalities in the embryos of young egg donors. Reprod Biomed Online, 2006; 12: 340–6. Ferraretti AP, Magli MC, Kopcow L, Gianaroli L. Prognostic role of preimplantation genetic diagnosis for aneuploidy in assisted reproduction technology outcome. Hum Reprod 2004; 19: 694–9.



HPE_Chapter18.qxp

7/13/2007

5:30 PM

Page 234

HPE_Chapter19.qxp

7/14/2007

5:51 PM

Page 235

19. Genomic imprinting and consequences for embryonic development Henry E Malter

EPIGENETICS AND ASSISTED REPRODUCTIVE TECHNOLOGY

Our understanding of epigenetic phenomena has progressed essentially in parallel with the application of assisted reproductive technology (ART) over the past three decades. It could be argued that had the current knowledge base concerning mammalian imprinting and epigenetics been known in advance of the application of ART in humans, the latter might have been viewed as at best questionable, and at worst, dangerous: some authors now suggest that this is in fact the case. However, it is difficult to reconcile the birth of millions of healthy human babies after ART with the presence of widespread or serious problems related to epigenetic phenomena. Nevertheless, proper epigenetic processing is a critical component of mammalian development. Much of this processing occurs during stages that concern ART, during the development of the gametes and early embryos, and epigenetic processing has an undisputed association with aberrant development and serious disease states in the human, including late onset diseases such as cancer.1 Furthermore, our understanding of epigenetic phenomena is far from complete, particularly as it relates to our understanding of the regulation of epigenetic phenomena, natural variation among individuals, developmental stages of similar age, and environmental triggers. This chapter reviews current information about epigenetic phenomena as it relates to human assisted reproduction. In the final analysis, it must be accepted that the jury is still out on this matter. While there is certainly some cause for concern, a balanced analysis of the data suggests that a clear connection between human ART techniques and epigenetic-related problems remains to be demonstrated. However, due to the potentially

serious nature of this issue, elucidating the true nature and source of potential epigenetic disturbance in ART should be an important research goal. We begin with a general description of basic epigenetic concepts. This is followed by a description of experimental and clinical evidence of a relationship between ART techniques and epigenetic perturbations. The review concludes with a discussion of the potential consequences of such perturbations in the application of ART.

EPIGENETICS: BASIC CONCEPTS

Differential gene expression is a basic and ubiquitous component of development and cellular function. Epigenetics describes processes through which gene expression is controlled or altered in a specific, usually stable, and sometimes reactionary fashion. Epigenetics concerns a specific type of expression control, in which genes or chromosomal regions are targeted for silencing or activation based on unique developmental programming that is ‘outside’ baseline Mendelian inheritance. The phenomenon of imprinting is a particular aspect of epigenetics, in which certain mammalian genes are expressed based on parent-of-origin – this has been raised as a primary concern in relation to ART.2,3 However, the breadth of epigenetic phenomena is still far from understood at present, and it is therefore prudent to think about possible epigenetic effects on gene expression in the broadest terms. The term ‘epigenetics’ was first coined to simply describe the generation of phenotype from genotype, and its definition has since changed over the years.4,5 Epigenetic processes are of course entirely normal, and absolutely integral and required components of development; therefore the current use of 

HPE_Chapter19.qxp

7/14/2007

5:51 PM

Page 236


the term to describe gene expression phenomena that seem to be ‘outside’ normal inheritance is a somewhat confusing distinction. Epigenetic phenomena can be thought of as an evolutionary ‘fine-tuning’ of gene expression. In terms of mechanism, epigenetics concerns the silencing/ activation of genes and chromosomal regions, and is related to ancient and ubiquitous processes of chromatin silencing and genetic control. As life evolved, selection pressure resulted in the requirement for silencing or activation of specific genetic information above and beyond baseline developmental and housekeeping control. The form of expression control we now term epigenetic is particularly involved in both the silencing/activation of genes based on external (i.e. environmental) stimuli, and in the control of gene dosage.2,6 The delineation of genetic phenomena that are classified as epigenetic is somewhat variable: the term is also used to describe processes such as X-inactivation, the silencing of foreign (i.e. viral DNA), and aberrant gene activation/ silencing in cancer.2,7,8 The best description of what constitutes an epigenetic control mechanism, beginning with environmental effects, can be provided by specific examples. In flowering plants, the time of flowering is a critical developmental decision which must be correlated with environmental cues. Cold-response vernalization affects this process by promoting flowering following a prolonged period of cold temperature. It is now recognized that vernalization involves epigenetic changes to the expression of certain genes that control flowering.9 Expression of the FLOWERING LOCUS C (FLC) gene in Arabidopsis is proportionally repressed in adult plants in response to cold treatment perceived by the germinating seed. This repression is stable, but high FLC expression is re-established in progeny plants. The exact mechanism of this repression remains to be elucidated, but it involves specific chromatin modification mediated by gene products like VERNALIZATION2, a protein related to the polycomb group proteins in Drosophila known to be involved with gene silencing.10 In this example of epigenetic control, the perception of an environmental cue during early development is transformed into chromatin modifications that specifically 

alter downstream gene expression in a stable fashion and bring about an appropriate response in the mature organism. Furthermore, these modifications are ‘reset’ in the next generation to allow for a new cycle of perception and response. These are hallmark components of an epigenetic process. In mammals, the best understood example of epigenetic programming involves the allele-specific expression of certain genes depending on the parent of origin.5 In the 1980s, nuclear transfer and uniparental disomy experiments in mice definitively demonstrated for the first time the absolute requirement for both a maternally and a paternally derived genome in directing normal mammalian development.11–13 Androgenetic or gynogenetic embryos (derived from diploid paternal or maternal genomes) exhibited aberrant and incomplete development, indicating that the two parental genomes were not equivalent.11,12 Mice that inherit uniparental disomies (UPD) (i.e. chromosome complements where a component is entirely derived from one parent) also exhibit a variety of developmental problems which are usually lethal.13 It is now understood that certain genes are only expressed when inherited from the maternal or paternal germline. These genes are described as being ‘imprinted’ by the germline of origin, with a stable and defining status that determines expression during development. For instance, androgenetic embryos (or those with an androgenetic UPD) lack expression of genes that are only active when derived from the maternal germline (and exhibit overexpression of paternally derived genes) and are thus developmentally incompetent. Current theory proposes that the origin of this epigenetic control lies in the conflict between paternal and maternal evolutionary interests during genome transmission.14 In mammals, this process necessarily involves the maternal contribution to the developing offspring during gestation. Paternal interests, apparently accelerated by polyandry (multiple male partners for each female), seek to maximize this maternal contribution to the development of each offspring. Paternal imprinting seems to promote the expression of genes with activities related to maximizing fetal growth and development.15 The differential maternal imprinting seeks to control fetal growth to protect

HPE_Chapter19.qxp

7/14/2007

5:51 PM

Page 237

GENOMIC IMPRINTING AND CONSEQUENCES FOR EMBRYONIC DEVELOPMENT

the mother’s own interests and future gestational activity. For example, the maternally derived copy of a gene product involved with the promotion of fetal growth, insulin-like growth factor 2 (Igf2), is imprinted to an inactive state, effectively reducing the dosage of this gene.16 Conversely, the gene for the Igf2 receptor (which binds to and inactivates Igf2) is inactivated on the paternal chromosome. This is a perfect example of a genomic ‘tug-of-war’ between the promotion and suppression of fetal growth. The result is that a complement of imprinted mammalian genes is only active in a single copy – the one bearing the active parent-of-origin imprint. As is discussed, dosage control for these genes is critical to normal development and perturbation to the process can have critical consequences.

EPIGENETICS: MECHANISTIC CONCERNS

A thorough description of epigenetic mechanisms from a molecular standpoint is well beyond the scope of this review. Epigenetic control mechanisms are among the most complex in molecular genetics, and are still only partly understood.1,6,17 For the most part, epigenetic silencing or activation involves stable covalent modification to the chromatin, creating specific states that have an over reaching effect on other mechanisms of expression control. Two hallmark epigenetic chromatin modifications are cytosine methylation (at CpG sequences) and methylation/ acetylation of histone proteins.18,19 Generally, the genome is divided into heavily methylated regions that are silent, and hypomethylated regions that are actively transcribed. However, as discussed below, in some cases uniquely methylated control regions are associated with gene activation. Cytosine methylation seems to be the principal protection mechanism for silencing foreign DNA (viral sequences, retrotransposons, etc.) in the genome, and such sequences are heavily methylated. In epigenetic regulation, methylation and other modifications come under allelespecific control schemes that ensure the maintenance of stable expression patterns. As the genome transits through the developing germline and in the developing embryo, waves of methylation, demethylation,

and other types of chromatin modification occur. These global chromatin modification processes are modified at imprinted loci to bring about specific active and silenced chromatin states. Once again, specific examples provide the best explanation. Returning to the Igf2 locus in mice, this gene is located on chromosome 7 in a region that is subject to imprinting effects.17,20 On chromosomes derived from a paternal germline, Igf2 is expressed; the gene is silenced on maternally derived chromosomes. A directly adjacent locus, H19, exhibits the reverse epigenetic expression pattern, with only maternal chromosome expression.21 These two loci and their complex interactions with germline-specific transregulatory proteins form an epigenetic molecular switch that controls Igf2 expression. Both genes require and compete for expression promoting enhancer proteins.22 A region adjacent to H19 exhibits differential methylation on the maternal and paternal chromosomes.21 This imprinting control region (ICR) acts as a binding site for regulatory proteins, such as the CCCTCbinding factor protein (CTCF), that create a boundary or insulator function between the two loci that blocks enhancer activity.23 On the maternal chromosome, the ICR is unmethylated, allowing CTCF binding and downstream silencing effects that prevent Igf2 expression.24 However, on the paternal chromosome, the ICR is methylated, CTCF binding is prevented, and enhancer mediated Igf2 expression results. Thus methylation-sensitive differential protein–chromatin interaction acts as a tight bipolar switch, initiating the creation of unique maternal and paternal chromatin conformation loops with the Igf2 locus in an active or silent state. CTCF ‘reads’ the imprinting marks and facilitates differential expression.25,26 CTCF is known to bind to all currently known imprinting-related control regions, including those involved with X-inactivation.27 Interestingly, a homolog of CTCF has recently been discovered in Drosophila.28 The Drosophila genome is highly compact, and basic expression control involves an almost ubiquitous use of ‘insulator’ and blocking elements similar to mammalian ICRs between genes. Apparently CTCF was one of many such proteins in Drosophila that has been co-opted to a similar but specific function in mammals. 

HPE_Chapter19.qxp

7/14/2007

5:51 PM

Page 238


An obvious question remains concerning the mechanism whereby the germline-unique imprints are created in the first place. Germline specific mechanisms must exist that allow existing imprints (from parental chromosomes) to be erased, and new imprints in the transmitted germ cell chromosomes to be established. This process is still only partly understood on a molecular level. A male germlinespecific protein, BORIS, has recently been identified that exhibits exquisitely identical DNA binding regions to CTCF, but a divergent functional sequence.29 The two proteins exhibit a mutually exclusive expression pattern during male germ cell development. In fact, BORIS is uniquely expressed in primary spermatocytes – the only cell type known where CTCF expression is absent. BORIS expression occurs at the time when existing parental methylation imprints are erased, and its down regulation coincides with the wave of methylation occurring at the round spermatid stage, when new paternal imprints may be established. BORIS may be involved with binding to ICR sequences and targeting them for methylation – in this case only in the male germline. CTCF binding has been shown to protect its target sequence from downstream methylation, so a reciprocal expression pattern with BORIS makes sense in this regard. A model can be envisioned in which the parentspecific germline expression pattern of ICR binding proteins such as CTCF and BORIS control methylation and other modifications at critical ICR sites during genome transmission. Therefore, as the maternal and paternal genomes transit through the early germline, gametogenesis, fertilization, and early development, an increasing number of global processes involving methylation and chromatin modification are modulated to produce the unique parentally imprinted loci observed. This model is most certainly an over simplification, and is clearly only one mechanism among several that describe the generation and maintenance of imprints and downstream expression control. In general, maternal germline-specific methylation seems to be a more common characteristic of imprinted genes.18,30 A wave of paternal genome demethylation occurs in the oocyte following fertilization, and this is speculated to be a component of the evolutionary 

conflict for control of imprinted loci.31 Other expression control mechanisms include direct repression by methylated DNA binding proteins, and the interaction of anti-sense non-coding transcripts in down regulating imprinted targets.1,17,19 In fact, the H19 locus codes for one such transcript, adding another level of complexity to that story. It is worth noting that in the absence of methylation (in methyltransferase knockout experiments) some imprinted genes are active from both alleles and some are silenced at both alleles.30 In general, epigenetic expression control should be thought of as arising from a complex synergy of target sequence-specific control regions and transcriptional effects, along with germline/ developmental specific protein expression patterns. Furthermore, the majority of normal epigenetic processing seems to take place during gamete development and early embryonic development. As discussed in the next section, this aspect clearly has implications for potential perturbation by ART procedures.

ART AND EPIGENETICS: EVIDENCE FOR PROBLEMS

The evidence for perturbations to epigenetic processing associated with assisted reproduction procedures comes from two main areas: research and large animal experiments related to assisted reproduction/ cloning, and the incidence of imprinting related disorders following human ART application. EVIDENCE FROM ANIMAL EXPERIMENTS

Assisted reproduction protocols in large animals such as the sheep and cow developed concurrently with, or followed the first successful use of assisted reproduction in the human. However, unlike in the human, such large animal protocols have been plagued by a variety of distinct developmental problems. It has become clear that some of these problems are directly related to abnormal imprinting and epigenetic processing associated with in vitro culture and gamete/embryo manipulation. Particularly during nuclear transfer cloning protocols (but also

HPE_Chapter19.qxp

7/14/2007

5:51 PM

Page 239


observed following basic in vitro embryo culture), aberrations of gestation length and in utero fetal growth control have been so frequent that a specific terminology, ‘large offspring syndrome’ (LOS), was coined to describe the situation in ruminants.32 In this syndrome, animals produced via ART are born after extended gestation periods with greatly augmented body mass and a variety of physical defects. It is now known that such animals manifest abnormalities in the epigenetic status and expression of imprinted genes. This particular relationship has not been observed in human ART, although reduced birth weight has been correlated with ART in singletons. In perhaps the best study using sheep, in vivo fertilized eggs were simply cultured in vitro for 5 days and then transferred to establish pregnancies.33 One-quarter of fetuses recovered just prior to term exhibited body weights 22–82% increased above the largest non-cultured control fetuses. In the LOS fetuses, IGF2 expression was normal. However, expression of the IGF2R gene was decreased 30–60% relative to controls, and this reduction was not observed in normal weight offspring also derived from in vitro culture. Reduced IGF2R expression was correlated with an aberrant loss of methylation at a key imprinting control region for the gene.33 A similar scenario has been observed in other large animal ART and cloning applications.32 Based on the large animal results, studies in the mouse have pursued a clear and specific connection between in vitro culture/manipulations and epigenetic abnormalities. The gist of these experiments is that transient preimplantation in vitro culture does seem to result in changes in gene expression in mice.34,35 Both imprinted and apparently nonimprinted genes can be affected. Furthermore, the type of culture media and serum supplementation used can affect this process. For example, one study showed that imprinting of the murine H19 gene was lost, with abnormal hypomethylation of the paternal allele following in vitro culture in a simple media (Whitten’s media). However, this abnormality was absent when culture took place in a complex amino acid-augmented media (KSOM).34 It is interesting to note that the abnormal bi-allelic H19 expression pattern observed in the Whitten’s media

embryos was lost following implantation, but remained in the extraembryonic tissues. Recently, mice derived from in vitro culture were blindly evaluated and compared with non-cultured littermates, delineated after analysis via a transgene maker.36 After exhaustive evaluation examining numerous developmental and behavioral markers, in vitro and in vivo derived mice were found to be essentially identical. No significant differences were found, after assessment of more than a dozen individual developmental milestones and complex behavioral/learning patterns. However, a slight but significant difference was observed in spatial learning retention, and in one component of an anxiety assessment maze, particularly in male animals.36 The authors of this study have recently reported that there was no distinguishable difference in lifespan between in vitro and in vivo derived animals.37 The actual origin of imprinting disturbance following in vitro manipulations in mice and ruminants is still undefined. It can be speculated that either basic epigenetic mechanisms are simply disturbed by ‘suboptimal’ culture conditions, or that normally functioning epigenetic mechanisms are responding to the altered environment of in vitro culture, resulting in gene expression changes. Large-scale cloning experiments in the mouse seem to indicate that gene expression abnormalities are exacerbated by the nuclear transfer procedure.38 Considering the importance of stage- and germlinespecific events in correct epigenetic processing – certainly challenged by the introduction of a somatic cell nucleus into an oocyte – this is not surprising. In fact, it is surprising that such protocols can produce developmentally functional embryos and offspring at all. As mentioned above, several recent studies have indicated widespread dysfunction of gene expression in cloned embryos, and many cloned animals exhibit developmental, late onset problems, and in some cases reduced lifespan.39 However, in other cases, nuclear transfer offspring, even those known to harbor clear gene expression abnormalities, have exhibited normal development and unremarkable life histories.38 Other experiments in the mouse have examined this issue from a different perspective by attempting to circumvent parental-specific imprinting controls. 

HPE_Chapter19.qxp

7/14/2007

5:51 PM

Page 240


Gynogenetic embryos, derived from the genome of early ‘non-growing’ stage oocytes (lacking later-stage epigenetic modifications), exhibit extended development well beyond that previously observed following parthenogenesis.40 These experiments have recently come to fruition, and via the direct manipulation of the expression pattern of only the Igf2 and H19 genes, resulted in the birth of apparently normal parthenogenetic offspring with a gynogenetic, i.e. diploid, maternal genome source.41 Despite their genetic makeup with supposedly only maternally imprinted chromosomes, the offspring from these experiments manifested apparently normalized epigenetic expression patterns at other imprinted genes. The expression of Igf2 and H19 was artificially manipulated via genetic modification to these loci. However, a second set of non-manipulated paternally imprinted genes, Dlk1 and Dtl2, exhibited normal methylation patterns and expression in surviving term offspring.41 Future experiments will hopefully provide further insight into how this epigenetic ‘normalization’ occurs. This has important relevance to ART and other protocols that involve early embryo development. Taken together, the data from large animal and mouse experiments suggests that in vitro embryo culture and perturbations can definitely result in epigenetic disturbance and gene expression alterations in these species. Furthermore, such epigenetic expression issues can result in both subtle and profound phenotypic consequences. However, considerable variation and flexibility seems to exist in both the origin and the downstream manifestations of such disturbance. It is also worth noting at this point that differences are known to exist between rodents, ruminants, and primates, including very specific epigenetic processing issues. For example, the IGF2R gene, which is known to exhibit epigenetic misregulation in the ruminant LOS is not subject to epigenetic control in primates. Despite the presence of an almost identical imprinting control region between the human and ruminant genes, IGF2R is not imprinted in the human – this regulation was apparently lost during the evolution of the primate lineage.42 Other genes continue to be identified that display differential epigenetic behavior between mice 

and humans.43 It is important to point out that the experimental findings are stage- as well speciesspecific, and that environmental factors in vitro may play a role. Embryologists experimenting with animal embryos tend to be less concerned about environmental criteria than human embryology specialists, and such differences may account for some of the observations. EVIDENCE FROM HUMAN POPULATION STUDIES

In 2002, two ICSI-conceived children were reported to be affected by an imprinting-related disorder;44 this was followed by a flood of reports that showed a slight but significant increase in the incidence of human imprinting-related disorders in children born following the application of ART.45–48 BeckwithWiedemann (BW) and Angelman (AS) syndromes are serious human developmental disorders caused by loss, uniparental disomies, or perturbations in the epigenetic expression of imprinted genes. Recent reports examining databases of affected individuals indicate that there is a significant increase in the incidence of both BW and AS when ART techniques were used.45 In most cases, molecular genetic analysis of affected individuals conceived by ART has confirmed an epigenetic origin for their syndrome phenotypes.49,50 Another study showed an increased ART-related incidence of retinoblastoma (another disorder with a known epigenetic component).51 However no molecular data linking the disorder with epigenetic perturbations were presented. In these studies, the overall incidence of these imprinting disorders was still very small. However, the incidence in ART offspring compared with baseline population levels was in every case significantly increased. The apparent incidence (based on a comparison using the overall numbers of ART births) seems to be approximately one per 10 000 births. However, one study of BW in Australia calculated an expected risk of BW following ART as approximately one per 4000.52 Of course, it is not possible for such comparisons to demonstrate a direct connection between ART techniques themselves and epigenetic disturbance, since entirely different populations are being compared.

HPE_Chapter19.qxp

7/14/2007

5:51 PM

Page 241


ART offspring are conceived from a population of individuals manifesting a variety of serious medical problems associated with their infertility. In some cases, the proximal cause of such infertility is cryptic and as yet poorly understood. Obviously, a predisposition for epigenetic aberrations could be an underlying cause of infertility in some couples, and other individuals may exhibit specific ‘sensitivity’ to the potential for perturbations caused by ovarian stimulation, in vitro culture, or other components of ART protocols. The increased incidence of apparent imprinting disturbance has been observed following a variety of ART interventions, including simple stimulation/insemination protocols without any in vitro culture or manipulation.53 In the majority of the studies published so far, detailed parental histories are lacking. However, in at least one case, there was a clear family history of unexplained infertility over multiple generations on the maternal side, potentially indicating a genetic predisposition for developmental (perhaps imprinting related) problems.54 It is of course also entirely possible that a certain level of epigenetic disturbance does directly result from ART protocols. Obviously, this must occur (or at least be manifested phenotypically) at a greatly reduced level compared with what has been observed in large animal and research experiments, and as mentioned, significant differences in imprinting are evident between the human and these animal model systems, and perhaps others are yet to be defined. As mentioned earlier, it is still unclear from animal experiments what the origin of in vitro culture-associated epigenetic disturbance might be. Also, the true correspondence between animal model systems and the human in terms of both in vitro manipulations and basic epigenetic processes also remains an open question.

CONCLUSIONS, CONSEQUENCES, AND FUTURE PROSPECTS

Taken as a whole, current knowledge about the molecular and developmental biology of epigenetic phenomena, the aberrations (as well as flexibility) observed in model systems of in vitro manipulation,

and the increased incidence of serious imprintingrelated disorders in human ART-associated offspring suggests that there is some cause for concern about a relationship between human ART and abnormal epigenetic regulation. Without question this issue deserves serious further attention in terms of basic research, population studies, follow-up studies on ART offspring, and in the assessment, counseling, and treatment of ART patients. As alluded to earlier, epigenetic gene expression phenomena can theoretically manifest themselves throughout an individual’s life history, including association with late onset diseases such as cancer. Furthermore, such phenomena can also be manifest across multiple generations. The first human ART offspring are now reaching adulthood and having children of their own, and therefore data collection on potential epigenetic abnormalities following ART is still very much ‘in progress’. However, drawing premature conclusions based on the data available to date is simply bad science, with the potential to manifest itself in compromised treatment of ART patients. For example, the suggestion that ICSI poses a danger and is specifically associated with imprinting aberration is based on very little evidence, and none of it is remotely conclusive as to a causative relationship. The fact that imprinting disorders have been observed following a range of ART treatments argues against a specific relationship.53 One direct study of the chromosomal region associated with Angelman’s syndrome showed no evidence of epigenetic misregulation in 92 ICSIconceived offspring.55 A study of this magnitude must be considered preliminary, since the disease frequency is observed 50–100 times more rarely. Although further population and continued follow-up studies are most definitely needed to provide a true assessment of the incidence of imprintingrelated consequences in ART offspring, this is only part of the picture. As discussed above, simple population incidence studies cannot determine whether a direct connection exists between ART procedures and abnormal imprinting, or whether ART patients represent a group with increased incidence or sensitivity to epigenetic abnormalities. One future challenge will be to address this issue in terms of both potential realities. In cases of completely unexplained 

HPE_Chapter19.qxp

7/14/2007

5:51 PM

Page 242


infertility in particular, it will be prudent to develop procedures for screening patients for potential imprinting issues and perhaps embryos via PGD. In addition, regardless of the current ambiguity, ART patients should be informed and counseled regarding the potential for an increased incidence of imprinting-related disorders. While the general success of human ART would suggest that more general epigenetic problems are not widespread, it is possible that subtle alterations in epigenetic processing related to ART may have some effect on development. For example, it is possible that such subtle alterations could be involved in the slightly decreased birth weight observed in ART singleton offspring in some studies.56 Our basic knowledge about imprinting status and more general epigenetic status during human gamete and embryo development needs to be greatly increased, particularly in relation to ART issues and protocols. Information about the timing, establishment, and maintenance of epigenetic modification needs to be gained in the relevant human material. Clearly, model systems do not provide an accurate or complete picture in this regard. For example, a recent study revealed significant differences between the human and the mouse in expression patterns of genes related to epigenetic processing such as methyltransferase and methylated DNA binding proteins.57 The imprinting-related methyltransferase DNMT3L was highly expressed in developing mouse oocytes, but did not appear in human material until after fertilization in developing embryos, indicating that the timing of imprint establishment may differ between the two species. Due to a variety of technical and ethical concerns, little direct analysis of epigenetic status has been conducted in human gametes and embryos. However, the studies that have been published indicate that, at least in the case of some genes that are known to be imprinted, correct epigenetic status appears to be undisturbed. An examination of the methylation-imprint status of the imprinted SNRPN gene (associated with Angelman syndrome) in human spermatozoa, oocytes, and preimplantation embryos revealed the expected pattern in all material analyzed.58 Furthermore, these results indicated that 

the correct maternal imprint was already stably established in human oocytes by the germinal vesicle stage, and not during later stages or following fertilization as had been previously thought.59 Another meticulous study examined the status of the imprinted H19 gene (associated with Beckwith-Wiedemann syndrome) in human male germ cells obtained by laser dissection of samples from the seminiferous tubules of individuals who exhibited impaired spermatogenesis, a situation of particular importance to the treatment of severe male factor patients. No evidence of abnormal imprinting at H19 was evident, even in cells from tubules that exhibited arrest at the spermatogonial stage.60 Studies such as these need to be extended in both number and scope. As indicated above, the timing and nature of normal epigenetic modifications in the human is a basic issue. Some global epigenetic events, such as paternal genome demethylation in the zygote, are known to occur during the in vitro culture period of ART protocols, and this process has been studied in a preliminary fashion in human zygotes using methylated DNA-specific antibodies.61 However, other epigenetic processing occurs prior to or following the period of ART manipulation, and a full and accurate picture of such processing in the human is currently lacking. In addition, based on the results from animal culture studies (with the caveat of potential cross-species differences), brief periods of transient postfertilization in vitro culture can theoretically cause perturbations to prior or subsequent epigenetic modifications. A variety of molecular methods are available for determining the imprint status at particular loci, based on an assessment of site-specific DNA methylation. Microarray techniques have recently been adapted to such protocols, allowing multiple targets to be assessed simultaneously.62 If such microarray protocols could be modified to allow for the greatly reduced target concentration that is available from single oocytes or early embryos, this would seem to provide an ideal methodology for pursuing epigenetic analysis in human ART-related material. A more daunting question concerns assessing the possibility of more general epigenetic perturbations, perhaps to cryptically imprinted or non-imprinted genes.63

HPE_Chapter19.qxp

7/14/2007

5:51 PM

Page 243


Single-cell expression analysis has been accomplished in human gametes and embryos, and data from such studies on both general variation as well as the specific expression of epigenetic-related gene products will need to be obtained and examined with these issues in mind.64 This type of data will be crucial in determining the real status of epigenetic disturbance during human ART protocols. It could also assist in the development of alternative protocols (culture media/conditions, etc.) that, if necessary, might be used to avoid epigenetic perturbation. Examining the situation in preimplantation embryos will provide only a partial analysis of potential epigenetic problems in ART, and it should perhaps be admitted that, due to the complexity of this issue, a truly complete assessment is simply impossible. As mentioned in the introduction, the basic worldwide status of human ART results to date, both anecdotally and in terms of formal follow-up studies, does not indicate the presence of any widespread or significant negative effect that could result from epigenetic aberration. Nevertheless, ART scientists and practitioners owe their patients an ongoing effort to ensure that the methodology in ART protocols ‘does no harm’. Understanding the true nature and ramifications of potential epigenetic problems related to ART protocols or in the ART patient population should now be seen as a critical component of this ongoing effort. REFERENCES 1. Jaenisch R, Bird A. Epigenetic regulation of gene expression: how the genome integrates intrinsic and environmental signals. Nat Genet 2003; 33: 245–53. 2. Surani MA. Imprinting and the initiation of gene silencing in the germ line. Cell 1998; 93: 309–12. 3. Maher ER. Imprinting and assisted reproductive technology. Hum Mol Genet 2005; 14: R133–8. 4. Lederberg J. The meaning of epigenetics. The Scientist 2001; 15: 6. 5. Waddington C. The genetic control of wing development in Drosophila. J Genet 1940; 41: 75–80. 6. Latham KE. Epigenetic modification and imprinting of the mammalian genome during development. Curr Top Devel Biol 1999; 43: 1–49. 7. Lee JT, Jaenisch R. The (epi)genetic control of mammalian Xchromosome inactivation. Curr Opin Genet Dev 1997; 7: 274–80. 8. Feinberg AP, Tycko B. The history of cancer epigenetics. Nat Rev Cancer 2004; 4: 143–53. 9. Finnegan EJ, Sheldon CC, Jardinaud F et al. A cluster of Arabidopsis genes with a coordinate response to an environmental stimulus. Curr Biol 2004; 14: 911–16.

10. Finnegan EJ, Kovac KA, Jaligot E et al. The downregulation of FLOWERING LOCUS C (FLC) expression in plants with low levels of DNA methylation and by vernalization occurs by distinct mechanisms. Plant J 2005; 44: 420–32. 11. McGrath J, Solter D. Completion of mouse embryogenesis requires both maternal and paternal genomes. Cell 1984; 37: 179–83. 12. Surani MAH, Barton SC, Norris ML. Development of reconstituted mouse eggs suggests imprinting of the genome during gametogenesis. Nature 1984; 308: 548–50. 13. Cattanach BM, Kirk M. Differential activity of maternally and paternally derived chromosome regions in mice. Nature 1985; 315: 496–8. 14. Moore T. Genetic conflict, genomic imprinting and establishment of the epigenotype in relation to growth. Reproduction 2001; 122: 185–93. 15. Moore T, Haig D. Genomic imprinting in mammalian development: a parental tug-of-war. Trends Genet 1991; 7: 45–9. 16. Haig D, Graham C. Genomic imprinting and the strange case of the insulin-like growth factor II receptor. Cell 1991; 64: 1045–6. 17. Pfeifer K. Mechanisms of genomic imprinting. Am J Hum Genet 2000; 67: 777–87. 18. Bestor T. Cytosine methylation and the unequal developmental potentials of the oocyte and sperm genomes. Am J Hum Genet 1998; 62: 1269–73. 19. Schübeler D, Elgin SCR. Defining epigenetic states through chromatin and RNA. Nat Genet 2005; 37: 917–18. 20. DeChiara TM, Robertson EJ, Efstratiadis A. Parental imprinting of the mouse insulin-like growth factor II gene. Cell 1991; 64: 849–59. 21. Tremblay KD, Saam JR, Ingram RS et al. A paternal-specific methylation imprint marks the alleles of the mouse H19 gene. Nat Genet 1995; 9: 407–13. 22. Leighton PA, Saam JR, Ingram RS et al. An enhancer deletion affects both H19 and Igf2 expression. Genes Dev 1995; 9: 2079–89. 23. Bell AC, West AG, Felsenfeld G. Insulators and boundaries: versatile regulatory elements in the eukaryotic genome. Science 2001; 291: 447–50. 24. Holmgren C, Kanduri K, Dell G et al. CpG methylation regulates the Igf2/H19 insulator. Curr Biol 2001; 11: 1128–30. 25. Kanduri C, Pant V, Loukinov D et al. Functional association of CTCF with the insulator upstream of the H19 gene is parent-of-origin specific and methylation-sensitive. Curr Biol 2000; 10: 853–6. 26. Murrell A, Heeson S, Reik W. Interaction between differentially methylated regions partitions the imprinted genes Igf2 and H19 into parent-specific chromatin loops. Nat Genet 2004; 36: 889–93. 27. Lee JT. Molecular links between X-inactivation and autosomal imprinting: X-inactivation as a driving force for the evolution of imprinting? Curr Biol 2003; 13: R242–54. 28. Moon H, Filippova G, Loukinov D et al. CTCF is conserved from Drosophila to humans and confers enhancer blocking of the Fab-8 insulator. EMBO Rep 2005; 6: 165–70. 29. Loukinov D, Pugacheva E, Vatolin S et al. BORIS, a novel male germline-specific protein associated with epigenetic reprogramming events, shares the same 11-zinc-finger domain with CTCF, the insulator protein involved in reading imprinting marks in the soma. Proc Natl Acad Sci USA 2002; 99: 6806–11. 30. Li E, Beard C, Forster AC et al. DNA methylation, genomic imprinting, and mammalian development. Cold Spring Harb Symp Quant Biol 1993; 58: 297–305. 31. Oswald J, Engeman S, Lane N et al. Active demethylation of the paternal genome in the mouse zygote. Curr Biol 2000; 10: 475–8. 32. Young LE, Wilmut I. Large offspring syndrome in cattle and sheep. Rev Reprod 1998; 3: 155–63.



HPE_Chapter19.qxp

7/14/2007

5:51 PM

Page 244


33. Young LE, Fernandes K, McEvoy TG et al. Epigenetic change in IGF2R is associated with fetal overgrowth after sheep embryo culture. Nat Genet 2001; 27: 153–4. 34. Erbach GT, Lawitts JA, Papaioannou VE et al. Differential growth of mouse preimplantation embryos in chemically defined media. Biol Reprod 1994; 50: 1027–33. 35. Doherty TL, Mann MR, Tremblay KD et al. Differential effects of culture on imprinted H19 expression in the preimplantation mouse embryo. Biol Reprod 2000; 62: 1526–35. 36. Ecker DJ, Stein P, Xu Z et al. Long-term effects of culture of preimplantation mouse embryos on behavior. Proc Natl Acad Sci USA 2004; 101: 1595–600. 37. Sommovilla J, Bilker WB, Abel T et al. Embryo culture does not affect the longevity of offspring in mice. Reproduction 2005; 130: 599–601. 38. Humpherys D, Eggan K, Akutsu H et al. Epigenetic instability in ES cells and cloned mice. Science 2001; 293: 95–7. 39. Rideout WM 3rd, Eggan K, Jaenisch R. Nuclear cloning and epigenetic reprogramming of the genome. Science 2001; 293: 1093–8. 40. Kono T, Obata Y, Yoshimzu T et al. Epigenetic modifications during oocyte growth correlates with extended parthenogenetic development in the mouse. Nat Genet 1996; 13: 91–4. 41. Wu Q, Kumagai T, Kwahara W et al. Regulated expression of two sets of paternally imprinted genes is necessary for mouse parthenogenetic development to term. Reproduction 2006; 131: 481–8. 42. Killian JK, Nolan CM, Wylie AA et al. Divergent evolution in M6P/IGF2R imprinting from the Jurassic to the Quaternary. Hum Mol Genet 2001; 10: 1721–8. 43. Arnaud P, Monk D, Hitchins M et al. Conserved methylation imprints in the human and mouse GRB10 genes with divergent allelic expression suggests differential reading of the same mark. Hum Mol Genet 2003; 12: 1005–19. 44. Cox GF, Burger J, Lip V et al. Intracytoplasmic sperm injection may increase the risk of imprinting defects. Am J Hum Genet 2002; 71: 162–4. 45. Neimitz EL, Feinberg AP. Epigenetics and assisted reproductive technology: a call for investigation. Am J Hum Genet 2004; 74: 599–609. 46. Jacob S, Moley KH. Gametes and embryo epigenetic reprogramming affect developmental outcome: implication for assisted reproductive technologies. Ped Res 2005; 58: 437–6. 47. Maher ER. Imprinting and assisted reproductive technology. Hum Mol Genet 2005; 14: R133–8. 48. Sutcliffe AG, Peters CJ, Bowdin S et al. Assisted reproductive technologies and imprinting disorders – a preliminary British survey. Hum Reprod 2006; 21: 1009–11.



49. DeBraun MR, Neimitz El, Feinberg AP. Association of in vitro fertilization with Beckman-Weidemann syndrome and epigenetic alterations of LIT1 and H19. Am J Hum Genet 2003; 72: 156–60. 50. Gicquel C, Gaston V, Mandelbaum J et al. In vitro fertilization may increase the risk of Beckwith-Wiedemann syndrome related to the abnormal imprinting of the KCN10T gene. Am J Hum Genet 2003; 72: 1338–41. 51. Moll AC, Imhof SM, Cruysberg JR et al. Incidence of retinoblastoma in children born after in-vitro fertilisation. Lancet 2003; 361: 309–10. 52. Halliday J, Oke K, Breheny S et al. Beckwith-Wiedemann syndrome and IVF: a case-controlled study. Am J Hum Genet 2004; 75: 526–8. 53. Chang AS, Moley KH, Wangler M et al. The association between Beckwith-Wiedemann syndrome and assisted reproductive technology: a series of 19 patients. Fertil Steril 2005; 83: 349–54. 54. Örstavik KH, Eiklid K, van der Hagen CB et al. Another case of imprinting defect in a girl with Angelman syndrome who was conceived by intracytoplasmic sperm injection. Am J Hum Genet 2003; 72: 218–9. 55. Manning M, Lissens W, Bonduelle M et al. Study of DNA-methylation patterns at chromosome 15q11–q13 in children born after ICSI reveals no imprinting defects. Mol Hum Reprod 2000; 6: 1049–53. 56. Schieve LA, Meikle SF, Ferre C et al. Low and very low birth weight in infants conceived with the use of assisted reproductive technology. N Engl J Med 2002; 346: 731–7. 57. Huntriss J, Hinkins M, Oliver B et al. Expression of mRNAs for DNA methyltransferases and methyl-CpG-binding proteins in the human female germline preimplantation embryos and embryonic stem cells. Mol Reprod Dev 2004; 67: 323–6. 58. Geuns E, De Ryke M, Van Steirteghem et al. Methylation imprints of the imprint control region of the SNRPN-gene in human gametes and preimplantation embryos. Hum Mol Genet 2003; 12: 2873–9. 59. El-Marrii O, Buiting K, Peery EG et al. Maternal methylation imprints on human chromosome 15 are established during or after fertilization. Nat Genet 2001; 27: 341–4. 60. Hartmann S, Bermann M, Bohle RM et al. Genetic imprinting during impaired spermatogenesis. Mol Hum Reprod 2006; 12: 407–11. 61. Fulka H, Mrazek M, Tepla O et al. DNA methylation pattern in human zygotes and developing embryos. Reproduction 2004; 128: 703–8. 62. Schumacher A, Kapranov P, Kaminsky Z et al. Micrroarray-based DNA methylation profiling: technology and applications. Nucleic Acids Res 2006; 34: 528–42. 63. Nikaido I, Saito C, Mizuno Y et al. Discovery of imprinted transcripts in the mouse transcriptosome using large-scale expression profiling. Genome Res 2003; 13: 1402–9. 64. Bermudez MG, Wells D, Malter H et al. Expression profiles of individual human oocytes using microarray technology. Reprod Biomed Online 2004; 8: 325–7.

HPE_Chapter20.qxp

7/17/2007

3:38 PM

Page 245

20. Selection of viable embryos and gametes by rapid, non-invasive metabolomic profiling of oxidative stress biomarkers James T Posillico and The Metabolomics Study Group for Assisted Reproductive Technologies

BACKGROUND FRAMING THE PROBLEM

In vitro fertilization (IVF) is often criticized for its generally low success rates and high treatment costs. The procedure is also confounded by high multiple birth rates that contribute to preterm delivery, small for gestational age babies, and increased neonatal mortality with its associated healthcare costs. These problems could be significantly ameliorated if single embryos of known viability and high developmental potential could be identified within a cohort group and selected for transfer. Moreover, the selection of such embryos by non-invasive techniques is necessary. Early embryonic wastage, which is usually caused by abnormalities in the number of chromosomes (aneuploidy), normally occurs in nature and is endemic in the IVF treatment procedure. Mounting scientific evidence suggests that as many as 60–75% of all oocytes generated in IVF are abnormal. Likewise, in nature the number of abnormal eggs increases dramatically with age and this is the primary cause of reduced fecundity in older women. Abnormalities in the genetic makeup of the egg are usually manifested as embryo wastage and/or congenital defects in the newborn. Since embryo aneuploidy is traceable back to oocyte aneuploidy, it follows that oocyte competence plays a central role in reproductive success. However, although a large number of strategies have been employed, there are no well-defined biological measurements or analytical

procedures available today that allow embryologists to absolutely distinguish between viable and nonviable oocytes and embryos. A major limitation in IVF is the inability to predict embryo viability. The ‘holy grail’ of IVF is embryo selection; more specifically, how to determine which embryos in a cohort group are viable, competent embryos that will implant and produce a pregnancy versus those that will not. Therefore, it follows that IVF treatments that (unknowingly) use non-viable, defective embryos will likely result in poor pregnancy outcomes and increased medical risk. Although there are a number of non-invasive methods to assess embryo quality, microscopic examination of embryo morphology remains the primary parameter of embryo selection.1 Genetic testing of the embryo by preimplantation genetic diagnosis is an invasive, controversial procedure that provides information about genetic traits of an embryo, but, with few exceptions (e.g. trisomy 22), not the embryo’s intrinsic biological competence. These factors contribute to the finding that only approximately 15% of embryos transferred in IVF result in a pregnancy.2 In order to overcome this limitation, the practice of multiple embryo transfer has been adopted as the standard of clinical practice. This, in turn, has led to the high incidence of multiple births that is typically observed in IVF. This situation has given rise to the ‘IVF dilemma’: how can IVF programs maintain high pregnancy rates while reducing the number of embryos transferred, to limit the incidence of multiple births?



HPE_Chapter20.qxp

7/17/2007

3:38 PM

Page 246


DEFINITIONS

Metabolomics is an emerging science that represents the continuation of the ever-expanding disciplines we call the ‘omics’.3–6 Metabolomics represents the study of cellular processes that are downstream events of those studied in the fields of genomics and proteomics. As a new scientific discipline, the nomenclature of metabolomics is rapidly evolving in parallel with the science itself, and is only now being defined. As a consequence, there is currently no universal definition of metabolomics, or its companion techniques; some of the more commonly used definitions that are used to frame the concept of metabolomics include: (1) the study of global metabolite profiles in a system (cell, tissue, or organism) under a given set of conditions; (2) the ‘systematic study of the unique chemical fingerprints that specific cellular processes leave behind’, specifically, the study of their small molecule metabolite profiles; and (3) the systematic analysis of the inventory of metabolites, as small molecule biomarkers, that represent the functional phenotype at the cellular level (Figure 20.1.) The metabolome 3 refers to the complete inventory of small-molecule, non-proteinaceous compounds, such as metabolic intermediates, ATP, fatty acids, glucose, cholesterol, hormones, and other signaling molecules, and secondary metabolites that are found within a biological sample. These molecules are the ultimate products of cellular metabolism.

Metabolomics Metabolomics can be thought of as the interface between genomics and a functional phenotype: Genomics → Transcriptomics → Proteomics → Metabolomics → Functional Phenotype Definition: systematic analysis of the inventory of metabolites – as small molecule biomarkers – that represent the functional phenotype at the cellular level. Figure 20.1 Definitions of ‘metabolomics’. Metabolomics is a natural offshoot of the fields of genomics and proteomics, and represents the downstream events of those cellular processes.



The metabolome is dynamic, changing from second to second, based upon the activation and interaction of different metabolic pathways within the cell; thus, it is of particular importance to the rapidly changing development of the embryo in nature and its interaction with the changing conditions in the laboratory. Ultimately, the metabolome represents the global population of biomarkers that can be assessed in the study of metabolomics. Mining of the metabolome provides information that best describes the phenotype, and this can also be used to infer gene function. The slightly different term metabonomics is commonly used in the context of drug toxicity assessment. There is some discussion over the exact differences between ‘metabolomics’ and ‘metabonomics’, but in general, the term metabolomics is more commonly used in the context of emphasis on comprehensive metabolic profiling, while metabonomics attempts to quantify the multiple metabolic changes caused by a biological perturbation. In practice, there is still a large degree of overlap in the way the terms are used, and they are often used interchangeably. THE CASE FOR METABOLOMICS

The dogma of molecular biology suggests that DNA is transcribed into RNA which is then translated into proteins. Protein and post-translational modifications of proteins ultimately give rise to small molecules within the cell which become the biomarkers that can be studied by metabolomics. It is generally agreed that there are more than 25 000 genes, 100 000– 200 000 transcripts, and up to 1 000 000 proteins, whereas there may be as few as 2500–3000 small molecule metabolites/biomarkers that make up the human metabolome. By systematically measuring this population of small molecule biomarkers or metabolites, scientists can establish ‘profiles’ or ‘signatures’ of healthy individuals versus those with specific illnesses. The realization that knowing the DNA sequence of the human genome, or that of other organisms, does not in itself explain the fundamental nature of many normal biological processes and disease states has triggered a marked increase in seeking paradigms that can relate gene expression to the phenotype.7

HPE_Chapter20.qxp

7/17/2007

3:38 PM

Page 247

METABOLOMIC PROFILING OF OXIDATIVE STRESS BIOMARKERS

Even with the sequencing of the total human genome, it is clear that our knowledge of gene function is limited; for example, as many as 35–40% of open reading frames have no known function.3 Consequently, there is a need to better understand the downstream effects of gene expression, which are translated via the transcriptome or proteome, and amplified at the level of the metabolome. This allows for a greater degree of sensitivity when mining the biomarker population of the metabolome compared with other levels of cellular regulation. More importantly, metabolic fluxes are not regulated by gene expression alone (considering post-transcriptional and post-translational events), so the metabolome is considered to be the closest expression of phenotype.8 Metabolomics can provide indications of a metabolic problem or lesion with high throughput at lower costs than genomics, transcriptomics, or proteomics, and is therefore well suited for widespread investigations in the life sciences. In summary, metabolomics is an extension of the growing understanding of the body’s many different genetic and molecular players, including genes and proteins. We know that small changes in the flux through an intracellular pathway can lead to large changes in the concentration of metabolites.9 These changes can be detected by metabolomic profiling of key biomarkers, thus signaling when crucial disturbances have occurred or are about to occur. In contrast, research focusing on genomics or proteomics alone only examines a small portion of the essential events that define normal biological function. While genes can tell us what we might become or the potential for a biological error, and proteins tell us which genes have been expressed, metabolomics provides us with a real-time ‘snapshot’ of the downstream events that characterize gene expression. For these reasons, metabolomics is expected to be a relevant tool in the management of various medical conditions, and, thus an important method for studying basic biological function in concert with other methods.

of Chemistry at the University of California at San Diego. Although it was not called metabolomics at the time, Robinson and Pauling published their first paper in 1971 in the Proceedings of the National Academy of Sciences.10 Robinson’s core concept was that information-rich data that reflect the functional status of a complex biological system reside in the quantitative and qualitative pattern of metabolites in body fluids. It was his expectation that body fluid analysis could be optimized to create a low cost, informative, and medically relevant means of measuring metabolic changes even when standard clinical chemistry markers were in the ‘normal range’. Later, the term ‘metabolomics’ was coined in 1998.11 Six years later, in 2004, the Metabolomics Society (http://www.metabolomicssociety.org) was formed to promote the study of metabolomics. The Human Metabolome Project (HMP) was launched in Canada in 2005 and has identified approximately 2000 metabolites that make up the human metabolome. The goal of HMP is to make this information available worldwide through its database (www. hmdb.ca). Many of the bioanalytical methods used for metabolomics research have been adapted from existing biochemical techniques and enhanced by the advent of newer methodologies, such as various forms of spectral analysis. Three characteristics common to metabolomic research are: (1) metabolite or small molecule biomarkers are profiled without bias towards a specific analyte or group of metabolites; (2) the biomarkers or metabolites are identified and quantified by various forms of analytical, biochemical and spectral analysis; and (3) relationships between the metabolites are characterized, primarily by multivariate mathematical methods often referred to collectively as bioinformatics or computational biology.

BIOINFORMATICS AND COMPUTATIONAL BIOLOGY

HISTORY

It is generally assumed that metabolomics was conceived in 1970 by Arthur Robinson of the Department

In 1952, the British neurophysiologists and Nobel Prize winners Hodgkin and Huxley constructed a mathematical model of the nerve cell. In 1960, 

HPE_Chapter20.qxp

7/17/2007

3:38 PM

Page 248


Denis Noble developed the first computer model of a beating heart. Since then, mathematical computations of biological systems have rapidly escalated to include analysis, quantification, and descriptions of complex molecular level events. These progressively important developments quickly gave rise to the field of bioinformatics and computational biology, which is defined as the use of techniques including applied mathematics, informatics, statistics, computer science, artificial intelligence, chemistry, and biochemistry to solve biological problems, usually at the molecular level. Research in computational biology often overlaps with systems biology. Major research efforts in the field include sequence alignment, gene finding, genome assembly, protein structure alignment, protein structure prediction, prediction of gene expression, and protein–protein interactions, and the modeling of evolution. The terms bioinformatics and computational biology are often used interchangeably. However, bioinformatics more properly refers to the creation and advancement of algorithms, transformations, discriminant analysis, computational, and statistical techniques and theory to solve formal and practical problems posed by or inspired from the management and analysis of biological data. Computational biology, on the other hand, refers to hypothesis-driven investigation of a specific biological problem using computers, carried out with experimental and simulated data, with the primary goal of discovery and the advancement of biological knowledge. A typical metabolomics experiment, for example, will generate volumes of data that need to be converted into useful information and knowledge.

SYSTEMS BIOLOGY

Systems biology studies the interactions between the components of a biological system, and how these interactions give rise to the function and behavior of that system; for example, the interactions between enzymes and metabolites in a specific metabolic pathway.12 Typically, a cellular network is modeled mathematically. Due to the large number of parameters, variables, and constraints in cellular networks, 

numerical and computational techniques are used. Other aspects of computer science are also used in systems biology, including text mining to find parameter data from literature, online databases, and repositories for sharing data and models. The systems biology approach is characterized by a cycle of theory, computational modeling and experiment to quantatively describe cells or cell processes.13 Since the objective is to model all the interactions in a system, the experimental techniques that most suit systems biology are those that are system-wide and attempt to be as complete as possible. Therefore, metabolomics, proteomics, and highthroughput techniques are used to collect quantitative data for the construction and validation of models. ANALYTICAL TECHNIQUES USED IN METABOLOMICS

Metabolite or biomarker analysis typically addresses two issues: (1) separation of the biomarkers, usually by a form of chromatography or electrophoresis, particularly capillary electrophoresis; and (2) detection and quantification of the biomarkers, depending on the methodologies used. A summary of the more commonly used separation and detection techniques is provided below. SEPARATION METHODS

GAS CHROMATOGRAPHY

Gas chromatography (GC) is often used in concert with mass spectrometry (MS). This combination method, often referred to as GCMS, is one of the most widely used and powerful methods. It offers very high chromatographic resolution, but requires chemical derivatization for many biomolecules: only volatile chemicals can be analyzed without derivatization. Many large and polar metabolites cannot be analyzed by GC. HIGH PERFORMANCE LIQUID CHROMATOGRAPHY

Compared with GC, high performance liquid chromatography (HPLC) has lower chromatographic

HPE_Chapter20.qxp

7/17/2007

3:38 PM

Page 249


resolution, but it does have the advantage that a much wider range of analytes can potentially be measured. CAPILLARY ELECTROPHORESIS

Capillary electrophoresis (CE) is the newest tool to be added to the armamentarium of methods used for metabolomic profiling. There are a number of advantages of CE: it has a higher theoretical separation efficiency of analytes than HPLC, and is suitable for use with a wider range of metabolite classes than is GC. As for all electrophoretic techniques, it is most appropriate for charged analytes. DETECTION METHODS

MASS SPECTROMETRY

Mass spectrometry is perhaps the most commonly used technique to identify and, in some (but not all) cases, to quantify metabolites after separation by GC, HPLC, or CE. GCMS is the most ‘natural’ combination of techniques, and was the first to be developed. Measuring time-of-flight of fragmentation pattern constituents also provides useful information regarding the makeup of the specimen being analyzed by MS. In addition, mass spectral fingerprint libraries exist, or can be developed, allowing identification of a metabolite according to its fragmentation pattern. MS is both sensitive and specific. There are also a number of metabolomic studies which use MS as a stand-alone technology: the sample is infused directly into the mass spectrometer with no prior separation. In this scenario MS serves to both separate and detect metabolites. A disadvantage of mass spectrometry is that the specimen is consumed in the analysis, and cannot be recovered. MS instrumentation for metabolomic analysis is generally very costly. NUCLEAR MAGNETIC RESONANCE SPECTROSCOPY

Nuclear magnetic resonance (NMR) has the added advantage that the metabolites don’t always need to be separated prior to analysis, and the sample can thus be recovered for further analyses. Several small

molecule metabolites can be measured simultaneously. NMR is close to being a universal detector, as it produces detailed structural information about the compounds. However, it also possesses one major disadvantage, which is that it is relatively insensitive compared with mass spectrometry-based techniques. NMR is not generally regarded as a reliable quantitative method. It is also the most complex and expensive instrumentation to acquire and maintain. FOURIER TRANSFORM INFRARED SPECTROSCOPY

Fourier Transform Infrared spectroscopy (FTIR) is a physicochemical method that measures vibrations around chemical bonds within different functional groups of molecules using light electromagnetic radiation at specific wavelengths. This method gives quantitative information about the total biochemical composition of a specimen without consuming it. However, substantial sample preparation is required before analysis can be initiated by FTIR, thus adding time and technical requirements to the analysis. RAMAN AND NEAR INFRARED SPECTROSCOPY

While MS and NMR are by far the two leading technologies used for metabolomics research, Raman and near infrared (NIR) spectroscopy have also been used successfully to identify and quantify biomarkers in metabolomics. These methods have similar levels of analytical sensitivity to the previous methods but also have several added advantages: direct sample measurement – no preparation is required; little chemical bias; instrumentation costs are considerably less; instrumentation size is much smaller and the instruments are easier to run and maintain; and rapid, simultaneous analysis of multiple biomarkers at one time. These techniques are not always suitable for all biomarker targets, but are becoming more widely used for a number of metabolomic applications. Our scientific team has successfully utilized Raman and NIR spectroscopy as the platform for its metabolomics technology and has coined the phrase ‘biospectroscopy-based metabolomics’ or BSM for such applications. 

HPE_Chapter20.qxp

7/17/2007

3:38 PM

Page 250


KEY APPLICATIONS

Metabolomics is an example of a true ‘disruptive technology’ the availability of which is expected to markedly impact discovery-driven as well as hypothesis-driven research paradigms. As a result, metabolomics is gaining interest across a wide variety of disciplines, including functional genomics, biomarker discovery, diagnostic and therapeutic monitoring, drug discovery and development, pharmacogenomics, and other life science applications. As metabolic biomarkers are identified they can be used to better understand specific disease processes based on the pathways involved. This information can give rise to therapeutic intervention through new drug discovery. Metabolomic studies of cells in culture has added immensely to the understanding of biological diversity and complexity in both basic science and medicine.

THERANOSTICS

Metabolomic technology will enable the healthcare industry to rapidly move towards the practice of personalized medicine, or ‘theranostics’, allowing the pharmaceutical and biotechnology industries to shift from products that only treat disease to new analytical methods that can detect, monitor, and help stratify illnesses. This trend in medical care looks at disease as a process rather than a state, and allows physicians to diagnose as well as track disease progression in order to provide the most appropriate treatment interventions on a patient-by-patient basis.14 Pharmaceutical and biotechnology companies are using metabolomics to develop ‘companion diagnostics’ so that their new therapeutic compounds can be used more effectively and safely, following FDA approval. DRUG DISCOVERY AND DEVELOPMENT: PHARMACODIAGNOSTICS

MOLECULAR DIAGNOSTICS

The sensitivity and specificity of metabolomic profiling is expected to lead to the development of a new class of molecular diagnostics that provides rapid, cost-effective assessment of the functional phenotype in health and disease. Metabolomic analysis can pinpoint changes in complex networks of metabolic reactions, where outputs from one enzymatic pathway are inputs to other chemical reactions, thus identifying deviations in normal biology. Metabolomics can be used to systematically distinguish between the often subtle differences that separate normal physiology from the onset or progression of disease or an individual’s response to a therapeutic compound, thus leading to (1) new diagnostic and monitoring paradigms; (2) the formation of new industry standards of specificity and sensitivity; (3) the generation of reliable, new measurements on which insurance providers can establish more accurate and thus broader reimbursement criteria; and (4) new designs for sensitive, high throughput instruments that yield cost-effective, user-friendly devices that produce rapid results without the need for specimen preparation.



Metabolomic profiling of biomarkers has the potential to revolutionize drug discovery and development through novel diagnostic testing paradigms referred to as pharmacodiagnostics. This approach is expected to generate more efficacious clinical trial designs that achieve the goals of cost-effective and expeditious clinical research. Pharmacodiagnostic strategies that are based on metabolomics and pharmacokinetic monitoring have the potential to reduce clinical trial failures and shorten drug development timelines, as well as to create new measurements of safety and efficacy that can be used to predict and monitor adverse events. With increased regulatory oversight, pharmacodiagnostics are also expected to be used increasingly in the design of phase IV or FDA-imposed postmarketing studies. BIOMARKER DISCOVERY

The identification of metabolites that could serve as biomarkers of disease is a key application of metabolomics. Over time, reliance on single analyte/biomarker efficiency has been replaced by the

HPE_Chapter20.qxp

7/17/2007

3:38 PM

Page 251


concept of multivariate biomarker profiles that are more sensitive and predictive. An initial approach in biomarker development is to create databases from metabolomic profiling of control populations, and then test populations having a specific disease or condition of interest. This approach statistically compares the metabolomic profile of the diseased population with the profile of healthy controls. The data are mined by bioinformatics analysis in order to identify the distribution of metabolites/biomarkers that can discriminate the disease state and which could then be used as predictors of that condition. The challenge in this process is to utilize rigorous statistical methodologies followed by validation of the database. The biomarkers can be further validated for their predictive or monitoring value for clinical applications, or can be adopted as surrogate markers for pharmacokinetic and pharmacodiagnostic applications. Finally, once the biological function of individual biomarkers is established, they may become viable molecular targets for drug development processes. TOXICITY ASSESSMENT/TOXICOLOGY

Metabolomic profiling of biological fluids can be used to detect the physiological changes caused by toxic insult of a chemical or a mixture of chemicals. These applications can address environmental, agricultural, or medical issues. In many cases, the observed changes can be related to specific syndromes, e.g. a specific lesion in the central nervous system (CNS), liver, or kidney. This is of particular relevance to pharmaceutical companies that need to test the efficacy as well as toxicity of potential drug candidates. Eliminating a compound on the grounds of poor efficacy or adverse toxicity before it reaches clinical trials could save enormous expense and time in otherwise faulty trials. FUNCTIONAL GENOMICS

Metabolomics is an excellent tool for determining the phenotype created in nature or by genetic manipulation such as gene deletion or insertion.

For example, metabolomics can be used to detect any phenotypic changes in a genetically modified organism or in a plant species intended for human or animal consumption. More exciting is the prospect of predicting the function of unknown genes by studying the metabolomic profiles of biomarkers that result from the deletion or insertion of known genes. Such advances are likely to come from models based on microorganisms,15 plants,16 and experimental animals. Transgenic mouse models of Huntington’s disease, for example, have already demonstrated the power of metabolomics to decipher the phenotypic human clinical picture in an animal carrying the transfected gene.17 Likewise atherosclerosis in laboratory mice has been diagnosed more efficiently and accurately using metabolomics than by genomics or proteomics alone.18 HOST–PATHOGEN INTERACTIONS

Metabolomic investigations are also expected to prove valuable in host–pathogen interactions.19 Using plant models, MS was used to identify discriminatory lipid metabolites and certain phospholipids that are know to be important in protecting plants from infectious diseases.20 Parallel applications in human infectious disease are equally plausible given the cascade of metabolic events that are mounted by a host’s defense mechanisms in response to infectious agents.21 NUTRIGENOMICS

This is a generalized term which links genomics, transcriptomics, proteomics, and metabolomics to human nutrition. In general the metabolome of a given body fluid is influenced by endogenous factors such as age, sex, body composition, and genetics as well as underlying pathologies. Diet and drugs are the main exogenous factors, and diet can then be further broken down to nutrients and nonnutrients. Metabolomics can be used to determine the biological endpoints that reflect the balance of these variables in an individual’s nutritional makeup.



HPE_Chapter20.qxp

7/17/2007

3:38 PM

Page 252


APPLICATIONS OF METABOLOMICS IN THE ASSISTED REPRODUCTIVE TECHNOLOGIES METABOLOMICS-BASED EMBRYO SELECTION

A technology that is capable of identifying viable gametes and embryos in an IVF program is expected to lead to a reduction in the number of embryos transferred and, eventually, the adoption of single embryo transfer (SET). The availability of rapid, non-invasive metabolomic profiling of embryos could potentially empower practitioners to overcome the IVF dilemma and move away from the current paradigm of multiple embryo transfer. The practical benefit of rapid, non-invasive metabolomic profiling in IVF is the improvement of embryo and gamete selection procedures thus allowing a reduction in the number of embryos transferred without compromising pregnancy rates. The longer term benefit of this technology to patients, physicians, and insurance providers is the potential to limit the incidence of multiple births and associated healthcare risks and expense.

(2)

(3)

CURRENT APPLICATIONS

The advent of metabolomics in the field of assisted reproduction is very recent. Only a small number of abstracts and papers have been published prior to the work of the Metabolomics Study Group for ART. The publications in the field so far are summarized below. (1) FTIR spectroscopy. Thomas et al22 reported substantial differences in the biochemical fingerprints of follicular fluid from large follicles (⬎17 mm) versus small (⬍15 mm) antral follicles in patients undergoing IVF. The differences in the distribution of steroid concentrations between large and small follicles were significant but did not follow protein bonding characteristics as observed by FTIR. While not labeled specifically as a metabolomics study, the authors concluded that the differences detected in metabolite profiles by FTIR reflect the developmental stage of the follicle, and possibly oocyte 

(4)

(5)

quality. FTIR has also been used successfully to examine biological fluids for other medical applications.23,24 HPLC. Houghton et al25,26 and Brison et al27 investigated amino acid turnover of embryo culture media to establish metabolomic profiles, or signatures, of developing human embryos. Using HPLC as their analytical tool, the authors observed differential amino acid compositions in the culture media of viable versus non-viable embryos and correlated their findings with pregnancy outcomes. The studies concluded that amino acid analysis may be used as a non-invasive test to predict embryonic viability. FTIR spectroscopy. Hollywood (in preparation) used FTIR spectroscopy to examine the metabolomic profile of biomarkers in embryo culture media. Cluster analysis of media specimen FTIR spectra from embryos that did not produce a pregnancy differed significantly from that of media from embryos that resulted in positive biochemical and clinical (ultrasound) pregnancies. The authors concluded that metabolomic fingerprinting of media can be used to assess the status of individual human embryos and their potential for producing a pregnancy. Respirometry (oxygen consumption). Lopes et al28,29 studied oxidative pathways of metabolism in individual bovine embryos by direct quantification of oxygen consumption. Using a non-invasive micro-electrode, these investigators postulated that embryonic respiration rates, in association with other embryo selection methods, allowed an accurate assessment of embryo quality. The authors discussed the potential use of this technology as a diagnostic tool for improving embryo selection in IVF. GCMS. In a related, but non-ART study, investigators used GCMS in metabolomic studies to assess biomarkers related to pre-eclampsia.30 The authors identified metabolomic profiles in plasma of pregnant women to discriminate those with normal pregnancies from those at risk for developing pre-eclampsia. The metabolomic data showed high sensitivity and specificity, targeting only three biomarkers in the analysis.

HPE_Chapter20.qxp

7/17/2007

3:38 PM

Page 253


DATA FROM THE METABOLOMICS STUDY GROUP FOR ART

The Metabolomics Study Group for ART has conducted proof of principle studies of gamete and embryo viability using a proprietary metabolomics platform and bioinformatics developed by Molecular Biometrics, LLC, Chester, NJ. Two retrospective studies (Yale University, Stanford University and Reproductive Biology Associates of Atlanta) and one prospective blinded study (Reproductive Medicine Associates of New Jersey) were performed.31,32 STUDY AIMS

The Metabolomics Study Group for ART was organized to test the hypothesis that biomarkers of oxidative stress (OS) could be targeted using metabolomics technology based on various forms of biospectroscopy, coupled with specific bioinformatics, to assess embryo and gamete viability in IVF; and that the resulting metabolomic profiles could adequately distinguish between viable and non-viable embryos and gametes within a cohort group. TECHNOLOGY PLATFORM

microenvironment of the gametes, i.e. in follicular fluid (oocyte) and in seminal plasma (sperm). Molecular Biometrics technology is based on the confluence of two scientific disciplines: (1) biospectroscopy, the application of different forms of spectral analysis in human biology that is used to identify, quantify, and validate proteomic and small molecule biomarkers; and (2) metabolomics, the science that systematically examines and integrates the dynamic interplay between multiple, small molecule biomarkers that are uniquely characteristic of complex biological functions that define the normal phenotype. The term ‘biospectroscopy-based metabolomics’ (BSM) was coined to describe this technology platform. These methodologies are used in concert to quantify a sample’s molecular biomarker makeup, specifically, biomarkers of oxidative stress, and then convert that data into a novel ‘metabolomic profile’ or ‘fingerprint’. Each profile is analyzed using proprietary bioinformatics and chemometrics to determine the probability of normal, competent cellular function versus the presence or predisposition to abnormal or even pathological states of cellular activity. The technology platform is illustrated in Figure 20.2. Technology Platform A methodology reduced to practice

Free radicals such as reactive oxygen species (ROS) exert their effect at the molecular level in all cell types and play a role in both normal physiological and pathological functions. Complex interactions between the pro-oxidants and antioxidants are crucial in the maintenance of intracellular homeostasis. An imbalance in these reactions results in oxidative stress, which has been implicated in a number of medical conditions and diseases.33,34 Moreover, biomarkers of OS have been found in the male and female reproductive tracts and are known to affect the quality of gametes, early embryo development, and implantation, which in turn affect pregnancy outcomes. ROS have been shown to influence sperm, oocytes and embryos and to modulate their interaction in their respective microenvironments.35 The technology platform of Molecular Biometrics targets biomarkers of oxidative stress in culture media of the developing embryo or in the respective

Hypothesis driven

Discovery driven

Biomarker targets (oxidative stress, proteomic, small molecules) Spectral analysis (Metabolomic profile by NMR, Raman, NIR) Chemimetrics and bioinformatics (Bayesian statistics, Haar transform, Matlab, Genetic algorithm) Phenotypic function

Figure 20.2 Technology platform. Diagrammatic representation of the pivotal steps that constitute the technology platform for assessing biomarkers of oxidative stress using metabolomic profiling. The platform can be driven from either a hypothesis or discovery approach and includes biomarker targets, methods of spectral analysis, chemometric and bioinformatic analysis of data, leading to an assessment of the phenotypic function of the cell or organism. NMR, nuclear magnetic resonance; NIR, near infrared. Courtesy of Molecular Biometrics, LLC, Chester, NJ, USA.



HPE_Chapter20.qxp

7/17/2007

3:38 PM

Page 254


Disposable sample cell Sample 2.1” Light

Ball lenses

OR electronic ‘telephone card’

Microprocessor LED display LED light source

Interference filter Array detector

USB To computer

Figure 20.3 Instrument prototype design. The instrument design for obtaining spectral signatures of biomarkers utilizes a disposable sample cell that contains 5–10 ␮l of specimen. The sample cell is inserted into a fixed position in the light path within the instrument. The stable light source (left) passes light (energy) into the sample, which then travels through a wavelength filter to select the specified wavelength for the analysis. The photons of energy which emerge from the filter are captured by a light-detecting diode and delivered to the bioinformatics microprocessor chip (far right) where the biomarker profiles are calculated. The total time of analysis is approximately 1 minute. The instrument can be interfaced with any computer system to download data to patient files. This prototype design has now been incorporated into a commercial product for large scale production.

INSTRUMENTATION

The metabolomics platform described here has been reduced to practice within the least expensive and most compact instrumentation available. The goal was to engineer a generation of affordable, point-of-care instruments for direct metabolomics applications that are accessible to all IVF laboratories. Unlike other metabolomics technologies that rely on very large and costly NMR, FTIR, and MS instruments, the current metabolomic applications are based on Raman and NIR spectroscopy. These instruments are robust, easy to use, do not require specimen preparation and do not destroy or consume the specimen. The device can interface with any computer system and can download data directly into patients’ medical records. 

Metabolomic profiling was performed using a small, prototype instrument (6 inches ⫻ 6 inches ⫻ 4 inches) that was programmed with proprietary spectroscopic and bioinformatics capabilities (Figure 20.3.) Initially, sample cells were filled with 30 microliters of discarded culture media obtained at the time of embryo assessment and selection. The original prototype has been replaced by a commercially manufactured instrument having a sample size of 5–10 microliters. This high-speed procedure takes about 1 minute to complete. STUDY DESIGN

Infertility patients enrolling in an IVF program were treated according to the standard IVF protocol of the participating program. Informed consent was

HPE_Chapter20.qxp

7/17/2007

3:38 PM

Page 255


obtained from each patient enrolled in the study and treatment information for that patient, including pregnancy outcome data, was noted in the patient’s medical record. Samples of normally discarded embryo culture media were collected at the time of embryo selection/transfer (Figure 20.4). All specimens were placed in Nunc vials and rapidly frozen in liquid N2 or dry ice and methanol and stored at ⫺80°C or liquid N2. Appropriate media controls were also prepared from media droplets that never received an embryo but were handled in a parallel manner with media droplets containing developing embryos. Pregnancy was confirmed via detection of fetal heartbeat by ultrasound. The pregnancy outcome groups were defined as follows: (1) no implantation or 0% pregnancy: regardless of the number of embryos transferred, no pregnancies were recorded; 50% pregnancy: four embryos were transferred and two implanted producing twins while the two corresponding embryos did not implant; and (2) 100% pregnancy: regardless of the number of embryos transferred, all embryos implanted and produced a

Sampling pipette (5–10 µl volume)

Oil

Embryo

Embryo culture media (35–65 µl microdrop)

Figure 20.4 Non-invasive study design. This illustration depicts the non-invasive procedure for collecting culture media for analysis of embryo viability. A small volume of media, usually 5–10 ␮l, is collected directly from the media microdrop containing the developing embryo. Care is taken to avoid aspiration of oil. The media is inserted directly into the disposable sample cell (as described in Figure 20.3) for analysis.

pregnancy (e.g. two embryos transferred, twins; three embryos transferred, triplets, etc.). All specimens were shipped to McGill University, Montreal Quebec for metabolomic measurements and bioinformatic analysis. METABOLOMIC PROFILING OF OS BIOMARKERS

Specimens were analyzed using both Raman and near infrared spectroscopy, which measures the bonds within functional groups of molecules at specific wavelengths. The spectra obtained from each instrument were separately analyzed by proprietary bioinformatics using a wavelength selective genetic algorithm (GA) to determine regions predictive of pregnancy outcome as determined by a logistic regression of light attenuation at specified wavelengths. To avoid random correlations, a leave-one out cross-validation was used. Sensitivity and specificity were calculated for each assay. RESULTS

Two different forms of spectroscopic analysis were used to report metabolomic parameters that were established on the basis of observed OS biomarker targets. Spectral profiles describing differences in -CH, -NH, -SH, C⫽C, and -OH concentrations showed distinct differences between culture media of embryos that resulted in pregnancy compared with those that did not. The ratio of C⫽C to -SH content in the media that is reflective of oxidative stress was also different between the two groups (Figure 20.5). Using GA with Raman spectroscopy, four spectral regions associated with these molecular species were reproducibly identified as predictive of pregnancy outcome in two pilot studies. Compiled outcomes from the leave-one-out cross-validation of the logistic regression using Raman measurements resulted in a specificity of 90% and a sensitivity of 100%. In a separate blinded study, analysis of media specimens by NIR required two wavelength regions identified by GA. The specificity and sensitivity of these assays was 83% and 73%, respectively. Significant differences were observed between 0 and 100% 

HPE_Chapter20.qxp

7/17/2007

3:38 PM

Page 256


Relative signal

A

1 0.5

NH

0 –0.5

C=C, SH groups

–1 0

500

1000

1500

2000

2500

3000

3500

Wavelength (nm)

Number of samples

B

2 1.5 1 0.5 0 –1

–0.5

0

0.5

1

1.5

2

Implantation index Figure 20.5 Raman spectral signature of three biomarkers in culture media. (A) A spectrum produced by Raman spectroscopy from a single embryo media specimen. Three distinct biomarker regions are identified in yellow depicting C苷C, -SH, and -NH bonding at specific wavelengths. The photon energy detected in these regions was captured and subjected to bioinformatic analysis. (B) The implantation index score is calculated by the bioinformatic algorithms for each specimen. The distribution of metabolomic data is shown for each of 20 media specimens: non-viable embryo scores are in blue vs viable embryo scores in red. The sensitivity and specificity for this group of specimens were 100% and 90%, respectively.

pregnancy groups in both day 3 and day 5 assays. Significantly, the 50% group had values that fell between the 0 and 100% groups (Figure 20.6). Using the data obtained in the pilot studies, Bayesian probability statistics were used to establish a predictive index of embryo viability based on a typical Gaussian distribution (Figure 20.7). A total of 432 samples from four different centers were analyzed (n ⫽ 211 embryo culture media; n ⫽ 133 seminal plasma; n ⫽ 88 follicular fluid) with assays that routinely achieved high sensitivity and specificity of ⬎80%. The results of these studies demonstrate a clear relationship between the



reproductive potential of human embryos and their modification of the media in which they have been cultured. These subtle modifications could be detected through BSM profiling of OS biomarkers that appear in the media during embryo culture. Specifically, there were detectable differences in the metabolomic profiles found in culture media obtained from embryos that resulted in pregnancy compared with those that did not. Likewise, significant differences were seen in the biomarker profiles of normal, healthy semen donors compared with those with various forms of male infertility, including idiopathic male factor infertility,

HPE_Chapter20.qxp

7/17/2007

3:38 PM

Page 257


A

Day

5

3

–0.6

–0.4

–0.2

0

0.2

0.4

0.6

0.8

1.0

1.2

Implantation index

% Implantation

B 100 Day 5 50 0 0.5

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

Implantation index

% Implantation

C 100

Day 3

50 0 –0.6

–0.4

–0.2

0

0.2

0.4

0.6

Implantation index Figure 20.6 Metabolomic results by NIR of 0, 50%, and 100% implantation rate embryos. Media from embryos that produced 0, 50%, and 100% pregnancies were analyzed in a blinded prospective study; n ⫽ 65 with an equal number of specimens per group. (A) A statistically significant difference was observed in the metabolomic profiles of day 3 embryos that produced a pregnancy compared with day 5 embryos. Day 5 embryos showed a greater implantations index score than day 3 embryos. t Statistic ⫺7.86; two-tailed p ⬍0.0001. (B) A comparison of implantation index scores of three groups of day 5 embryos: 0%, 50%, and 100% pregnancy outcomes. The 0% and 100% groups are statistically different from each other (t Statistic 3.85; p ⫽ 0.0085).The implantations index score of the 50% group falls between the 0 and 100% group scores. (C) A comparison of implantation index scores of three groups of day 3 embryos: 0%, 50%, and 100% pregnancy outcomes. The 0% and 100% groups are statistically different from each other (t Statistic 3.23; p ⫽ 0.003), while the implantation index score of the 50% group fell between the 0 and 100% group scores.

varicocele, and vasectomy reversal (Figure 20.8). Profiles characteristic of oocyte quality were also detectable in follicular fluid and correlated to pregnancy outcomes of their respective transferred embryos (Figure 20.9).

CONCLUSION

This technology offers significant potential as a tool to allow rapid non-invasive assessment of embryonic reproductive potential prior to transfer.



HPE_Chapter20.qxp

7/17/2007

3:38 PM

Page 258


+Pregnancy +Non-pregnancy

00..9 8

00. .76 0 00.0.3.4.5 2

300

1 0.

Integrated mean centered SH

5 000.0..43.2 1 0.

400

200

00.8.9

0.9 0.8 0.7 0.6 4 0. 0.3 0.2 0.1

00..67 00.5 0. 00.2.3.4 1

100 0 –100 –2500

–2000

–1500

–1000

–500

Probability of viability contours

0

Integrated mean centered CH Figure 20.7 Bayesian probability statistics. A distribution of all results from embryos (pregnancy and non-pregnancy) was used to create this Bayesian probability model. Only results from -SH and -CH biomarker populations were included to construct this curve. The Bayesian database was then used to assess a single embryo’s metabolomic score, based on comparison with a representative population of embryo scores between 0 and 100% implantation, to predict that embryo’s relative probability of implantation.

0.05

Relative absorbance

0.04 0.03

Varicocele Idiopathic Vasectomy reversal Normal donors

0.02 0.01 0 –0.01 –0.02 550 600 650 700 750 800 850 900 9501000 1050 Wavelength (nm)

Figure 20.8 Near infrared spectra from seminal plasma. Seminal plasma from four groups of men with infertility were studied by metabolomic profiling: normal healthy donors, patients with idiopathic male factor infertility, patients with varicocele, and patients who underwent vasectomy reversal. A qualitative and quantitative difference in metabolomic profiles could be discerned between all groups of male infertility compared with normal donors. The most pronounced difference was observed between donors and patients with idiopathic male factor infertility. The groups were statistically different from each other with assay sensitivity and specificity of ⬎95% (data not shown).



Enhanced selection parameters based on metabolomic profiling might allow a reduction in transfer order, eventually to single embryo transfer, with the desirable effect of lowering multiple gestation rates while maintaining or raising pregnancy rates. The application of metabolomic profiling to assess gamete viability appears to be equally promising. For example, this technique could be adapted as a rapid diagnostic screening test for male factor infertility. The ability to determine oocyte viability prior to fertilization is expected to increase the number of viable embryos available for selection, transfer, and cryopreservation. Conceivably, metabolomics could work in concert with, or in place of, PGD as a form of functional genomics testing. In certain European countries where limitations on embryo transfer are tightly regulated, viable oocyte selection could provide a relevant means of sustaining IVF success rates through improved oocyte fertilization and cryopreservation techniques. Viability testing of gametes and embryos used in cryopreservation cycles is also a logical application of metabolomics in IVF.

HPE_Chapter20.qxp

7/17/2007

3:38 PM

Page 259


A Relative absorbance

1

0

–1 650

700

750

800

850

900

950

Wavelength (nm)

Number of samples

B

6 4

No pregnancy

Pregnancy

2 0 Implantation index

Figure 20.9 Near infrared spectra of follicular fluid. (A) A spectrum produced by NIR spectroscopy from a single follicular fluid specimen. Six distinct biomarker regions are identified in yellow. The photon energy detected in these regions was captured and subjected to bioinformatic analysis. (B) The implantation index score was calculated by the bioinformatic algorithms for each specimen. The distribution of metabolomics data of each of 88 follicular fluid specimens showing the pregnancy outcome data of oocytes from their respective follicular fluid that did not result in a pregnancy (blue) vs oocytes that produced viable embryos and a pregnancy (red). The sensitivity and specificity for this group of specimens were 81% and 77%, respectively.

ADDITIONAL ART STUDIES

Confirmation of these observations is planned through a large prospective study involving several ART centers. Currently, a multicenter, multinational study is underway in nine countries using a common Institutional Review Board (IRB), approved protocol and informed consent. IVF programs in the USA, Canada, Japan, Belgium, Italy, The Netherlands, Finland, Spain, and Sweden are expected to add up to 500 additional specimens to the current database. In addition, more than half of the IVF cycles will be single embryo transfer cycles. This refined study design will provide unequivocal data demonstrating the relationship between an embryo’s metabolomic profile and its implantation potential. A Metabolomic Study Group for ART

is being formed in Europe to facilitate this and further investigations in the field using metabolomic profiling. ADDITIONAL INDICATIONS IN REPRODUCTIVE HEALTH

Beyond the immediate applications of embryo and gamete viability testing, the Study Group will pursue the use of metabolomics for functional genomics testing and to assess endometrial receptivity for implantation. In the former scenario, metabolomics is viewed as a possible adjunct or replacement technology for aneuploidy screening, and perhaps other genetic testing. With regard to implantation, study designs employing metabolomics are being reviewed that allow non-invasive examination of the endometrial lining of the uterus just prior to embryo transfer. 

HPE_Chapter20.qxp

7/17/2007

3:38 PM

Page 260


In the field of maternal fetal medicine, Molecular Biometrics (Chester, New Jersey, USA), has applied metabolomics to assess fetal development by examining biomarkers in amniotic fluid. In several hundred patients studied to date, final fetal birth weight, intrauterine growth retardation/small for gestational age, gestational diabetes mellitus, and preterm labor have been characterized by metabolomic profiling (Molecular Biometrics, data on file). INDICATIONS FOR USE IN OTHER FIELDS

This same metabolomics platform has also been applied to the assessment of neurodegenerative disease, using biomarker profiles to distinguish between patients with Alzheimer’s disease, Parkinson’s disease, mild cognitive impairment, and age-matched controls. Similarly, pulmonary edema and lactate levels have been accurately diagnosed and monitored non-invasively using this same metabolomics technique (Molecular Biometrics, data on file.) These indications are also entering clinical development programs for development of their respective medical applications. ACKNOWLEDGMENTS

The Metabolomics Study Group for ART is compromised of the following investigators: Ashok Agarwal, The Cleveland Clinic Foundation, Cleveland, Ohio; Barry Behr, Stanford University, Palo Alto, California; David Burns, McGill University, Montreal, Quebec, Canada; Joe B Massey, Reproductive Biology Associates, Atlanta, Georgia; Peter Nagy, Reproductive Biology Associates, Atlanta, Georgia; Denny Sakkas, Yale University, New Haven, Connecticut; Richard J Scott, Reproductive Medicine Associates of New Jersey, Morristown, New Jersey; and Emre Seli, Yale University, New Haven, Connecticut. REFERENCES 1. Sakkas D, Gardner DK. Noninvasive methods to assess embryo quality. Curr Opin Obstet Gynecol 2005; 17: 283–8. 2. Patrizio P, Kovalevsky G. High rates of embryo wastage with use of assisted reproductive technology: a look at the trends between 1995 and 2001 in the United States. Fertil Steril 2005; 84: 325–30.



3. Hollywood K, Brison D, Goodacre R. Metabolomics: current technologies and future trends. Proteomics 2006; 6: 4716–23. 4. Harington GG. Metabolomics Metabolomics: a “systems” contribution to pharmaceutical discovery and drug development. Drug Discov World 2005; 39–46. 5. Lindon JC, Nicholson JK, Holmes E et al. Metabonomics: metabolic processes studied by NMR spectroscopy of biofluids. Concepts Magn Reson 2000; 12: 289–320. 6. Dunn WB, Bailey NJC, Johnson HE. Measuring the metabolome: current analytical technologies. Analyst 2005; 130; 606–25. 7. Nicholson JK, Lindon JC, Holmes E. Metabolomics: understanding the metabolic responses of living systems to pathophysiological stimuli via multivariate statistical analysis of biological NMR data. Xenobiotica 1999; 29: 1181–9. 8. Ter Kuile BH, Westerhoff HV. Transcriptome meets metabolome: hierarchical and metabolic regulation of the glycolytic pathway. FEBS Letts 2001; 500: 169–71. 9. Raamsdonk LM, Teusink B, Broadhurst D et al. A functional genomics strategy that uses metabolome data to reveal the phenotype of silent mutations. Nat Biotechnol 2001; 19: 45–50. 10. Pauling LC, Robinson AB, Teranishi R et al. Quantitative analysis of urine vapor and breath by gas-liquid partition chromatography. Proc Natl Acad Sci 1971; 68: 2374–6. 11. Oliver SG, Winson MK, Kell DB et al. Systematic functional analysis of the yeast genome. Trends Biotechnol 1998; 16: 373–8. 12. Snoep JI, Westerhoff HV. From isolation to integration, a systems biology approach for building the silicon cell. In: Alberghina L, Westerhoff HV, eds. Systems Biology: Definitions and Perspectives. SpringerVerlag, 2005: 7. 13. Kholodenko BN, Bruggeman FJ, Sauro HM. Mechanistic and modular approaches to modeling and inference of cellular regulatory networks. In Alberghina L, Westerhoff HV, eds. Systems Biology: Definitions and Perspectives. Springer-Verlag, 2005: 143. 14. Burrill GS. Biotech 2005, Life Sciences. San Francisco: Burrill & Company, LLC, 2005: 198. 15. Delneri D, Brancia FL, Oliver SG. Towards a truly integrative biology through the functional genomics of yeast. Curr Opin Biotechnol 2001; 12: 87–91. 16. Jenkins H, Hardy N, Beckmann M et al. A proposed framework for the description of plant metabolomics experiments and their results. Nat Biotechnol 2004; 22: 1601–6. 17. Underwood BR, Broadhurst D, Dunn WB et al. Huntington’s disease patients and transgenic mice have similar procatabolic serum metabolite profiles. Brain 2006; 129: 877–8. 18. Clish CB. Integrative biological analysis of the APOE*3-leiden transgenic mouse. Omics 2004; 8: 3–13. 19. Forst CV. Host-pathogen systems biology. Drug Discov Today 2006; 11: 220–7. 20. Allwood JW, Ellis DI, Heald JK et al. Metabolomic approaches reveal that phosphatidic and phosphatidyl glycerol phospholipids are major discriminatory metabolites in responses by Brachypodium distachyon to challenge by Magnaporthe grisea. Plant J 2006; 46: 351–68. 21. Goodacre R, Timmins EM, Burton R et al. Rapid identification of urinary tract infection bacteria using hyperspectral, whole organism fingerprinting and artificial neural networks. Microbiology 1998; 144: 1157–70. 22. Thomas N, Goodacre R, Timmins EM et al. Fourier transform infrared spectroscopy of follicular fluids from large and small antral follicles. Hum Reprod 2000; 15: 1667–71. 23. Jackson M, Sowa MG, Mantsch HH. Infrared spectroscopy: a new frontier in medicine. Biophys Chem 1997; 68: 109–25.

HPE_Chapter20.qxp

7/17/2007

3:38 PM

Page 261


24. Timmins EM, Howell SA, Alsberg BK. Rapid differentiation of closely related candida species and strains by pyrolysis-mass spectroscopy and Fourier transform-infrared spectroscopy. J Clin Microbiol 1998; 36: 367–74. 25. Houghton FD, Hawkhead JA, Humpherson PG et al. Non-invasive amino acid turnover predicts human embryo developmental capacity. Hum Reprod 2002; 17: 999–1005. 26. Houghton FD, Leese, HJ. Metabolism and developmental competence of the preimplantation embryo. Eur J Obstet Gynecol Reprod Bio 2004; 115 (Suppl 1): S92–6. 27. Brison DR, Houghton FD, Falconer D et al. Identification of viable embryos in IVF by non-invasive measurement of amino acid turnover. Hum Reprod 2004; 19: 2319–24. 28. Lopes AS, Greve T, Callesen. Quantification of embryo quality by respirometry. Theriogenology 2007; 67: 21–31. 29. Lopes AS, Madsen SE, Ramsing NB et al. Investigation of respiration of individual bovine embryos produced in vivo and in vitro and correlation with viability following transfer. Hum Reprod 2007; 22: 558–66.

30. Kenny LC, Dunn WB, Ellis DI et al. Novel biomarkers for pre-eclampsia detected using metabolomics and machine learning. Metabolomics 2005; 1: 227–34. 31. Seli E, Sakkas D, Behr B et al. Non-invasive metabolomic profiling of human embryo culture media correlates with pregnancy outcome. Initial results of the Metabolomics Study Group for ART. Fertil Steril 2006; 86 (Suppl); 117. 32. Scott, RT, Miller K, Picnic S. A prospective blinded evaluation of the relationship between metabolomic profiling of spent embryo culture media and human embryonic reproductive potential. Fertil Steril 2006; 86 (Suppl); 235. 33. Halliwell B, Gutteridge JMC, eds. Free Radicals in Biology and Medicine. Oxford: Oxford University Press, 1999. 34. Singh KK, ed. Oxidative Stress, Disease and Cancer. World Scientific Publishing Company, 2006. 35. Agarwal A, Gupta S, Sharma R. Oxidative stress and its implications in female infertility – a clinician’s perspective. Reprod BioMed Online 2005; 11: 641–50.



HPE_Chapter20.qxp

7/17/2007

3:38 PM

Page 262

HPE_Chapter21.qxp

7/13/2007

5:31 PM

Page 263

21. Gene expression analysis in the human oocyte and embryo Nury M Steuerwald

INTRODUCTION

Large-scale analyses of gene expression during human oogenesis and embryogenesis has been hampered by a scarcity of material for analysis coupled with a lack of highly sensitive investigational tools. Historically, a variety of classical molecular techniques have been employed to examine cellular mRNA levels.1–3 However, these methods were generally crude in nature and thus lacked the precision to accurately and reproducibly detect or quantify transcripts in single cells. Reverse transcriptionpolymerase chain reaction (RT-PCR) provided a highly sensitive means to distinguish rare mRNA species in individual cells.4 Together with fluorescent dyes, RT-PCR can now be used to reliably quantify gene expression in individual oocytes.5,6 Subsequently, PCR-based methods, and, more recently, analysis tools that provide genome-wide perspectives such as microarray techniques have been used to glean a wealth of knowledge regarding the global expression profile of the oocyte and embryo.7–9 However, a great deal of work remains to be done in order to fully dissect the complex regulatory pathways that direct early human development. This chapter reviews our current understanding of these developmental programs. GENE EXPRESSION DURING FOLLICULOGENESIS

The human follicular cycle culminates with the expulsion of a single mature oocyte at ovulation, which is arrested in metaphase II. However, the process of follicular maturation begins in utero many years, and even decades earlier. Primordial germ cells

(PGCs) migrate to the genital ridge in the developing embryo, where they actively engage in mitosis. The oogonia subsequently initiate meiosis, and the resulting primary oocytes arrest at the diplotene phase of prophase I, invest themselves with granulosa cells, and thus give rise to primordial follicles. Concomitantly, those oogonia that do not enter meiosis I undergo atresia, resulting in the eradication of all remaining oogonia by the time of birth. Shortly thereafter, primordial follicle formation ceases, but the process of primary follicle development has already begun. During this process, primordial follicles develop an extracellular matrix, the zona pellucida, and a differentiated granulosa layer. Further proliferation of the granulosa layer gives rise to a secondary follicle, which then also acquires a thecal layer.10 Although the origin of PGCs and formation of primordial follicles is not yet well understood at a molecular level, several factors that are expressed in adjoining somatic tissues have been identified which may affect their development. Tumor necrosis factor-␣ (TNF␣) expression has been confirmed in human primordial follicles.11 TNF␣ appears to be involved in the apoptosis of random oocytes, and this is believed to facilitate primordial follicle assembly.12 Leukemia inhibitory factor (LIF) has been shown to induce proliferation and differentiation.13 Expression profiles of Kit ligand (KL) and its tyrosine kinase receptor (c-Kit) in the human ovary suggest that they are involved in human fertility.14 While their precise roles in the human are not known, studies from mouse mutants have revealed that KL and c-Kit are important in the establishment of the PGCs within the ovary, oocyte survival and growth, granulosa cell proliferation, theca cell recruitment, and the maintenance of meiotic arrest.14 

HPE_Chapter21.qxp

7/13/2007

5:31 PM

Page 264


Genes expressed by the PGCs and oocytes themselves are believed to play a role in the development of the oocyte and follicle. Human Factor in the germline alpha (FIGLA)15 appears to be associated with cell–cell interactions and survival during the process of primordial follicle formation.16 The murine counterpart of FIGLA also regulates the expression of genes encoding zona pellucida proteins. This role may possibly be conserved among mammals.17 It has been reported that there are four human zona pellucida genes (ZP1, ZP2, ZP3, and ZPB), not three as previously predicted by the murine model.18 The zona pellucida serves as the first point of contact for sperm–egg interactions, and it is also involved in establishing a block to polyspermy and providing protection for the preimplantation embryo. The human Dazla (deleted in azoospermia like autosome) gene is the autosomal homolog of the Daz (deleted in azoospermia) gene which maps to the long arm of the Y chromosome.19 It encodes a cytoplasmic protein with RNA binding

Primordial germ cell

Primordial follicle

Primary follicle

Ovarian somatic cells: Tnfa, Lif, KL, c-Kit

Oocyte: Gdf9, Bmp15, bFgf, Nobox

PGCs: Figla, ZP genes, Dazla

Cumulus cells: Has2, Ptgs2, Grem1

domains and is believed to be involved with translational regulation. Dazla is expressed in oogonia as well as in oocytes and granulosa cells of primordial follicles. Its expression pattern in humans is similar to that in mouse, where Dazla mutations cause male and female sterility, suggesting that there may also be functional parallels during gametogenesis.19 The transition from primordial to mature follicle is further mediated by the expression of several key genes in the oocyte. Growth differentiation factor (Gdf-9) and bone morphogenic protein (Bmp-15) are oocyte-expressed genes that are both members of the transforming growth factor ␤ superfamily. In the human, transcripts of these genes are present in oocytes of primary follicles and persist following ovulation.20,21 GDF-9 is believed to promote granulosa cell proliferation, as well as induce cumulus cell differentiation and expansion.22 GDF-9 modulates the expression of cumulus granulosa genes such as hyaluronic acid synthase 2 (Has2), prostaglandin-endoperoxide synthase 2 (Ptgs2), and

Secondary follicle

Antral follicle

Figure 21.1 Genes implicated in follicle development. Follicles at various stages of development are depicted. Ovarian somatic tissues, primordial germ cells (PGC), oocytes, and cumulus cells express various factors throughout folliculogenesis. Genes expressed by specific cell types are listed.



HPE_Chapter21.qxp

7/13/2007

5:31 PM

Page 265

GENE EXPRESSION ANALYSIS IN THE HUMAN OOCYTE AND EMBRYO

gremlin (Grem1). The expression of these GDF-9 regulated genes has been shown to correlate with morphological and physiological criteria consistent with higher grade embryos, and this may serve as a biochemical marker for embryo development during in vitro fertilization procedures.23 Bmp-15 shares a high degree of homology with Gdf-9 and is also thought to regulate critical somatic cell activities during folliculogenesis, as evidenced by a recent study implicating Bmp-15 mutations in the pathogenesis of premature ovarian failure in a large cohort of women.24 Basic fibroblast growth factor (BFGF) localizes to oocytes, granulosa, and theca cells in the human ovary.25 Human bFgf temporal and spatial expression patterns hint at a potential role in primordial follicle activation, as predicted by studies in rodents.26 Likewise, studies in the mouse have identified another gene, Nobox (newborn ovary homeobox), expressed in primordial and growing oocytes, which appears to be essential for oocyte survival and follicular development.27 The human gene encoding NOBOX has been identified, but its precise function remains to be elucidated.28 Figure 21.1 presents a schematic representation of selected genes that have been implicated in oogenesis and folliculogenesis.

MESSENGER RNA EXPRESSION, LOCALIZATION, STABILITY, AND REGULATION DURING OOGENESIS

Following many years of arrest at meiotic prophase, the luteinizing hormone (LH) surge attending ovulation induces the resumption of meiosis by modulating cAMP levels, which is believed to bring about the activation of maturation promoting factor (MPF).29 MPF is a highly conserved complex of proteins consisting of two subunits, one which serves a regulatory role (cyclin) and another which possesses kinase activity.30,31 Activated MPF mediates germinal vesicle breakdown by phosphorylating several proteins: histone protein phosphorylation results in chromosomal condensation, and phosphorylation of nuclear envelope lamin proteins

results in its depolymerization and breakdown.32 However, meiosis is again arrested at metaphase II, presumably due to the action of a cytostatic factor (CSF) which contains the protein products of the c-mos and Cdk-2 genes.33,34 CSF prevents the degradation of cyclin until fertilization, when the increase in calcium, probably initiated by the inositol phosphate pathway,35 activates a CSF protease which allows meiosis to proceed.36 It is undeniable that oocytes of several organisms, including certain insects, nematode worms, and amphibians are polarized with respect to the distribution of maternal transcripts and proteins, and this ultimately defines the axes of the embryo.37 However, the details regarding the polarized nature of mammalian oocytes have only recently begun to emerge. Polarization in mammals is not as dramatic morphologically as in low-order animals. Nevertheless, it has been described at the molecular level in human oocytes.38–40 Non-uniform allocation of protein products encoded by the leptin and Stat3 genes could play a critical role in axis determination in the oocyte as well as in the differentiation of the embryo.38 Their ultimate localization to trophectodermal cells has prompted suggestions for their use as trophectodermal markers. Members of several classes of proteins serving a variety of functions have also been found in polarized domains in human oocytes and embryos.39 These include growth factors (TGF-␤2 and VEGF), growth factor receptors (C-ERBB and C-KIT), and apopotosis proteins (BCL-X and BAX). Asymmetric transcript distributions of ␤-human chorionic gonadotropin and ␤-LH amongst blastomeres in cleaving embryos have also been reported.36,40 Perturbations of the spatial aspects and/or concentration gradients in the oocyte could adversely affect its quality and developmental potential.41 Several structural features of eukaryotic mRNA can affect the stability and may be responsible for early embryonic message degradation and turnover. These include polyadenylation and the formation of secondary structures. The association of mRNAs with cytoplasmic factors may also result in the formation of complexes which may confer stability to the message, or may promote its degradation. 

HPE_Chapter21.qxp

7/13/2007

5:31 PM

Page 266


In mammals, altered polyadenylation is critical in controlling the translation of the oocyte’s mRNA,42 the efficiency of which is regulated by the 3´ untranslated region (UTR).43 Indeed, it has been demonstrated that this is a fairly generalized phenomenon in the mammalian oocyte and embryo.44 Messages with shortened poly(A) tails are repressed until meiotic maturation and fertilization. At the appropriate time, an inversion occurs and those messages that had been actively translated lose their poly(A) tails and cease to be expressed, whereas the messages that had been stored in a quiescent form become functional.45 Two signals in the UTR are required to regulate poly(A) length, the hexanucleotide AAUAAA and the sequence known as the cytoplasmic polyadenylation element (CPE).46,47 A variety of proteins may bind the CPE (CPE binding proteins, CPEB) and thus exert translational control of specific messages throughout development.48,49 A human gene encoding CPEB protein has been identified, and its expression in immature oocytes has been confirmed.50 This human gene, by extension, may be involved in modulating expression throughout early embryonic development. The advent of microarray techniques has allowed the human oocyte to be even more extensively analyzed.51 The study of Bermudez et al51 identified a dataset of 1361 transcripts expressed in common among all of the human oocytes examined, and 406 of these were independently confirmed by other work. The detection of these commonly expressed genes suggests that specific physiological pathways may be functional in human oocytes. Conversely, 7432 transcripts were not found to be present in oocytes, which could be indicative of pathways that are not active during oogenesis. Alternately, the genes may not have been activated due to patient-specific responses to ovarian stimulation protocols, diverse in vitro conditions, and selection criteria among study samples. More recently, these investigators determined by microarray analysis of oocytes from younger versus older patients that several major functional categories of genes are influenced by maternal age including genes involved in cell cycle regulation, energy pathways, stress responses, cytoskeletal structure and transcription.108 A comparison of gene 

expression in immature and mature human oocytes using microarray techniques revealed 52 genes with progressively increasing expression during oocyte maturation.52 In addition, genes were found to be differentially expressed in cumulus cells versus oocytes. In particular, cumulus cells were found to over-express cell-to-cell signaling genes, numerous complement components, semaphorins, and CD antigen genes. Likewise, another study found evidence of differential expression between human oocytes and embryos.53 Transcripts detected by these genome-wide studies included those involved in apoptosis (Bcl), cell cycle regulation (cyclins, cyclindependent kinases, and checkpoint components), signaling pathways (GTPases), and enzyme activation (casein, calmodulin, and serine/threonine kinases), demonstrating that many conserved regulators of basic cellular processes are active at early stages of oocyte maturation. Bioinformatics tools provide a means of carefully scrutinizing datasets obtained by microarray methods, and this may yield valuable information regarding processes governing oocyte maturation; in turn, this may lead to improvements in in vitro methods and diagnostics.

CHECKPOINTS DURING MEIOSIS: SPINDLE AND DNA DAMAGE

Following germinal vesicle breakdown (GVBD), maintaining the orderly segregation of chromosomes during meiosis is critical for oocyte health and viability. The signal transduction machinery which regulates the resumption of meiosis during oocyte maturation must remain in close communication with defect sensing systems in the oocyte at all times, otherwise errors in chromosome distribution may result. However, the very existence of these surveillance mechanisms in the oocyte has been a matter of controversy. A multitude of genetic and biochemical experiments have identified a signal transduction cascade that is activated in response to spindle damage, and this halts further cell cycle progression.54–60 The spindle assembly checkpoint ensures that chromosomes are accurately segregated by preventing

HPE_Chapter21.qxp

7/13/2007

5:31 PM

Page 267


anaphase onset until bipolar attachment to the spindle is achieved.61 Several human genes encoding kinetochore-associated proteins were initially identified as members of this regulatory apparatus, including mitotic arrest deficient (Mad) and budding uninhibited by benomyl (Bub).57,62,63 In the absence of appropriate kinetochore–microtubule interactions, the spindle checkpoint is activated, resulting in cell cycle arrest. Checkpoint gene products are believed to broadcast a wait signal from the unattached kinetochore. The checkpoint proteins could then potentially become modified and released to act as soluble inhibitors.58,64 The primary target of this inhibitory signal is the anaphase promoting complex (APC), the enzyme responsible for the regulated proteolysis of cyclins.54–56 However, once proper chromosomal alignment is achieved, the checkpoint is quenched, bringing about the initiation of anaphase.57,60,62,63 Conflicting results have been reported concerning checkpoint mechanisms in oocytes, arguing both for65–67 and against68–71 their existence. The high rate of chromosomal aberrations routinely observed during preimplantation development would certainly tend to support the latter conclusion.72 However, these contradictory findings could possibly be explained by surveillance systems in oocytes that are more permissive than those present in somatic cells. Thus it would follow that oocytes may be more susceptible to cell cycle error, environmental assault, or cytoplasmic manipulations. Additional studies have bolstered the argument that checkpoint surveillance is indeed functional in mammalian oocytes.73,74 These investigators confirmed the expression of checkpoint proteins during the metaphase-anaphase transition and demonstrated that the mouse oocyte responded to spindle aberrations in a checkpoint competent manner. Compelling evidence supporting checkpoint activity during meiosis came from work in yeast.75,76 These researchers hypothesized that an age-dependent loss of the spindle checkpoint may be the causative factor in the occurrence of birth defects such as Down syndrome. This premise led to the partly successful attempt to rescue aging-associated meiotic chromosome misalignment in senescence-accelerated mouse

oocytes by nuclear transplantation to the ooplasm of young mice of another strain.77,78 Their findings indicate a potential cytoplasmic deficiency in the senescence-accelerated mouse oocyte, including extensive mitochondrial dysfunction, which may impair spindle checkpoint function. Likewise, a possible link between the incidence of age-related aneuploidy and the decline in spindle checkpoint function led to the quantification of checkpoint gene expression during oocyte maturation.79 The results of these experiments appear to indicate that the steady-state levels of these transcripts within stages of maturation decrease as the oocyte ages, perhaps due to degradation over time. Microarray analysis by these investigators confirmed that the expression of spindle checkpoint genes is affected by maternal age.108 Another study also revealed that postovulatory aging results in a quantitative decrease in Mad2 transcripts and an increased incidence of premature centromere separation (PCS), leading to a predisposition to aneuploidy.59 Considerable data have shown that PCS represents a predisposition to chromosome missegration resulting in aneuploidy. All in all, these studies substantiate the view that oocytes do possess a surveillance system, albeit not necessarily a robust one. Securin is another key substrate of the APC, a target of checkpoint regulation.55 In addition to cyclin degradation, the proteolysis of securin must occur to allow further progression through the cell cycle. Securin regulates the protein separase, which in turn mediates cleavage of the chromosomal cohesion complex. (Separase becomes active only following securin destruction.) Sister chromatids are held together by the cohesion protein complex; when chiasmata or crossover sites form between homologous chromosomes, cohesion is instrumental in keeping the paired homologs together long enough for them to orient properly with respect to the meiotic spindle. Consequently, cohesions are critical in ensuring the accurate segregation of chromosomes and averting the production of aneuploid oocytes. The mammalian cohesin complex formed during meiosis has not been fully characterized.80 However, it is believed to include meiosis-specific variants of their mitotic counterparts, encoded by Smc1␤ 

HPE_Chapter21.qxp

7/13/2007

5:31 PM

Page 268


(structural maintenance of chromosome),81 Rec8,82 and Stag3 (stromal antigen 3).82 SMC1␤ is a cohesin subunit that is concentrated at the arms and centromeric regions of chromosomes during meiosis I. It remains associated with the centromeric regions through metaphase II. REC8 and STAG3 co-localize in fibers between sister chromatids which eventually associate with synaptonemal complex, the proteinaceous structure that forms between synapsed chromosomes during prophase I. Rec8 is believed to promote sister chromatid cohesion during meiosis I and II in a variety of species. It is also thought to be involved in maintaining centromeric cohesion. However, its precise role in humans has yet to be described. Figure 21.2 provides an illustration of selected genes involved with cell cycle regulation in oocytes. A recent study by Hodges et al83 presented compelling evidence for the involvement of cohesins

DNA damage Spindle checkpoint checkpoint (Mad, Bub) (Atm, Tp53, Brca1) MPF (Cyclin, Cdk1/Cdc2)

APC Securin Separase

CSF (c-mos, Cdk2)

Cohesin complex (Smc1b, Rec8, Stag3)

Anaphase onset

Figure 21.2 Diagram of selected genes thought to be involved in cell cycle regulation in oocytes. The anaphase promoting complex (APC) is responsible for the proteolysis of cyclin (a component of maturation promoting factor, MPF) and securin (which is bound to and regulates separase). Securin degradation activates separase, resulting in the cleavage of the chromosomal cohesion complex and triggering the onset of anaphase. The activity of the APC is inhibited by the spindle checkpoint which is activated in response to the improper attachment of the chromosomes on the spindle. The cell cycle can also be halted in response to DNA damage. The oocyte is arrested at metaphase II at ovulation due to the action of cytostatic factor (CSF).



in the incidence of age-related aneuploidy. These investigators used a mouse model with a mutated Smc1␤ gene to examine the role of this cohesin subunit in oogenesis. Coincidentally, they observed an age-dependent increase in the percentage of univalent or unpaired chromosomes and single chromatids, as well as a reduced number of chiasmata per oocyte in SMC1␤ deficient mice. Thus, they postulated that the cohesin complex containing SMC1␤ protein is required for the maintenance of cohesion. Their findings support the hypothesis that a progressive loss of sister chromatid cohesion may underlie the age-related increase in aneuploidy observed in human female meiosis. Double-stranded DNA breaks (DSBs) precede recombination initiation, which is necessary to ensure faithful chromosome segregation during meiosis I in yeast. DSBs also probably promote meiotic recombination in mammals.84 DSB repair genes mediate recombination repair events, with the creation of recombination intermediates. Meiosisspecific MutS homologs such as Msh4 and Msh5 are believed to be involved in this process. It has been hypothesized that the products of these genes process nascent recombination events by stabilizing recombination intermediates.85 MutL homologs, Mlh1 and Mlh3, act downstream of these genes by resolving these intermediates into crossovers.86–88 Atm (ataxia-telangiectasia-mutant) encodes a protein kinase which is also believed to be involved in DSB repair and cell cycle control.89 Its expression in human oocytes and embryos has been examined, and was found to be increased in fragmented embryos.90 It should be noted that differences have been observed between male and female gametes with regards to faulty DSB repair during recombination. Oocytes tend to be more tolerant of failed progression or completion of recombination than sperm.91

GENE EXPRESSION DURING PREIMPLANTATION DEVELOPMENT

The zygote relies on maternally derived messages to direct the initial cleavage divisions, until its genome

HPE_Chapter21.qxp

7/13/2007

5:31 PM

Page 269


is activated in later cleavage stages.92,93 The transition between the dependence on maternally inherited transcripts and the initiation of embryonic gene expression does not appear to be absolute, as maternal RNAs and their products persist in advanced phases of preimplantation development.94 Nevertheless, the switch from maternal to embryonic genome control can be characterized by the selective degradation of maternally derived macromolecules and by the initiation of embryonic gene expression, as well as by a shift in protein synthesis patterns.92,95 By postponing translation of stored maternal transcripts, the temporal coordination of gene expression required for mammalian oocyte– embryo transition can be achieved. A group of genes collectively referred to as maternal effect genes are of particular interest. These genes encode factors that are stored in the oocyte, and are required for proper embryonic development. A variety of maternal effect genes have been identified in insects, nematodes, and amphibians, but relatively few have been identified in mammals. Zygote arrest 1 (ZAR1) is an ovaryspecific maternal factor that functions during the oocyte–embryo transition, and is necessary for the completion of fertilization.96 This gene displays an evolutionarily conserved expression pattern, and its protein exhibits a high degree of structural homology among divergent vertebrate species, including human, suggesting that it plays a critical role in development. MATER (maternal antigen that embryos require), a maternal effect gene expressed solely in the oocyte, was initially identified as an autoantigen in a mouse model of autoimmune premature ovarian failure.97 It is believed to be involved in embryonic genome activation due to the developmental arrest noted in 2-cell mouse embryos lacking MATER.98 A homolog of Mater has been identified in human.99 Several other maternal effect genes with human homologs have been identified, all of which are believed to be master controllers in development (Table 21.1). Interest in human embryonic gene expression has intensified in recent years, as the search for accurate clinical tools that can be used to predict embryo viability and health continues. For years,

Table 1.1 Maternal effect genes with human homologs Gene symbol

Gene name

Reference

Zar1 Mater

zygote arrest 1 maternal antigen that embryos require Spindlin Formin-2

Wu et al, 200396 Tong and Nelson, 1999;97 Tong et al, 200299

Spin Fmn2 Hsf1 Dnmt Nom2

Oh et al, 1997109 Ryley et al, 2005;110 Leader et al, 2002111 heat shock factor-1 Christians et al, 2000;112 Rallu et al, 1997113 DNA methytransferases Hamatani et al, 2004114 Nucleoplasmin 2 Burns et al, 2003115

Maternal effect genes encode RNAs and proteins which are stored in the oocyte for later use. The genes listed above all have human homologues and are believed to influence early developmental processes.

clinicians have relied on morphological criteria as a primary basis for embryo selection. Typically, the embryo has been evaluated by assessing the number of cells present and the degree of fragmentation, However, a variety of additional parameters have been examined, including the presence of cytoplasmic inclusions, refractile bodies and/or vacuoles, degree of granularity, symmetry of the blastomeres, and multinucleation. Although careful morphological scrutiny of the embryo has met with some success, there continues to be a significant degree of embryonic wastage in in vitro fertilization. As a result, embryologists have resorted to the transfer of multiple embryos, risking the incidence of high-order births in order to increase pregnancy rates. Extended culture of human embryos to the blastocyst stage has also been used as a means of providing a longer period of evaluation and selection.100 This is an intuitively attractive approach, since it allows embryos to be transferred at a stage that is in synchrony with the recipient uterus. However, development in culture to blastocyst stage does not guarantee that the embryo is genetically normal.101 Furthermore, extended culture places the embryo at increased risk of exposure to potentially suboptimal conditions for a longer period of time. Chromosomal screening through preimplantation genetic diagnostic (PGD) methods has been used successfully for years to identify embryos harboring 

HPE_Chapter21.qxp

7/13/2007

5:31 PM

Page 270


lethal genetic anomalies.102 However PGD has its limitations, since it can only detect gross chromosomal aberrations. A technique capable of detecting subtle yet critical irregularities is still lacking. The search for novel diagnostic tools prompted the analysis of gene expression and its relationship to abnormal morphology in embryos.90 Wells et al examined the expression of multiple genes in 50 embryos at various stages of preimplantation development. Using cluster analysis, the embryos were sorted into distinct groups consistent with their stage of development. However, subgroups were identified which contained embryos displaying abnormal features. The embryos were also grouped according to their abnormal characteristics in order to determine if a specific abnormality was associated with any particular expression pattern, and a correlation was found between increased fragmentation and altered expression of several genes. The strongest correlation found was between fragmentation and the expression of Tp53 (tumor protein p53), a transcription factor involved in cell cycle regulation, programmed cell death (apoptosis), and DNA repair. Similarly, a relationship between the expression of Brca1 (breast cancer 1, early onset) gene and fragmentation was also revealed. Like TP53, Brca1 is also involved in the regulation of the cell cycle and repair of damaged DNA. Finally, as previously mentioned, an increase in Atm (ataxia telangiectasia mutated) expression was observed in embryos with moderate amounts of fragmentation, providing further evidence for a role for DNA damage in the etiology of this morphological characteristic. Atm is believed to play a central role in the control of cell cycle checkpoint signaling pathways that are required for genome stability and for cellular response to DNA damage. The search for clinically relevant molecular markers has prompted more comprehensive analyses of gene expression in human embryos. A recent study by Dobson et al103 revealed that microarray analysis could clearly identify stage-specific patterns of gene expression in human embryos. Wells et al104 confirmed this finding by real-time PCR analysis of selected genes. Both studies reported a sharp decrease in gene expression postfertilization, followed 

by an increase after the presumptive activation of the embryonic genome. However, Dobson’s group found that even arrested embryos displayed expression patterns consistent with their day of development, as long as they had cleaved at least once. Surprisingly, these investigators observed comparable levels of expression for markers of normal embryonic genome activation between normal and arrested 2-, 3- and 4-cell embryos. They concluded that embryonic arrest in human embryos may not be linked to a failure in global activation of the genome, as is the case in other mammalian species.96,99,105 Dobson et al103 instead hypothesized that the loss of embryos without gross chromosomal abnormalities may be attributed to the aberrant expression of individual genes. Following embryonic genome activation, the cleavage stage embryo becomes compacted and the blastomeres are no longer distinct, giving rise to the morula. Shortly thereafter, mammalian embryos enter a key developmental phase characterized morphologically by the formation of two distinct cell lineages, the inner cell mass (ICM) and the trophectoderm (TE) of the blastocyst. Differential gene expression is believed to be responsible for these morphological transitions. However, the precise nature of these stages has yet to be defined at the molecular level. A recent study examining global expression patterns has been undertaken to expose the molecular cues that signal these transformations in human embryos.106 These researchers detected ICM specific (i.e. Oct4/Pou5F1, Nanog, Hmgb1 and Dppa5) and TE specific (i.e. Cdx2, Atp1b3, Sfn and Ipl) marker transcripts, and they hypothesized that complex metabolic and signaling pathways were responsible for the emergence of ICM and TE cell lineages. Another study analyzed the proteome of individual human blastocysts using mass spectrometry.107 Consistent with the previous studies that examined gene expression patterns in embryos, these investigators observed differential protein expression profiles corresponding to the stage of development and morphology. Additional work will be necessary to permit the thorough molecular dissection of the pathways responsible for blastocyst formation.

HPE_Chapter21.qxp

7/13/2007

5:31 PM

Page 271


CONCLUDING REMARKS

All of the studies reviewed here have produced an immense repository of information, enabling massive parallel mining of biological data. Further inspection of the results may form the basis for selective analysis, using more traditional hypothesis-driven approaches. In the process, it is anticipated that a better explanation of how regulatory networks are activated throughout human development may be obtained. Likewise, insight into these physiological pathways may provide new tools for use in clinical diagnostics. The complex interplay between the developing oocyte and its surroundings is only now beginning to be exposed, and a few key players during human embryo development have been uncovered. Many genes have been identified that are involved in folliculogenesis, oocyte maturation, and early preimplantation development. Undoubtedly, many more remain to be discovered. A better understanding of these processes could lead to further advances in the management and treatment of human infertility. REFERENCES 1. Davidson EH, Hough BR. High sequence diversity in the RNS synthesized at the lampbrush stage of oogenesis. Proc Natl Acad Sci USA 1969; 63: 342–9. 2. Thomas PS. Hybridization of denatured RNA and small DNA fragments transferred to nitrocellulose. Proc Natl Acad Sci USA 1980; 77: 5201–5. 3. White BA, Bancroft FC. Cytoplasmic dot hybridization. J Biol Chem 1982; 257: 8569–72. 4. Rappolee DA, Brenner CA, Schultz R et al. Developmental expression of PDGF, TGF-alpha, and TGF-beta genes in preimplantation mouse embryos. Science 1988; 241: 1823–5. 5. Steuerwald N, Cohen J, Herrera RJ et al. Analysis of gene expression in single oocytes and embryos by real-time rapid cycle fluorescence monitored RT-PCR. Mol Hum Reprod 1999; 5: 1034–9. 6. Steuerwald N, Barritt JA, Adler R et al. Quantification of mtDNA in single oocytes. Zygote 2000; 8: 209–15. 7. Goto T, Jones GM, Lolatgis N et al. Identification and characterization of known and novel transcripts expressed during the final stages of human oocyte maturation. Mol Reprod Dev 2002; 2002: 13–28. 8. Metcalfe AD, Bloor DJ, Lieberman BA et al. Amplification of representative cDNA pools from single human oocytes and pronucleate embryos. Mol Reprod Dev 2003; 65: 1–8. 9. Nielson L, Andalibi A, Kang D et al. Molecular phenotype of the human oocyte by PCR–SAGE. Genomics 2000; 63: 13–24.

10. Adashi EY, Rock JA, Rosenwaks Z. Reproductive Endocrinology: Surgery and Technology. Philadelphia: Lippincott-Raven Publishers, 1996: 18–40. 11. Serafica MD, Goto T, Trounson AO. Transcripts from a human primordial follicle cDNA library. Hum Reprod 2005; 20: 2074–91. 12. Hussein MR. Apoptosis in the ovary: molecular mechanisms. Hum Reprod Update 2005; 11: 162–78. 13. Skinner MK. Regulation of primordial follicle assembly and development. Hum Reprod Update 2005; 11: 461–71. 14. Hutt KJ, McLaughlin EA, Holland MK. Kit ligand and c-Kit have diverse roles during mammalian oogenesis and folliculogenesis. Mol Hum Reprod 2006; 12: 61–69. 15. Huntriss J, Gosden R, Hinkins M et al. Isolation, characterization and expression of the human Factor In the Germline alpha (FIGLA) gene in ovarian follicles and oocytes. Mol Hum Reprod 2002; 8: 1087–95. 16. Bayne RAL, Martins da Silva SJ, Anderson RA. Increased expression of the FIGLA transcription factor is associated with primordial follicle formation in the human fetal ovary. Mol Hum Reprod 2004; 10: 373–81. 17. Liang L, Soyal SM, Dean J. FIG␣, a germ cell specific transcription factor involved in the coordinate expression of the zona pellucida genes. Development 1997; 124: 4939–47. 18. Conner SJ, Lefièvre L, Hughes DC et al. Cracking the egg: increased complexity in the zona pellucida. Hum Reprod 2005; 20: 1148–52. 19. Brekhman V, Itskovitz-Eldor J, Yodko E et al. The DAZL1 gene is expressed in human male and female embryonic gonads before meiosis. Mol Hum Reprod 2000; 6: 465–8. 20. Aaltonen J, Laitinen MP, Vuojolainen K. Human growth differentiation factor 9 (GDF-9) and its novel homolog GDF-9B are expressed in oocytes during early folliculogenesis. J Clin Endocrinol Metab 1999; 84: 2744–50. 21. Dube JL, Wang P, Elvin J et al. The bone morphogenetic protein 15 gene is X-linked and expressed in oocytes molecular. Endocrinology 1998; 12: 1809–17. 22. Pangas SA, Matzuk MM. The art and artifact of GDF9 activity: cumulus expansion and the cumulus expansion-enabling factor. Biol Reprod 2005; 73: 582–5. 23. McKenzie LJ, Pangas SA, Carson SA. Human cumulus granulosa cell gene expression: a predictor of fertilization and embryo selection in women undergoing IVF. Hum Reprod 2004; 19: 2869–74. 24. Pasquale ED, Rossetti R, Marozzi A et al. Identification of new variants of human BMP15 gene in a large cohort of women with premature ovarian failure. J Clin Endocrinol Metab 2006; 10: 2005–650. 25. Yamamoto S, Konishi I, Komatsu NK et al. Immunohistochemical localization of basic fibroblast growth factor (bFGF) during folliculogenesis in the human ovary. Gynecol Endocrinol 1997; 11: 223–30. 26. Ben-Haroush A, Abir R, Ao A et al. Expression of basic fibroblast growth factor and its receptors in human ovarian follicles from adults and fetuses. Fertil Steril 2005; 84: 1257–68. 27. Rajkovic A, Pangas SA, Ballow D et al. NOBOX deficiency disrupts early folliculogenesis and oocyte-specific gene expression. Science 2004; 20: 1157–9. 28. Suzumoria N, Yana C, Matzuk M et al. Nobox is a homeobox-encoding gene preferentially expressed in primordial and growing oocytes. Mech Dev 2002; 111: 137–41. 29. Schultz RM, Montgomery RR, Belanoff JR. Regulation of mouse oocyte maturation: implications of a decrease in oocyte cAMP and protein dephosphorylation in commitment to resume meiosis. Dev Biol 1983; 97: 264-73. 30. Smith LD, Ecker RE. Role of the oocyte nucleus in the physiological maturation in Rana pipiens. Dev Biol 1969; 19: 281–309.



HPE_Chapter21.qxp

7/13/2007

5:31 PM

Page 272


31. Masui Y, Markert CL. Cytoplasmic control of nuclear behavior during meiotic maturation of frog oocytes. J Exp Zool 1971; 177: 129–46. 32. Dorée M, Galas S. The cyclin-dependent protein kinases and the control of cell division. FASEB J 1994; 8: 1114–21. 33. Masui Y. A cytostatic factor in amphibians: its extraction and partial characterization. J Exp Zool 1974; 187: 141–7. 34. Paules RS, Buccione R, Moscel RC et al. Mouse mos protoncogene product is present and functions during oogenesis. Proc Natl Acad Sci USA 1989; 86: 5395–9. 35. Chiapa B, Chiri S. Egg activation; upstream of the fertilization calcium signal. Biol Cell 2000; 92: 215–33. 36. Watanabe N, Vande Woude GF, Ikawa Y et al. Specific proteolysis of the c-mos proto-oncogene product by calpain on fertilization of Xenopus eggs. Nature 1989; 342: 505–17. 37. Edwards RG. Ovarian differentiation and human embryo quality. 1. Molecular and morphogenetic homologies between oocytes and embryos in Drosophila, C. elegans, Xenopus and mammals. Reprod Biomed Online 2001; 3: 138-60. 38. Antczak M, Van Blerkom J. Oocyte influences on early development: the regulatory proteins leptin and STAT3 are polarized in mouse and human oocytes and differentially distributed within the cells of the preimplantation stage embryo. Mol Hum Reprod 1997; 3: 1067–86. 39. Antczak M, Van Blerkom J. Temporal and spatial aspects of fragmentation in early human embryos: possible effects on developmental competence and association with the differential elimination of regulatory proteins from polarized domains. Hum Reprod 1999; 14: 429–47. 40. Hansis C, Edwards RG. Cell differentiation in the preimplantation human embryo. Reprod Biomed Online 2002; 6: 215–20. 41. Eichenlaub-Ritter U. Manipulation of the oocyte: possible damage to the spindle apparatus. Reprod Biomed Online 2002; 5: 117–24. 42. Richter JD. Cytoplasmic polyadenylation in development and beyond. Microbiol Mol Biol Rev 1999; 63: 446–56. 43. Trudel M, Constantini F. A 3´ enhancer contributes to the stagespecific expression of the human β-globin gene. Genes Dev 1987; 1: 954–61. 44. Oh B, Hwang SY, McLaughlin J et al. Timely translation during the mouse oocyte-to-embryo transition. Development 2000; 127: 3795–803. 45. Vassalli JD, Huarte J, Belin D et al. Regulated polyadenylation controls mRNA translation during meiotic maturations of mouse oocytes. Genes Dev 1989; 3: 2163-71. 46. Bachvarova RF. A maternal tail of poly(A): the long and the short of it. Cell 1992; 69: 895–7. 47. Huarte et al. 1992. 48. Gebauer F, Richter JD. Mouse cytoplasmic polyadenylation element binding protein: an evolutionarily conserved protein that interacts with the cytoplasmic polyadenylation elements of c-Mos mRNA. Proc Natl Acad Sci USA 1996; 93: 14602–7. 49. Stutz A, Conne B, Huarte J et al. Masking, unmasking, and regulated polyadenylation cooperate in the translational control of a dormant mRNA in mouse oocytes. Genes Dev 1998; 12: 2535–48. 50. Welk JF, Charleswortha A, Smith GD et al. Identification and characterization of the gene encoding human cytoplasmic polyadenylation element binding protein. Gene 2001; 263: 113–20. 51. Bermudez MG, Wells D, Malter H et al. Expression profiles of individual human oocytes using microarray technology. Reprod Biomed Online 2004; 8: 325–37. 52. Assou S, Anahory T, Pantesco V. The human cumulus–oocyte complex gene-expression profile. Hum Reprod 2006; 21: 1705–19.



53. Li SS, Liu Y, Tseng C et al. Analysis of gene expression in single human oocytes and preimplantation embryos. Biochem Biophys Res Commun 2006; 340: 48–53. 54. Cleveland DW, Mao Y, Sullivan KF. Centromeres and kinetochores: from epigenetics to mitotic checkpoint signaling. Cell 2003; 112: 407–21. 55. Hagting A, den Elzen N, Vodermaier HC et al. Human securin proteolysis is controlled by the spindle checkpoint and reveals when the APC/C switches from activation by Cdc20 to Cdh1. J Cell Biol 2002; 157: 1125–37. 56. Li Y, Gorbea C, Mahaffey D et al. MAD2 associates with the cyclosome/anaphase-promoting complex and inhibits its activity. Proc Natl Acad Sci USA 1997; 94: 12431–36. 57. Ouyang B, Lan Z, Meadows J et al. Human Bub1: a putative spindle checkpoint kinase closely linked to cell proliferation. Cell Growth Differ 1998; 9: 877–85. 58. Rieder CL, Cole RW, Khodjakov A. et al. The checkpoint delaying anaphase in response to chromosome monoorientation is mediated by an inhibitory signal produced by unattached kinetochores. J Cell Biol 1995; 130: 941–48. 59. Steuerwald N. Meiotic spindle checkpoints for assessment of aneuploid oocytes. Cytogenet Genome Res 2005; 11: 623–30. 60. Zhang D, Ma W, Li YH et al. Intra-oocyte localization of MAD2 and its relationship with kinetochores, microtubules, and chromosomes in rat oocytes during meiosis. Biol Reprod 2004; 71: 740–48. 61. Hartwell LH, Weinert TA. Checkpoints: controls that ensure the order of cell cycle events. Science 1989; 246: 629–34. 62. Li Y, Benezra R. Identification of human mitotic checkpoint gene: hsMAD2. Science 1996; 274: 246–8. 63. Jablonski SA, Chan GKT, Cooke CA et al. The hBUB1 and hBUBR1 kinases sequentially assemble onto kinetochores during prophase with hBUBR1 concentrating at the kinetochore plates in mitosis. Chromosoma 1998; 107: 386–96. 64. Hardwick KG. The spindle checkpoint. Trends Genet 1998; 14: 1–4. 65. Eichenlaub-Ritter U, Betzendahl I. Chloral hydrate induced spindle aberrations, metaphase I arrest and aneuploidy in mouse oocytes. Mutagenesis 1995; 10: 477–86. 66. Fulka J Jr, Moor RM, Fulka J. Sister chromatid separation and the metaphase-anaphase transition in mouse oocytes. Dev Biol 1994; 165: 410–7. 67. Fulka J Jr, Moor RM, Fulka J. Mouse oocyte maturation: meiotic checkpoints. Exp Cell Res 1995; 219: 414–19. 68. Fulka J Jr, Moor RM, Fulka J. Replicating DNA does not block germinal vesicle breakdown in mouse oocytes. J Exp Zool1995; 272: 245–8. 69. Fulka J Jr, Kalab P, First NL et al. Damaged chromatin does not prevent the exit from metaphase I in fused mouse oocytes. Hum Reprod 1997; 12: 2473–6. 70. LeMaire-Adkins R, Radke K, Hunt PA. Lack of checkpoint control at the metaphase/anaphase transition: a mechanism of meiotic nondisjunction in mammalian females. J Cell Biol 1997; 139: 1611-9. 71. Yin H, Cukurcam S, Betzendahl I et al. Trichlorfon exposure, spindle aberrations and nondisjunction in mammalian oocytes. Chromosoma 1998; 107: 514–22. 72. Handyside AH, Delhanty JDA. Preimplantation genetic diagnosis: strategies and surprises. Trends Genet 1997; 13: 270–75. 73. Brunet S, Pahlavan G, Taylor S et al. Functionality of the spindle checkpoint during the first meiotic division of mammalian oocytes. Reproduction 2003; 126: 443–50. 74. Wassmann K, Niault T, Maro B. Metaphase 1 arrest upon activation of the Madz-dependent spindle checkpoint in mouse oocytes. Curr Biol 2003; 13: 1596–1608.

HPE_Chapter21.qxp

7/13/2007

5:31 PM

Page 273


75. Shonn MA, McCarroll R, Murray AW. Requirement of the spindle checkpoint for proper chromosome segregation in budding yeast meiosis. Science 2000; 289: 300–3. 76. Sluder G, McCollum D. The mad ways of meiosis. Science 2000; 289: 254–6. 77. Liu L, Keefe DL. Ageing–associated aberration 1 meiosis of oocytes from senescence–accelerated mice. Hum Reprod 2002; 17: 2678–85. 78. Liu L, Keefe DL. Nuclear origin of aging-associated meiotic defects in senescence-accelerated mice. Biol Reprod 2004; 71: 1724–9. 79. Steuerwald N, Cohen J, Herrera R et al. Association between spindle assembly checkpoint expression and maternal age in human oocytes. Mol Hum Reprod 2001; 7: 49–55. 80. Revenkova E, Jessberger R. Keeping sister chromatids together: cohesins in meiosis. Reproduction 2005; 130: 783–90. 81. Revenkova E, Eijpe M, Heyting C et al. Cohesin SMC1b is required for meiotic chromosome dynamics, sister chromatid cohesion and DNA recombination. Nat Cell Biol 2004; 6: 555-62. 82. Prieto I, Tease C, Pezzi N et al. Cohesin component dynamics during meiotic prophase I in mammalian oocytes Chromosome Res 2004; 12: 197–213. 83. Hodges CA, Revenkova E, Jessberger R et al. SMC1␤-deficient female mice provide evidence that cohesins are a missing link in age-related nondisjunction. Nat Genet 2005; 37: 1351–5. 84. Di Giacomo M, Barchi M, Baudat F et al. Distinct DNA-damagedependent and -independent responses drive the loss of oocytes in recombination-defective mouse mutants Proc Natl Acad Sci USA 2005; 102: 737–42. 85. Snowden T, Acharya S, Butz C et al. hMSH4-hMSH5 recognizes holliday junctions and forms a meiosis-specific sliding clamp that embraces homologous chromosomes. Mol Cell 2004; 15: 437–51. 86. Lenzi ML, Smith J, Snowden T et al. Extreme heterogeneity in the molecular events leading to the establishment of chiasmata during meiosis I in human oocytes. Am J Hum Genet 2005; 76: 112–27. 87. Roig I, Liebe B, Egozcue J et al. Female-specific features of recombinational double-stranded DNA repair in relation to synapsis and telomere dynamics in human oocytes. Chromosoma 2004; 113: 22–33. 88. Tease C, Hartshorne GM, Hultén A. Patterns of meiotic recombination in human fetal oocytes. Am J Hum Genet 2002; 70: 1469–79. 89. Lavin MF, Birrell G, Chen P et al. ATM signaling and genomic stability in response to DNA damage. Mut Res 2005; 569; 123–32. 90. Wells D, Bermúdez MG, Steuerwald N et al. Association of abnormal morphology and altered gene expression in human preimplantation embryos. Fertil Steril 2005; 84: 343–55. 91. Hunt P, Hassold TJ. Sex matters in meiosis. Science 2002; 21: 2181–3. 92. Telford N, Watson A, Schultz G. Transition from maternal to embryonic control in early mammalian development: a comparison of several species. Mol Reprod Dev 1990; 26: 90–100. 93. Braude P, Bolton V, Moore S. Human gene expression first occurs between the four- and eight-cell stages of preimplantation development. Nature 1988; 332: 459–61. 94. Nothias JY, Majumder S, Kaneko KJ et al. Regulation of gene expression at the beginning of mammalian development. J Biol Chem 1995; 270: 22077–80. 95. Wang Q, Latham KE. Translation of maternal messenger ribonucleic acids encoding transcription factors during genome activation in early mouse embryos. Biol Reprod 2000; 62: 969–78. 96. Wu X, Wang P, Brown CA et al. Zygote arrest 1 (Zar1) is an evolutionarily conserved gene expressed in vertebrate ovaries. Biol Reprod 2003; 69: 861–7.

97. Tong ZB, Nelson LM. A mouse gene encoding an oocyte antigen associated with autoimmune premature ovarian failure. Endocrinology 1999; 140: 3720–6. 98. Tong ZB, Gold L, De Pol A et al. Developmental expression and subcellular localization of mouse MATER, an oocyte-specific protein essential for early development. Endocrinology 2004; 145: 1427–34. 99. Tong ZB, Bondy CA, Zhou J et al. A human homologue of mouse Mater, a maternal effect gene essential for early embryonic development. Hum Reprod 2002; 17: 903–11. 100. Gardner DK, Surrey E, Minjarez D et al. Single blastocyst transfer: a prospective randomized trial. Fertil Steril 2004; 81: 551–5. 101. Sandalinas M, Sadowy S, Alikani M et al. Developmental ability of chromosomally abnormal human embryos to develop to the blastocyst stage. Hum Reprod 2001; 16: 1954–8. 102. Munne S. Preimplantation genetic diagnosis and human implantation—a review. Placenta 2003; 24: S70–6. 103. Dobson AT, Raja R, Abeyta MJ et al. The unique transcriptome through day 3 of human preimplantation development. Hum Mol Genet 2004; 13: 1461–70. 104. Wells D, Bermudez MG, Steuerwald N. Expression of genes regulating chromosome segregation, the cell cycle and apoptosis during human preimplantation development Hum Reprod 2005; 20: 1339–48. 105. Schramm RD, Paprocki AM, VandeVoort CA. Causes of developmental failure of in vitro matured rhesus monkey oocytes: impairments in embryonic genome activation. Hum Reprod 2003; 18: 826–33. 106. Adjaye J, Huntress J, Herwig R et al. Primary differentiation in the human blastocyst: comparative molecular portraits of inner cell mass and trophectoderm cells. Stem Cells 2005; 23: 1514–25. 107. Katz-Jaffe MG, Gardner DK, Schoolcraft WB. Proteomic analysis of individual human embryos to identify novel biomarkers of development and viability. Fertil Steril 2006; 85: 101–7. 108. Steuerwald N, Bermúdez MG, Wells D, Munné S, Cohen J. Maternal age-related differential global expression profiles observed in human oocytes. Reprod BioMed Online 2007 [e-pub ahead of print on 10 April 2007]. 109. Oh B, Hwang SY, Solter D et al. Spindlin, a major maternal transcript expressed in the mouse during the transition from oocyte to embryo. Development 1997; 124: 493–503. 110. Ryley DA, Wu HH, Leader B et al. Characterization and mutation analysis of the human FORMIN-2 (FMN2) gene in women with unexplained infertility. Fertil Steril 2005; 83: 1363–71. 111. Leader B, Lim H, Carabatsos MJ et al. Formin-2, polyploidy, hypofertility and positioning of the meiotic spindle in mouse oocytes. Nat Cell Biol 2002; 4: 921–8. 112. Christians E, Davis AA, Thomas SD et al. Maternal effect of Hsf1 on reproductive success. Nature 2000; 12: 693–4. 113. Rallu M, Loones MT, Lallemand Y et al. Function and regulation of heat shock factor 2 during mouse embryogenesis Proc Natl Acad Sci USA 1997; 94: 2392–97. 114. Hamatani T, Falco G, Carter MG et al. Age-associated alteration of gene expression patterns in mouse oocytes. Hum Mol Genet 2004; 13: 2263–78. 115. Burns KH, Viveiros MM, Ren Y et al. Roles of NPM2 in chromatin and nucleolar organization in oocytes and embryos. Science 2003; 300:633–6.



HPE_Chapter21.qxp

7/13/2007

5:31 PM

Page 274

HPE_Chapter22.qxp

7/13/2007

4:53 PM

Page 275

22. Mitochondria in reproduction: future assays for embryo selection Brian Dale, Loredana Di Matteo and Martin Wilding

INTRODUCTION

Although techniques such as preimplantation genetic diagnosis (PGD) have been used to aid in the selection for transfer of human embryos developing in vitro,1 the technology and expertise required limit their use to specialized laboratories; selection criteria in the routine IVF laboratory are largely based on parameters that the biologist can observe under the light microscope. These parameters include oocyte morphology (position and size of polar body, morphology of metaphase II plate, presence of imperfections in the oocyte cytoplasm),2 zygote morphology (the size and position of pronuclei, presence of a ‘cytoplasmic flare’, and nucleolar morphology),2 the rate of development (number of cell cleavages/day), and morphology of embryos (such as inequalities in cleavage and the presence of cell fragments). These morphological parameters are based on cellular physiology, and research in recent years has revealed many aspects of embryonic physiology that shed light on the basis for these observations.2 Current data suggest that the mitochondrial physiology of preimplantation embryos may play an important role in human preimplantation embryo development, and in determining embryo quality. In this chapter, we discuss the role of mitochondria, and in particular mitochondrial physiology in human embryology. Some aspects of morphological analysis are determined by the morphology of mitochondria within the cytoplasm, and we suggest that assays of mitochondrial activity and analysis of mitochondrial morphology may become important parameters in the analysis of developing human embryos and their selection for transfer into the uterus. Unfortunately, assays for mitochondrial respiration that can be incorporated into the IVF laboratory routine

are not yet available, and we look at future possibilities for mitochondrial assessment in embryos developing in vitro. MITOCHONDRIA

All eukaryotic cells contain mitochondria as intracellular organelles. Mitochondria are now known to have a wide variety of critical physiological roles, including: the triggering of apoptosis; synthesis of pyrimidines, heme, and steroid hormones; redox regulation; protein glycosylation; and metabolism of neurotransmitters, calcium, and iron.3–6 Many of these mitochondrial pathways are tissue specific, and all are essential for human health and development. In yeast, mitochondrial defects prevent sporulation – a complex developmental program that involves cellular co-operation, pattern formation, and differentiation.7 In humans, disorders of mitochondrial function are now known to contribute to maladies such as childhood developmental disorders, seizures, diabetes, hearing loss, blindness, heart disease, kidney disease, liver disease, gastrointestinal dysmotility disorders, antiviral drug toxicity, delayed wound healing, multiorgan system failure during sepsis, cancer, stroke, dementia, and possibly even infertility.6,8–12 A major mitochondrial role is to supply energy to the cell cytoplasm through the production of adenosine trisphosphate (ATP). Mitochondria produce ATP through aerobic respiration; ATP can also be produced by anaerobic respiration in mammalian cells. In the oocyte cytoplasm, the two processes initiate with the conversion of glucose to pyruvate, and the production of ATP in all mammalian cells, including human embryos, probably includes components of both aerobic and anaerobic respiration at any time.13 Aerobic respiration is significantly more efficient than 

HPE_Chapter22.qxp

7/13/2007

4:53 PM

Page 276


anaerobic respiration, producing an energy output (⌬G) of ⫺686 kcal/mol glucose compared with ⌬G of ⫺47 kcal/mol glucose for anaerobic respiration. Therefore, it is logical to suggest that aerobic respiration is the preferred choice when sufficient oxygen is available. Mitochondria are membrane-bound intracellular organelles, and also contain a second inner membrane. The outer mitochondrial membrane is permeable to molecules with molecular weights of up to 10 000 Da.14 The inner mitochondrial membrane is relatively impermeable and surrounds the ‘matrix’. The inner mitochondrial membrane contains most of the enzymes required for the respiratory chain, and has foldings, or ‘cristae’, which give this membrane a greater surface area for the electron transport chain. The number of cristae in a mitochondrion indicates the degree of oxidative phosphorylation taking place, and corresponds to its capacity to produce ATP. In fact, local mitochondrial configurations often represent local energy requirements in the body.15 The mitochondrial matrix also contains a small amount of DNA termed ‘mitochondrial DNA’ (mtDNA), 16 560 kb of double stranded DNA that codes for 13 proteins, 22 transfer RNA (tRNA) species, and two ribosomal RNA (rRNA) species.16–19 Only a small portion of the enzymes required for the respiratory chain are coded for by mtDNA; the majority are derived from the nuclear DNA. The current consensus is that one copy of mtDNA is present for every single mitochondrion in the mammalian oocyte, and this therefore provides a useful means of estimating the density of mitochondria within these cells.20 Mitochondrial DNA is subject to degeneration via the action of free radicals produced within the mitochondrial matrix during oxidative phosphorylation.21,22 It is thought that mtDNA has little in the way of DNA repair or protection mechanisms.23,24 Free radical attack can influence the production of some of the proteins involved in the respiratory chain, and over long periods this could result in reduced efficiency of the respiratory pathway.13,25,26 The respiratory chain consists of a series of enzymes that catalyze a cyclic pathway (‘Krebs’ cycle)27 

in which carbon dioxide and energy are produced from substrates of oxygen and carbon atoms in molecules such as pyruvate and glucose. The energy produced through the respiratory chain is converted to ATP via an indirect pathway, in which H⫹ ions produced by the respiratory chain are expelled from the matrix. This process creates an electrophysiologically negatively charged environment within the matrix, causing the expelled H⫹ ions to be attracted towards the mitochondrial matrix. The H⫹ ions return to the matrix through the action of an ATPase enzyme, thus linking ATP production to the activity of the respiratory chain. Therefore, according to Mitchell’s chemiosmotic hypothesis,28,29 the H⫹ potential (⌬⌿) of the inner mitochondrial membrane is highly correlated with the capacity of individual mitochondria to produce ATP.

MITOCHONDRIA IN MAMMALIAN OOCYTES AND EMBRYOS

The mitochondrial ‘bottleneck’ theory states that the mitochondrial content of primordial oocytes is progressively reduced to a pool of 3–4 copies of mitochondria which then replicates to form the complement of mature oocytes.30,31 Therefore, the relative ‘fitness’ of these original components strongly determines the quality of mitochondria within the mature oocyte. Although some paternal mitochondria may survive after fertilization,32 the majority of spermatozoal mitochondria are either targeted for destruction after fertilization,32 or simply become dysfunctional, so that inheritance of mitochondria is nearly 100% maternal. By measuring the mtDNA content of individual oocytes, it has been estimated that mature oocytes contain between 20 000 and 800 000 mitochondria.26,33,34 Interestingly, oocytes from patients with fertilization failure for unknown causes were often found to contain low copy numbers of mtDNA, suggesting that mtDNA copy number, and therefore number of mitochondria, may be indicative of fertilization potential.33 Mitochondria in human oocytes and preimplantation embryos undergo distinct changes in localization during each phase of development, and are

HPE_Chapter22.qxp

7/13/2007

4:53 PM

Page 277

MITOCHONDRIA IN REPRODUCTION: FUTURE ASSAYS FOR EMBRYO SELECTION

spatially reorganized during mammalian oocyte maturation. In germinal vesicle (GV) stage oocytes, the cytoplasm surrounding the germinal vesicle is very dense (Figure 22.1A), and our data confirm previous reports demonstrating that this clustering represents the localization of mitochondria26,36 (Figure 22.1A and B).After GV breakdown (BD), mitochondria reorganize and tend to distribute more evenly throughout the cytoplasm (Figure 22.1C and D). Some groups have observed a higher density of mitochondria in the region of the metaphase I and II meiotic apparatus,26,36 but this is not always visible (Figures 22.1

and 22.2); we have observed a mitochondria-free region of cytoplasm directly surrounding the meiotic apparatus (Figure 22.2C and D). Human metaphase II oocytes have a wide variety of morphological configurations, and this appears to be reflected in the density and morphology of mitochondria, as visualized under fluorescence microscopy (Figure 22.2). Using a variety of mitochondria specific fluorescence dyes and confocal microscopy, we have observed that not

I II

II

II

A

B

I

I A

B

I A

II

II

I

I

C

C

D

Figure 22.1 Mitochondrial morphology in human immature oocytes. Confocal microscopy images of immature human oocytes donated for research (donations were made prior to 2004, after which legislation forbade research on human embryos)35 loaded with the mitochondrial specific fluorescence dye 5,5',6,6'-tetrachloro-1,1',3,3'-tetraethylbenzimidazolylcarbocyanine iodide (JC-1). (A) and (B) Images of a germinal vesicle (GV) stage human oocyte. (C) and (D) Images of a metaphase I (MI) stage human oocyte. (A) and (C) show the bright-field images of the oocytes (non-confocal) and (B) and (D) show a single confocal section of the mitochondrial membrane potential-insensitive channel (green fluorescence output). ‘I’ shows mitochondrial type I morphology. ‘II’ shows mitochondrial type II morphology (bright-field visualization in (A) and (C'), and fluorescence images in (B) and (D)). White bars represent 20 ␮m.

II II

A II

D

Figure 22.2 Mitochondrial morphology in human mature oocytes. Confocal microscopy images of fresh mature human oocytes (metaphase II stage) donated for research (donations were made prior to 2004, after which legislation forbade research on human embryos).35 Oocytes were loaded with the mitochondrial specific fluorescence dye JC-1. (A) and (B) Images of a fresh metaphase II (MII) stage human oocyte with polarized granularity. (C) and (D) Images of a MII stage human oocyte with a ‘smooth’ (non-polarized) cytoplasm. (A) and (C) show the bright-field images of the oocytes (non-confocal) and (B) and (D) show a single confocal section of the mitochondrial membrane potential-insensitive channel (green fluorescence output). ‘I’ shows mitochondrial type I morphology. ‘II’ shows mitochondrial type II morphology (bright-field visualization in (A) and (C), and fluorescence images in (B) and (D)). ‘A’ denotes an area surrounding the meiotic apparatus in which no mitochondria are observed (see (D)). White bars represent 20 ␮m.



HPE_Chapter22.qxp

7/13/2007

4:53 PM

Page 278


only the density, but also the morphological appearance of mitochondria correlate with features observed under the light microscope37 (Figures 22.1–22.4). Two distinct mitochondrial morphologies can be seen throughout human oogenesis and preimplantation embryo development37 (Figures 22.1–22.4). Type I mitochondria appear as loosely packed clusters, usually towards the periphery of the cytoplasm (Figures 22.1–22.4). This morphology appears to correlate with a granular appearance of the oocyte cytoplasm when visualized under light microscopy.

I

I

II II A

II D

Figure 22.3 Mitochondrial morphology in human zygotes. Confocal microscopy images of human zygotes donated for research, 19–20 hours after insemination (donations were made prior to 2004, after which legislation forbade research on human embryos).35 Zygotes were loaded with the mitochondrial specific fluorescence dye JC-1. (A) and (B) Images of a human zygote with contracted cytoplasm surrounding the pronuclei. (C) and (D) Images of a human zygote with smooth cytoplasm (no cytoplasmic contraction) surrounding the pronuclei. (A) and (C) show the bright-field images of the oocytes (nonconfocal) and (B) and (D) a single confocal section of the mitochondrial membrane potential-insensitive channel (green fluorescence output). ‘I’ shows mitochondrial type I morphology. ‘II’ shows mitochondrial type II morphology (bright-field visualization in (A) and (C), and fluorescence images in (B) and (D)). White bars represent 20 ␮m.



II

A

B

Figure 22.4 Mitochondrial morphology in human embryos. Confocal microscopy images of a human six blastomere embryo 67 hours after insemination (embryos were donated for research prior to 2004, after which legislation forbade research on human embryos).35 Embryos were loaded with the mitochondrial specific fluorescence dye JC-1. (A) Shows the bright-field image of the embryo (non-confocal) and (B) a single confocal section of the mitochondrial membrane potential-insensitive channel (green fluorescence output). ‘I’ shows mitochondrial type I morphology. ‘II’ shows mitochondrial type II morphology (bright-field visualization in (A), and fluorescence images in (B)). White bars represent 20 ␮m.

B

II C

I II

I

Type II mitochondria are small and compact and exist at a very high density. This morphological type appears to correlate with a smooth appearance of the oocyte cytoplasm.37 Whereas GV oocytes are characterized by type I morphology towards the periphery of the oocyte, and some small areas of type II mitochondrial morphology surrounding the GV (Figure 22.1B), the cytoplasm of metaphase I and II oocytes is often polarized for the two distinct morphologies37 (Figures 22.1 and 22.2). Attempts have been made to correlate these morphological observations with developmental characteristics (see Table 22.1 and below). After fertilization in mammalian oocytes, changes in the distribution of mitochondria occur in synchrony with each round of cell division. In the zygote, mitochondria may be densely packed around the pronuclei, although this is often absent26,36 (Figure 22.3). This morphology often appears when the blastomere nucleus re-forms after each round of division is completed, alternating with a more even distribution of mitochondria at the time of nuclear membrane breakdown. Interestingly, mitochondria cluster around the presumed location of the centrosome in the preimplantation

HPE_Chapter22.qxp

7/13/2007

4:53 PM

Page 279


Table 22.1 Correlation between oocyte morphology and developmental characteristics during intracytoplasmic sperm injection (ICSI). Data refer to ICSI cycles where all oocytes observed prior to microinjection were of a single morphological type

Number of patients Age in years (mean ⫾ SD) Number of ICSI cycles Number of oocytes retrieved (mean ⫾ SD) Number of oocytes with normal fertilization (% fertilization) Number of embryos transferred (mean ⫾ SD) Number of grade I embryos transferreda (% of transferred) Number of pregnancies Number of gestational sacs (implantation rate per embryo transferred, %)

Type I

Type II

Type III

56 33.4 ⫾ 5.5 56 634 (11.1 ⫾ 3.5) 480 (75.8%)

44 32.8 ⫾ 4.9 44 515 (10.8 ⫾ 4.0) 284 (55.1%)

32 33 ⫾ 4.0 32 305 (9.2 ⫾ 3.6) 122 (40%)

152 (2.7 ⫾ 0.4) 135 (88.8%)

128 (2.9 ⫾ 0.3) 91 (71.1%)

89 (2.8 ⫾ 0.5) 65 (73.3%)

22 (39.3%) 36 (23.7%)

16 (36.4%) 28 (21.8%)

11 (34.4%) 13 (14.6%)

Type I, oocytes with granularity to one side of the cytoplasm; type II, oocytes where granularity was observed in the center of the oocyte; type III, oocytes where granularity was not observed. aGrade I embryos were scored by both morphology and growth rate using previously described methods.38

embryo, suggesting that this structure requires a high energy input (Figure 22.4). Both active (as revealed by Rh123, Mitotracker, or JC-1 staining)37,39,40 and inactive mitochondria (as visualized with nonyl acridine orange (NAO) staining),39 are reorganized in the same manner during development, suggesting that mitochondrial relocalization is not activity dependent, although highly active mitochondria may be localized to peripheral regions.40 The role of mitochondrial reorganization during each phase of development may reflect the energy requirement of distinct regions of the cytoplasm,26,36 and therefore the normal physiology of oocyte and embryo development.

RELATIONSHIP BETWEEN MITOCHONDRIAL LOCALIZATION, ACTIVITY, AND HUMAN EMBRYO DEVELOPMENT

The above data demonstrate that mitochondria localize to distinct regions of the oocyte and embryo cytoplasm during preimplantation embryo development and that they are active during this period. However, evidence regarding the role of these two features of mitochondria in mammalian preimplantation embryo development is less well characterized.

MITOCHONDRIAL LOCALIZATION

Observations on human metaphase II oocytes prior to intracytoplasmic sperm injection (ICSI) reveal a wide variety of morphologies in the oocyte cytoplasm, and that the cytoplasm may be smooth or granular in appearance under the light microscope. The distribution of these two types of cytoplasm is highly diverse, but is often polarized to distinct regions of the cytoplasm. Our group and others have recorded the appearance of the human oocyte prior to ICSI in an attempt to correlate these distinct types of morphology with fertilization, development, and implantation potential of individual oocytes.41–43 Human oocytes present a wide variety of morphological features, even within cohorts of oocytes from individual patients. Data from fluorescence labeling suggest that these granular features of oocytes represent distinct zones of mitochondrial localization. In our laboratory, we characterized the different oocyte morphologies into three groups for simplicity: granularity towards one side of the oocyte (type I), granularity in the center of the oocyte (type II), and complete absence of granularity (type III, Table 22.1). Patients may have a mixture of morphologies within a cohort of oocytes, but in some cases have oocytes with a single morphological type. When a single 

HPE_Chapter22.qxp

7/13/2007

4:53 PM

Page 280


morphological type is present, we can correlate clinical and biological parameters after fertilization and embryo transfer. In 132 patients from a total of 190 observed to have a single morphological type within the analysis period (Table 22.1), correlations between these oocyte types and clinical and biological parameters suggest that fertilization rates and embryo development do not show obvious differences. However, there was a significant difference in the clinical outcome based on oocyte morphology. Therefore, features of oocyte morphology may provide an insight into the developmental history, and in consequence the implantation potential of each individual oocyte. An accumulation of cytoplasm around the pronuclei can also be observed in the zygotes of human and other mammalian species, and this and other features have been applied to selection criteria in the IVF laboratory. Staining with mitochondrial dyes has shown that this accumulated material includes a large proportion of mitochondria. However, this effect is not always seen in human zygotes. Zygotes which have abnormal, or no cytoplasmic accumulation around the pronuclei show slower rates of development, abnormal distributions of mitochondria within the embryo, and significant differences in ATP content between blastomeres40,44 (Table 22.2). Therefore, the zygote stage could be considered an essential stage in the developmental programming

of the fertilized oocyte.Again, the cytoplasm of human preimplantation embryos is often characterized by granularity observed in the nuclear region. These morphological features continue through embryo development and may be highly indicative of human embryo implantation potential.2 MITOCHONDRIAL ACTIVITY

As described above, mitochondrial activity, measured by either mitochondrial membrane potential or oxygen consumption, is a measure of the level of aerobic respiration within oocytes. Correlations between the state of perifollicular vascularity of developing follicles, the dissolved oxygen content of follicular fluid, and the quality of oocytes retrieved from these follicles suggests that oxidative phosphorylation plays an important role in oogenesis,44,45 and aerobic respiration appears to be a vital source of energy during oocyte maturation. After the luteinizing hormone (LH) surge, the available evidence also suggests that mitochondrial metabolism has a fundamental role in the final maturation of the oocyte, since the rate of oxygen consumption, a measure of mitochondrial metabolism, increases dramatically in the isolated oocyte after the LH surge.13 Although an increase in lactic acid is also observed after the LH surge, an indication of anaerobic respiration,13

Table 22.2 Correlation between zygote morphology and developmental characteristics during ICSI. Data refer to ICSI cycles where zygotes of a single morphological type were observed 19–20 hours after microinjection and selected for transfer on day 2.

Number of patients Age in years (mean ⫾ SD) Number of ICSI cycles Number of zygotes selected (mean ⫾ SD) Number of grade I embryosa (% of zygotes) Number of embryos transferred Number of pregnancies Number of gestational sacs (implantation rate per embryo transferred, %)

Type I

Type II

Type III

64 32.3 ⫾ 5.0 64 189 (2.9 ⫾ 0.4) 157 (83.1%)

55 33.5 ⫾ 7.6 55 143 (2.6 ⫾ 0.6) 109 (76.2%)

13 32.8 ⫾ 6.0 13 37 (2.8 ⫾ 0.3) 25 (67.6%)

189 25 (39.1%) 47 (24.9%)

143 21 (38.2%) 26 (18.2%)

37 3 (23.1%) 4 (10.8%)

Type I, zygotes with pronuclei surrounded by granular cytoplasm in the center of the cell; type II, zygotes with pronuclei surrounded by granular cytoplasm to one side of the cell; type III, zygotes where granularity was not observed. aGrade I embryos were scored by both morphology and growth rate using previously described methods.38



HPE_Chapter22.qxp

7/13/2007

4:53 PM

Page 281


blocking anaerobic respiration with iodoacetate does not inhibit oocyte maturation, suggesting that anaerobic respiration is not necessary for oocyte maturation.46 In contrast, suppressing aerobic respiration does block oocyte maturation.13 However, the aerobic respiration measured here probably reflects the activity of granulosa cells surrounding the oocyte within the follicle. After release from the follicle and fertilization, mammalian oocytes and embryos are no longer dependent on helper cells, and start to produce and utilize their own energy. Although aerobic respiration is clearly present in human preimplantation embryos, its role in development is controversial. Mitochondria within human oocytes have few matrix membrane foldings and appear relatively inactive.47 In fact, estimates of the contribution of mitochondrial respiration to the energetic requirement of mammalian embryo development suggest that as little as 10% of glucose is metabolized through aerobic respiration in the early stages of development, although this rises to 85% in the blastocyst.36 However, since aerobic respiration is 14-fold more efficient than anaerobic respiration with respect to ATP production, the 10% of glucose passing through aerobic respiration produces more energy in the form of ATP than the 90% of glucose metabolized without mitochondria. Mammalian oocytes and preimplantation embryos prefer pyruvate as an energy source,36 and this preference for pyruvate does not eliminate either aerobic or anaerobic respiration as a possible route to ATP production. Lactic acid is present in fluid sampled from human fallopian tubes,48 and is also produced during mammalian embryo development in vitro,36 suggesting that there is a component of anaerobic respiration in the ATP-generating mechanism. Mitochondria within early cleavage stage preimplantation embryos do not replicate, and contain poorly formed cristae, so the level of mitochondrial respiration may be low during this stage.36 Aerobic respiration appears to be upregulated at the stage of blastocyst development: the mitochondrial cristae become more compact, mitochondria start to replicate, and the utilization of glucose as a carbohydrate substrate increases.49 These data suggest that both aerobic and anaerobic respiration pathways

are active during oocyte maturation and embryo preimplantation development, but aerobic respiration is upregulated during blastocyst development and implantation. Despite the fact that anaerobic respiration can permit some degree of mammalian embryo development in the absence of aerobic respiration, current evidence suggests that a minimum level of aerobic respiration is required to enable embryo implantation and subsequent development of the mammalian fetus.13 This hypothesis may explain the ‘maternal age effect’ observed in human fertility.13 Oocyte mitochondria replicate from a small initial pool of organelles residing in primordial follicles created in the fetus; some primordial follicles reside in the human ovary for up to 40 years, and are therefore subject to the degeneration of the mtDNA through free radical attack. It has therefore been suggested that the reduction in fertility associated with advanced maternal age is caused by a slow loss in the capacity of mitochondria to provide the energy required.13 Although a precise relationship is yet to be defined, there are some molecular data to support this hypothesis.50–52 Physiological data also suggest that there is a relationship between mitochondrial activity and maternal age;37 recent data suggest that the level of aerobic respiration may be directly correlated with maternal age,37 with a significant influence on embryo quality.53 Mitochondrial activity decreases linearly with increasing age and influences both embryo development and IVF outcome,13,26,37,53 and therefore appears to exert a strong influence on the inherent developmental potential of individual oocytes. Our previous data using fresh oocytes and embryos donated to research suggest that mitochondrial activity does not change between the maturing oocyte and the first stages of human embryo development,37 and this indicates that measuring the level of aerobic respiration in individual human oocytes could provide a strong indication of the implantation potential of developing embryos.37,53 Experimental elimination of mitochondrial activity in human and mammalian embryos does not immediately block embryo development. Evidence from mice suggests that embryos can develop, and even implant adequately in the absence of mitochondrial respiration, 

HPE_Chapter22.qxp

7/13/2007

4:53 PM

Page 282


although fetal development does not continue until birth.54–56 These data suggest that anaerobic respiration can support preimplantation embryo development, at least in the mouse. In the human, eliminating energy substrates in the culture medium lowers the mitochondrial membrane potential and slows development.57 It must be assumed that the lack of aerobic respiration can be compensated for, at least in the short term, by anaerobic respiration, enabling sufficient energy production to permit the completion of the cell cycle. Our present hypothesis suggests that aerobic and anaerobic respiration are coupled, and despite the fact that embryo development can persist in the absence of aerobic respiration,13 implantation requires a minimum level of aerobic respiration.

CAN MITOCHONDRIAL MORPHOLOGY OR ACTIVITY BE USED AS A PARAMETER TO ASSIST EMBRYO SELECTION?

The above data demonstrate that human oocyte and embryo morphology is related to the morphology of mitochondria within the oocyte cytoplasm, and suggest that the activity of mitochondria is important for embryo development. Could these parameters then be used to aid in embryo selection? Although we do not suggest that a single criterion can be used in selection, it is possible that the selection of embryos for transfer may be assisted by adding parameters related to mitochondria. Currently, embryo selection criteria include the rate of embryo development in vitro, the morphology of these embryos (blastomere volumes, presence of fragments, etc.)2, and to a lesser extent zygote morphology.2 Although some analysis of oocytes is included during ICSI protocols,2 we propose that oocyte cytoplasmic morphology could be added as a selection criterion. Our data suggest that the cytoplasmic morphology of human oocytes is an indicator of development potential (Table 22.1), and the presence of ‘polarized’ cytoplasm in the oocyte seems to be a positive indicator of implantation potential. However, the current evidence suggests that oocyte morphology alone is not highly predictive, as high quality embryos and pregnancy often result from a variety of cytoplasmic morphologies 

(Table 22.1). Therefore, assessment of oocyte cytoplasmic morphology should be added to the embryology routine, but not applied uniquely. The level of mitochondrial activity is not currently tested, but this could prove to be a highly useful tool in embryo selection criteria, even before fertilization. However, technical difficulties limit the application of these assays in the clinical laboratory. At present, oxygen consumption and fluorescence techniques can be used to measure the activity of mitochondria in mammalian oocytes and embryos. Fluorescence techniques measure the electrophysiological membrane potential across the mitochondrial matrix, and are therefore an indirect measure of mitochondrial activity. However, this technique involves the introduction of potentially hazardous dyes into the oocyte cytoplasm, and therefore it cannot be applied clinically. The advantage of the technique is that measurements can be made quickly in single oocytes and embryos, permitting individual variations in activity to be observed. Data compiled from fluorescence techniques have demonstrated that the activity of mitochondria in oocytes and preimplantation embryos is both patient and oocyte specific.37,58 The measurement of oxygen consumption is a direct measure of mitochondrial activity, and this can be done via two techniques. The first uses an oil soluble, non-toxic quarternary benzoid compound, pyrene (Sigma-Aldrich, St. Louis, MO, USA), to measure fluorescence that is reversibly quenched by oxygen.59 Pyrene’s oil solubility is an advantage, as this facilitates the technique’s application in IVF culture, where an oil overlay is often employed. However, in order to measure oxygen consumption, oocytes and embryos must be cultured in an air-tight apparatus for appreciable lengths of time (up to 4 hours) and in minute quantities of culture media in the absence of air.69 Even under these circumstances, the minute amounts of oxygen consumed by individual oocytes and embryos (19.6 pmol/embryo/ hour)60 mean that a minimum of three embryos are required to allow the measurement of oxygen consumption.60 Therefore, the oxygen consumption of individual human oocytes and embryos has not yet been determined. A second technique utilizes the fluorescent dye tris-1,7-diphenyl-1,10-phenanthroline ruthenium (II)

HPE_Chapter22.qxp

7/13/2007

4:53 PM

Page 283


chloride (BD Biosciences, Bedford, MA, USA) the fluorescence of which is also reversibly quenched by oxygen. The advantage of this technique is that the dye can be embedded in a gas-permeable silicone polymer and used to coat standard cell culture plates for analysis. Theoretically, the set-up and analysis of the material can be performed inside a standard incubator, although minute quantities of cell culture media and sealed conditions are again required. This system has not been tested on human oocytes or embryos at the present time. A third technique for measuring mitochondrial activity is through the use of coenzymes such as nicotinamide adenine dinucleotide (NAD), nicotinamide adenine dinucleotide phosphate (NADP), or flavine adenine dinucleotide (FAD), which are vital to metabolic pathways through their properties of reversible oxidation. The relative proportion of these compounds in the cytoplasm can be directly measured by non-invasive fluorescence techniques. NAD(P)H emits fluorescence at 445 nm when excited at a wavelength of 340 nm, whereas FAD⫹⫹ emits fluorescence at 520 nm when excited at a wavelength of 458 nm. Both NAD(P) and FAD are linked to the aerobic and anaerobic metabolic pathways.13 Therefore, the level of fluorescence derived from individual oocytes could be used to determine the relative levels of aerobic and anaerobic metabolism in cells. However, excitation of NAD(P)H requires ultraviolet light (340 nm), and this unfortunately poses risks for clinical use. These risks could theoretically be overcome by the use of newer techniques such as two-photon fluorescence microscopy.61 FAD⫹⫹ fluorescence can be measured with visible light wavelengths, and this is therefore safe. Both NADH and FAD⫹⫹ fluorescence have been successfully measured in mouse eggs,62 but calibrating the fluorescence output in terms of mitochondrial activity is a problem that remains to be solved. Genetically, the measurement of mitochondrial DNA mutations or the expression of mitochondrial genes could be used to determine the health of an embryo. Both large mitochondrial DNA mutations and point mutations have been detected in human oocytes and preimplantation embryos.50,52,63 Although no definite relationship has yet been defined between these

mutations and reproductive capacity, the fact that the incidence of point mutations in human oocytes was correlated with maternal age appears significant.52 The expression of several mitochondrial genes has also been measured in unfertilized oocytes and arrested embryos from assisted reproduction programs.64 Data suggesting a correlation may in the future serve as a diagnostic tool for the selection of embryos for transfer.

CONCLUSIONS

Physiological techniques that can be applied to the selection of viable embryos for transfer during IVF cycles are currently scarce. In contrast, morphological and chromosomal data are abundant. Although morphological and chromosomal analyses are easily applied in the IVF laboratory, they do not reliably determine an important parameter – the implantation potential of the human embryo. Identifying the implantation potential of the human preimplantation embryo from the stage of the unfertilized human oocyte could prove to be a highly effective tool in embryo selection, especially in combination with diverse techniques applied to the embryology routine in order to assist selection.2 We suggest that measuring an aspect of oocyte physiology such as mitochondrial activity could prove to be a very useful addition to the complement of selection techniques available. Although our data suggest that characteristics of oocyte and embryo cytoplasm are related to mitochondrial morphology, this observation is limited by the fact that mitochondrial morphology, at the level revealed by fluorescence microscopy, is not related to mitochondrial activity. In recent years, there has been an increased interest in the role of mitochondrial metabolism during preimplantation human embryogenesis. Current data suggest that measuring mitochondrial metabolism may be useful as an aid in selecting human oocytes and embryos for transfer, if the technical difficulties inherent in measuring mitochondrial metabolism can be overcome to the extent that analyses can be performed during routine IVF cycles. We feel that this could become an important tool for successful in 

HPE_Chapter22.qxp

7/13/2007

4:53 PM

Page 284


vitro fertilization. In combination with morphological assessment and chromosomal analysis, the assessment of mitochondria may add an important step towards the goal of IVF – the transfer of a single embryo.

ACKNOWLEDGMENTS

We thank the Fondazione Nuovi Orizzonti, Naples, Italy and Vincenzo Monfrecola for technical support. REFERENCES 1. Munné S, Magli MC, Cohen J et al. Positive outcome after preimplantation diagnosis of aneuploidy in human embryos. Hum Reprod 1999; 14: 2191–9. 2. Scott L. The biological basis of non-invasive strategies for selection of human oocytes and embryos. Hum Reprod Update 2003; 9: 237–49. 3. Liu H, Trimarchi J, Keefe D. Involvement of mitochondria in oxidative stress-induced cell death in mouse zygotes. Biol Reprod 2000; 62: 1745–53. 4. Mohamad N, Gutierrez A, Nunez M et al. Mitochondrial apoptotic pathways. Biocell 2005; 29: 149–61. 5. Pozzan T, Magalhaes P, Rizzuto R. The comeback of mitochondria to calcium signalling. Cell Calcium 2000; 28: 279–83. 6. Naviaux RK. Mitochondrial DNA disorders. Eur J Pediatr 2000; 159 (Suppl 3): S219–26. 7. Codon AC, Gasent-Ramirez JM, Benitez T. Factors which affect the frequency of sporulation and tetrad formation in Saccharomyces cerevisiae baker’s yeasts. Appl Environ Microbiol 1995; 61: 630–8. 8. Singh PJ, Santella RN, Zawada ET. Mitochondrial genome mutations and kidney disease. Am J Kidney Dis 1996; 28: 140–6. 9. Pulkes T, Hanna MG. Human mitochondrial DNA diseases. Adv Drug Deliv Rev 2001: 49: 27–43. 10. Shoffner JM. An introduction: oxidative phosphorylation diseases. Semin Neurol 2001: 21: 237–50. 11. Wallace DC. The mitochondrial genome in human adaptive radiation and disease: on the road to therapeutics and performance enhancement. Gene 2005; 354: 169–80. 12. Maassen JA, Janssen GM, ‘t Hart LM. Molecular mechanisms of mitochondrial diabetes (MIDD). Ann Med 2005; 37: 213–21. 13. Wilding M, Di Matteo L, Dale B. The maternal age effect: a hypothesis based on oxidative phosphorylation. Zygote 2005; 13: 1–7. 14. Lodish H, Baltimore D, Berk A et al. Molecular Cell Biology, 3rd edn. New York: Freeman and Company, 1995. 15. Mechler F, Fawcett PR, Mastaglia FL et al. Mitochondrial myopathy. J Neurol Sci 1981; 50: 191–200. 16. Anderson S, Bankie AT, Barrell BG. Sequence and organisation of the mitochondrial genome. Nature 1981; 290: 457–65. 17. Saraste M. Oxidative phosphorylation at the fin di siecle. Science 2000; 283: 1488–92. 18. Clayton D. Transcription and replication of mitochondrial DNA. Hum Reprod 2000; 15 (Suppl 2): 11–7. 19. Trounce L. Genetic control of oxidative phosphorylation and experimental models of defects. Hum Reprod 2000; 15 (Suppl 2): 18–28.



20. Cummins J. The role of maternal mitochondria during oogenesis, fertilisation and embryogenesis. Reprod Biomed Online 2002; 4: 176–82. 21. Wallace DC. Maternal genes: mitochondrial diseases. In: McKusick VA, Roderick TH, Mori J, Paul MW, eds. Medical and Experimental Mammalian Genetics: A Perspective. New York, USA: Liss for March of Dimes Foundation, 1987; 23: 137–90. 22. Tarin JJ. Aetiology of age-associated aneuploidy: a mechanism based on the free-radical theory of ageing. Hum Reprod 1995; 10: 1563–5. 23. Linnane AW, Marzuki S, Ozawa T et al. Mitochondrial mutations as an important contributor to ageing and degenerative diseases. Lancet 1989; i: 642–5. 24. Linnane AW, Zhang C, Baumer A et al. Mitochondrial DNA mutations and the ageing process: bioenergy and pharmacological intervention. Mutat Res 1992; 275: 195–208. 25. Harman D. The biologic clock: the mitochondria? J Am Geriatr Soc 1972; 20: 145–7. 26. Van Blerkom J. Mitochondria in human oogenesis and preimplantation embryogenesis: engines of metabolism, ionic regulation and developmental competence. Reproduction 2004; 128: 269–80. 27. Krebs HA. Gluconeogenesis. Expos Annu Biochim Med 1965; 26: 13–30. 28. Mitchell P and Moyle J. Chemiosmotic hypothesis of oxidative phosphorylation. Nature 1967; 213: 137–9. 29. Mitchell P. Keilins respiratory chain concept and its chemiosmotic consequences. Science 1979; 206: 1148–59. 30. Hauswirth W, Laipis P. Rapid segregation of heteroplasmic bovine mitochondria. Nucleic Acids Res 1982; 17: 7325–31. 31. Jansen R. Germline passage of mitochondria: quantitative considerations and possible embryological sequelae. Hum Reprod 2000; 15 (Suppl 2): 112–28. 32. Cummins J. Fertilisation and elimination of the paternal mitochondrial genome. Hum Reprod 2000; 15 (Suppl 2): 92–101. 33. Reynier P, May-Panloup P, Chretien M et al. Mitochondrial DNA content affects the fertilisability of human oocytes. Mol Hum Reprod 2001; 7: 425–9. 34. Barritt JA, Kokot M, Cohen J et al. Quantification of human ooplasmic mitochondria. Reprod Biomed Online 2002; 4: 243–7. 35. Legge 19 Feb 2004, n. 40. Norme in materia di procreazione medicalmente assistita. Gazz Uff Ital 2004; 45. 36. Bavister BD, Squirrell JM. Mitochondrial distribution and function in oocytes and early embryos. Hum Reprod 2000; 15 (Suppl 2): 189–98. 37. Wilding M, Dale B, Marino M et al. Mitochondrial aggregation patterns and activity in human oocytes and preimplantation embryos. Hum Reprod 2001; 16: 909–17. 38. De Placido G, Wilding M, Strina I et al. High outcome predictability after IVF using a combined score for zygote and embryo morphology and growth rate. Hum Reprod 2002; 17: 2402–9. 39. Barnett DK, Kimura J, Bavister B. Translocation of active mitochondria during hamster preimplantation embryo development studied by confocal laser scanning microscopy. Dev Dyn 1996; 205: 64–72. 40. Van Blerkom J, Davis P, Mathwig V et al. Domains of high-polarised and low-polarised mitochondria may occur in mouse and human oocytes and early embryos. Hum Reprod 2002; 17: 393–406. 41. De Sutter P, Dozortsev D, Qian C et al. Oocyte morphology does not correlate with fertilisation rate and embryo quality after intracytoplasmic sperm injection. Hum Reprod 1996; 11: 595–7. 42. Serhal PF, Ranieri DM, Kinis A et al. Oocyte morphology predicts outcome of intracytoplasmic sperm injection. Hum Reprod 1997; 12: 1267–70. 43. Loutradis D, Drakakis P, Kallianidis K et al. Oocyte morphology correlates with embryo quality and pregnancy rate after intracytoplasmic sperm injection. Fertil Steril 1999; 72: 240–4.

HPE_Chapter22.qxp

7/13/2007

4:53 PM

Page 285


44. Van Blerkom J. Intrafollicular influences on human oocyte developmental competence: perifollicular vascularity, oocyte metabolism and mitochondrial function. Hum Reprod 2000; 15 (Suppl 2): 173–88. 45. Van Blerkom J, Antczak M, Schrader R. The development potential of the human oocyte is related to the dissolved oxygen content of follicular fluid: association with vascular endothelial growth factor levels and perifollicular blood flow characteristics. Hum Reprod 1997; 12: 1047–55. 46. Tsafriri A, Lieberman ME, Ahren K et al. Disassociation between LHinduced aerobic glycolysis and oocyte maturation in cultured Graafian follicles of the rat. Acta Endocrinol (Copenhagen). 1976; 81: 362–6. 47. Motta P, Nottola S, Makabe S et al. Mitochondrial morphology in human fetal and adult female germ cells. Hum Reprod 2000; 15 (Suppl 2): 129–47. 48. Beier HM. Oviductal and uterine fluids. J Reprod Fertil 1974; 37: 221–37. 49. Houghton FD, Leese HJ. Metabolism and developmental competence of the preimplantation embryo. Eur J Obstet Gynecol Reprod Biol 2004; 115 (Suppl 1): S92–6. 50. Brenner C, Wolny YM, Barritt JA et al. Mitochondrial DNA deletion in human oocytes and embryos. Mol Hum Reprod 1998; 4: 887–92. 51. Seifer DB, DeJesus V, Hubbard K. Mitochondrial deletions in luteinised granulosa cells as a function of age in women undergoing in vitro fertilisation. Fertil Steril 2002; 78: 1046–8. 52. Barritt JA, Cohen J, Brenner CA. Mitochondrial DNA point mutations in human oocytes is associated with maternal age. Reprod Biomed Online 2000; 1: 96–100. 53. Wilding M, De Placido G, Di Matteo L et al. Chaotic mosaicism in human preimplantation embryos is correlated with a low mitochondrial membrane potential. Fertil Steril 2003; 79: 340–6. 54. Piko L, Chase DH. Role of the mitochondrial genome during early development in mice. Effects of ethidium bromide and chloramphenicol. J Cell Biol 1973; 58: 357–78.

55. Larsson N, Wang J, Wilhelmsson H et al. Mitochondrial transcription factor A is necessary for mtDNA maintenance and embryogenesis in mice Nat Genet 1998; 18: 231–6. 56. Li K, Li Y, Shelton J et al. Cytochrome c deficiency causes embryonic lethality and attenuates stress-induced apoptosis. Cell 2000; 101: 389–99. 57. Wilding M, Fiorentino A, De Simone ML et al. Energy substrates, mitochondrial membrane potential and human preimplantation embryo division. Reprod Biomed Online 2002; 5: 39–42. 58. Van Blerkom J, Davis P, and Lee J. ATP content of human oocytes and developmental potential and outcome after in vitro fertilisation and embryo transfer. Hum Reprod 1995; 10: 415–24. 59. Houghton FD, Thompson JG, Kennedy CJ et al. Oxygen consumption and energy metabolism of the early mouse embryo. Mol Reprod Dev 1996; 44: 476–85. 60. Butcher L, Coates A, Martin KL et al. Metabolism of pyruvate in the early human embryo. Biol Reprod 1998; 58: 1054–6. 61. Williams RM, Piston DW, Webb WW. Two-photon molecular excitation provides intrinsic 3-dimensional resolution for laser-based microscopy and microphotochemistry. FASEB J 1994; 8: 804–13. 62. Dumollard R, Marangos P, Fitzharris G et al. Sperm-triggered [Ca2⫹] oscillations and Ca2⫹-homeostasis in the mouse egg have an absolute requirement for mitochondrial ATP production. Development 2004; 131: 3057–67. 63. Barritt JA, Brenner CA, Cohen J et al. Mitochondrial DNA rearrangements in human oocytes and embryos. Mol Hum Reprod 1999; 5: 27–33. 64. Hsieh RH, Au HK, Yeh TS et al. Decreased expression of mitochondrial genes in human unfertilised oocytes and arrested embryos. Fertil Steril 2004; 81 (Suppl.1): 912–8.



HPE_Chapter22.qxp

7/13/2007

4:53 PM

Page 286

HPE_Chapter23.qxp

7/13/2007

5:31 PM

Page 287

23. Future genetic and other technologies for assessing embryos Dagan Wells

OVARIAN STIMULATION AND HIGH-ORDER MULTIPLE PREGNANCY

The use of ovarian stimulation has provided a means of obtaining multiple oocytes per assisted reproductive technology (ART) treatment cycle. In most cases the majority of oocytes retrieved will be successfully fertilized, and as a result, several embryos are typically available for transfer. Because the probability of an individual embryo forming a viable pregnancy is relatively low (approximately 15%) it has long been standard practice to transfer more than one embryo per cycle. This approach has been highly successful at increasing take-home-baby rates, thus reducing the likelihood that a patient will need to face the emotional, physical, and financial burden of multiple treatment cycles. Unfortunately, the transfer of multiple embryos has also led to an explosion in the incidence of highorder multiple gestations. Since the introduction of in vitro fertilization (IVF) more than 20 years ago the incidence of such pregnancies has increased 400% for women in their thirties and 1000% for women in their forties (National Center for Health Statistics). Although some patients initially welcome the prospect of a ‘ready made family’, the medical reality is that pregnancies of this type carry a significantly elevated risk of serious complications for both the mother and children. For the mother there is increased risk of problems such as preeclampsia, gestational diabetes, and vaginal and uterine hemorrhaging. Fetal complications include a high-risk of prematurity and very low birth weight, as well as significantly elevated risks of miscarriage, infant mortality, and cerebral palsy. High-order multiple pregnancies also have financial implications for both the parents and the health care provider.1–3

For example, a triplet pregnancy will typically cost a health care provider in the United States approximately $300 000.

EMBRYO VIABILITY ASSESSMENT – MORPHOLOGY

The problem of multiple gestations in IVF stems from the fact that current methods for predicting embryo viability and implantation potential are relatively poor. Multiple embryos are transferred because this is the only way to maximize the probability that at least one viable embryo has been selected. It is inevitable that in some cases all of the embryos transferred will have the potential to implant. There is an increasing desire within the field of reproductive medicine to overcome the problem of multiple gestations. Additionally, in some countries, legislation to limit the number of embryos transferred has come into force.4 Although it may be generally accepted that high-order multiple pregnancies are not a desirable outcome of IVF treatment, many laboratories remain concerned about the impact of a reduction in the number of embryos transferred on pregnancy rates. This is particularly true in countries where IVF success rates are published and influence the number of patients seeking treatment at a particular center. If pregnancy rates are to be maintained while the number of embryos transferred is reduced, ultimately culminating in the routine use of single embryo transfer at some point in the future, it will be necessary to dramatically improve our ability to assess embryo viability. One of the most important challenges encountered each day in IVF laboratories 

HPE_Chapter23.qxp

7/13/2007

5:31 PM

Page 288


is the determination of which embryos to select for transfer to the uterus. Which embryos have the greatest probability of providing the patient with a successful pregnancy? In the majority of laboratories, microscopic assessment of developmental stage and morphology of the embryo is the most important guide to future viability. Many different systems for scoring zygotes and embryos based on their morphology have been described, and in most cases these schemes have been demonstrated to assist in distinguishing viable embryos from non-viable to some extent.5–11 However, morphological analysis remains a relatively inefficient means of assessing embryos. All physicians and embryologists working in IVF can recount stories of patients who received two embryos of extremely poor morphology and became pregnant with twins. Similarly, we are all familiar with cases in which embryo morphology was perfect and yet no pregnancy ensued.

EMBRYO VIABILITY ASSESSMENT – PREIMPLANTATION GENETIC SCREENING

One reason why morphologically normal, apparently healthy, IVF embryos may fail to form a pregnancy is the presence of aneuploidy – an incorrect number of chromosomes. During the early 1990s, research using fluorescence in situ hybridization (FISH) revealed that chromosome anomalies in human preimplantation embryos are startlingly common. Analysis of between three and six of the 24 types of chromosome (22 pairs of autosomes plus the X and Y), revealed that more than 50% of cleavage stage embryos contained at least one abnormal cell.12–15 Later investigations using comparative genomic hybridization (CGH), a more comprehensive cytogenetic method that permits the entire chromosomal set to be screened in a single cell, suggested that between two-thirds and threequarters of human embryos contain abnormal cells on day 3 postfertilization.16,17 Most of the anomalies detected have been shown to be lethal,18 and their high prevalence in embryos produced during 

IVF treatment is thought to explain many cases of embryonic arrest, implantation failure, and early spontaneous abortion. A critical point of note is that most chromosomally abnormal embryos are morphologically indistinguishable from their euploid counterparts. A major breakthrough in embryo assessment came with the development of preimplantation genetic diagnosis (PGD) for the detection of chromosome abnormalities, known as PGD-AS (PGDaneuploidy screening) or PGS (preimplantation genetic screening). The most widely applied PGS strategy involves the biopsy of a single blastomere on day 3 postfertilization. The blastomere can be subjected to a wide range of genetic analyses. In the case of chromosomal screening, the cell is spread onto a microscope slide, fixed, and then analyzed using FISH. At present, the most experienced laboratories assess up to ten chromosomes per cell. This is achieved by employing two sequential rounds of hybridization using chromosome-specific probes labeled with distinct fluorescent dyes. The use of PGS as a guide to selecting embryos to be considered for transfer, supplementing (not replacing) morphological analysis, has been shown to significantly improve several IVF outcomes. Not surprisingly, the screening of chromosomes 13, 18, 21, X, and Y has led to a reduction in the number of pregnancies affected by aneuploid syndromes (i.e. Patau, Edward, Down, Turner and Klinefelter).19 This is particularly evident in IVF patients where the female partner is over 35 years of age, an age after which the risk of producing a chromosomally abnormal oocyte begins to increase rapidly. It is also clear that certain groups of IVF patient display a decline in the incidence of spontaneous abortion following PGS. This is not surprising, given that chromosome imbalance is known to be the major cause of early miscarriage in natural cycles. Of first trimester pregnancy losses, 60–70% display chromosome anomalies.20–22 Of most significance to the issue of multiple pregnancies is the impact of PGS on embryo implantation rates. Although some controversy remains, there is mounting evidence that PGS provides specific groups of IVF patient with a significant increase

HPE_Chapter23.qxp

7/13/2007

5:31 PM

Page 289

FUTURE GENETIC AND OTHER TECHNOLOGIES FOR ASSESSING EMBRYOS

in implantation rate per embryo transferred.23,24 Apparent contradictions in the literature regarding the impact of PGS on implantation rate seem to be a consequence of methodological differences between studies.25 Data suggest that increased implantation is only conveyed by PGS strategies employing biopsy of a single blastomere (rather than two) and analysis of at least seven chromosomes. If PGS truly has the ability to identify embryos with high implantation potential, then it is likely that this type of screening can help to reduce the incidence of multiple pregnancy, while maintaining an acceptable pregnancy rate. A recent study performed by Munné et al has provided preliminary data in support of this suggestion. The patients taking part in the study were divided into two groups: (1) embryos assessed using PGS and transfer limited to one to two embryos per cycle; and (2) a control group in which embryos were transferred without any genetic screening and no restriction was placed on the number of embryos transferred to the patient. Data from over 100 cycles were documented in each group, with patients matched for a variety of factors including maternal age. As with earlier studies, embryos transferred after PGS displayed an approximate doubling of implantation rate compared with the control group. However, despite the increased implantation rate for the PGS embryos, pregnancy rates for the two groups were essentially identical (30% for the PGS group, 33% for the control). This is because fewer embryos were transferred per cycle in the PGS group (average of 1.5 per cycle for PGS versus 3.6 for control). Importantly, no high-order multiple pregnancies were seen in the PGS group, whereas two triplet pregnancies and more twins were present in the control group (Munne et al, personal communication).

PREIMPLANTATION GENETIC SCREENING – FUTURE DIRECTIONS

Although PGS has provided significant benefits in terms of embryo screening and IVF outcome, existing protocols assess less than half of the chromosomal complement. Examination of the chromosomes

not usually tested during PGS in a research context, has shown that any chromosome can display aneuploidy during preimplantation development. Thus, it is inevitable that some aneuploid embryos are incorrectly classified ‘normal’ using current PGS methods, and may be inadvertently transferred. Such embryos are believed to have little potential for forming a viable pregnancy and most likely fail to implant or spontaneously abort at an early stage of gestation. It seems logical that an expanded chromosomal screen, permitting assessment of every chromosome, would improve the ability of PGS to identify viable embryos. However, a comprehensive chromosomal screen cannot be achieved using FISH due to the restricted number of spectrally distinct fluorochromes available for probe labeling. Although this problem can be partially overcome by performing sequential FISH experiments on the same cell, accuracy declines with each additional hybridization and consequently most PGS laboratories limit the number of rounds of FISH to two. Although methods such as G-banding, multiplex-FISH (M-FISH), and spectral karyotyping (SKY) provide information on every chromosome, they are dependent on the presence of cells in metaphase. Unfortunately the vast majority of blastomeres sampled from day 3 embryos are found to be in interphase. During this phase of the cell cycle chromosomes are contained within the nucleus and cannot be distinguished from one another. On the rare occasions when a blastomere in metaphase is biopsied, artifacts of the method used for fixation and spreading, most notably loss of chromosomes, are often observed. Comparative genomic hybridization (CGH) is a method related to FISH that permits analysis of the entire chromosome complement. Aneuploidy detection is achieved via a competitive hybridization of differentially labeled DNA samples to normal metaphase chromosomes26 (Figure 23.1). In the case of PGS, DNA from the sample (blastomere) is labeled with a green fluorochrome, while chromosomally normal reference DNA (usually from 46 XY lymphocytes) is labeled with a red fluorochrome. The two labeled samples are simultaneously hybridized 

HPE_Chapter23.qxp

7/13/2007

5:32 PM

Page 290


Test DNA

Normal DNA 46 XY

Normal

Trisomy

Monosomy

1:1

3:2 Ratio red : green

1:2

Figure 23.1 Comparative genomic hybridization (CGH) involves the simultaneous hybridization of differentially labeled DNA samples – a chromosomally normal DNA (labeled with a red fluorochrome) and sample DNA (labeled with a green fluorochrome). The relative number of red and green DNA fragments hybridizing to a given chromosome is related to the number of copies of the chromosome in question in the sample compared with the control. An excess of chromosomal material in the sample causes a green coloration, while a deficiency leads to a red coloration. Identical copy numbers of sample (green) and control (red) chromosomes results in equal hybridization of red and green fragments and a yellow coloration. In most cases a computer is used to assist in the calculation of red : green fluorescence intensity along the length of each chromosome.

to normal metaphase chromosomes on a microscope slide. Red and green DNA fragments compete for hybridization to their complementary sequences on the chromosomes, causing each chromosome to adopt a coloration related to its copy number in each of the samples. For example, if the green (blastomere) DNA was from an embryo with an extra copy of chromosome 16, there would be relatively more green copies of chromosome 16 DNA fragments than red. This will cause green DNA to outcompete red DNA for hybridization to chromosome 16. The overall effect is that chromosome 16 appears more green than the other chromosomes. Chromosome imbalances of this type are most clearly indicated by analyzing the ratio of green : red fluorescence along the length of each chromosome. CGH represents an excellent method for the simultaneous assessment of all chromosomes, and can 

even reveal errors involving fragments of chromosomes, provided they are at least 5 Mb in size. Unfortunately, CGH requires ~1 ␮g of DNA whereas a single blastomere contains only 5–10 pg. For this reason it is necessary to perform whole genome amplification (WGA) prior to CGH analysis of individual cells. The method of choice for this purpose is known as degenerate oligonucleotide primed (DOP) PCR.27,28 DOP-PCR utilizes a semidegenerate oligonucleotide primer that has the ability to anneal at many sites throughout the genome. With the addition of a thermostable DNA polymerase, DNA synthesis is initiated at each site of primer annealing. Thermal cycling using the principles of the polymerase chain reaction (denaturation, annealing, synthesis) permits a dramatic amplification of DNA from the original sample and provides enough material for subsequent CGH analysis.

HPE_Chapter23.qxp

7/13/2007

5:32 PM

Page 291


Studies applying CGH to the analysis of cleavage stage embryos of good morphology confirmed that aneuploidy could affect any chromosome during human preimplantation development. Abnormalities included chromosome breakage and varieties of aneuploidy that are not seen in prenatal samples or in material from spontaneous abortions (e.g. imbalance affecting the larger chromosomes of groups A and B).16,17 These unfamiliar forms of abnormality are presumed to be lethal during early stages of development. Analysis of the published CGH data reveals that 25–30% of embryos carry chromosome abnormalities that would not be detectable using the ninechromosome PGS examination employed by most PGD laboratories.29,30 The fact that current PGS protocols are not 100% successful in preventing the transfer of aneuploid embryos argues for the development of methods that permit a comprehensive analysis of blastomere chromosomes. However, it may be possible to significantly improve embryo selection simply by adjusting the combination of chromosomes screened. In most cases the chromosomes selected for screening are included because they are frequently found to be abnormal in prenatal samples and miscarriages. However, these chromosomes are not necessarily the most relevant in terms of implantation failure.31 Unfortunately, there are insufficient published CGH data to allow the incidence of aneuploidy to be precisely calculated for individual chromosomes, and consequently it is difficult to determine precisely which chromosomes should be prioritized for screening. However, pooling all existing CGH data can provide a rough indication of the most common aneuploidies at the cleavage stage. In most cases this analysis confirms the importance of the chromosomes currently assessed by PGS. However, it also suggests that screening might be improved if some chromosomes were substituted by others with higher preimplantation aneuploidy rates. Chromosomes that have not been extensively studied in human embryos, but appear to show above average aneuploidy rates include chromosomes 2, 4, 7, 9, 12, and 20. This observation is based on a small number of samples and requires confirmation using

CGH or FISH on a much larger series of embryos before adjustment of clinical PGS strategies. In addition to the CGH data, some FISH studies have also indicated that a revision of the chromosomes assessed by PGS may be necessary.32 From the pooled CGH data, one can predict that the proportion of aneuploid embryos detected could be increased from 70–75% to ⬎95% if the number of chromosomes screened by FISH was expanded from nine to 15. However, although it may be possible to achieve high detection rates for aneuploid embryos without screening the entire chromosome complement, it is inevitable that maximum accuracy will only be achieved if all the chromosomes are evaluated. To date CGH has been the most promising method for this purpose. However, as with FISH there are limitations to this technology. The principal difficulties in applying CGH clinically are the complexity of the technique and the length of time required (approximately 5 days), which is incompatible with the restricted timeframe available for PGS. One strategy for overcoming the problem of timeframe involves cryopreservation of embryos after biopsy, with embryo transfer occurring in a subsequent cycle.29,30,33 The main drawback of this approach is that freezing and thawing can lead to a reduction in embryo implantation potential, a problem that is exacerbated by embryo biopsy. An alternative strategy for the clinical application of CGH is based upon assessment of the first polar body (PB).34 The first PB is available for analysis 3 days earlier than biopsied blastomeres, providing sufficient time to perform CGH without the need for cryopreservation. However, confirmatory FISH analysis of blastomeres is advisable since chromatid anomalies detected in meiosis I have only a 50% chance of leading to an aneuploid embryo. There is also the potential for misdiagnosis due to a meiosis II error or a chromosome abnormality of paternal origin. Whatever the CGH strategy employed for PGS, the complexity of the technique remains a major problem. The protocol is labor intensive and necessitates expertise in both molecular genetic and cytogenetic methods that are not generally available to fertility clinics. A less complex approach will be 

HPE_Chapter23.qxp

7/13/2007

5:32 PM

Page 292


necessary if CGH is to be widely applied. At the present time, the best hope for a simplified methodology is microarray CGH. As with conventional CGH, this method involves the competitive hybridization of differentially labeled test and reference DNA samples. However, in this case the labeled DNAs are hybridized to DNA probes fixed to a microscope slide rather than metaphase chromosomes. Each probe is specific to a different chromosomal region and occupies a discrete spot on the slide. Chromosomal loss or gain is revealed by the color adopted by each spot after hybridization (ratio of red : green fluorescence). Importantly, the evaluation of red : green fluorescence is easily automated, circumventing the need for an experienced cytogeneticist. Microarray CGH has been successfully applied for the detection of aneuploidies in single cells after whole genome amplification using DOP-PCR or an alternative method known as multiple displacement amplification (MDA).35–37 Using this approach, comprehensive chromosome analysis has been achieved in less than 48 hours (i.e. within the timeframe necessary for PGS) and consequently the future application of microarray CGH for PGS appears to be extremely encouraging. Although comprehensive chromosomal analysis would seem to be advantageous for PGS, some physicians have expressed concern that a fullchromosome screening will increase the number of embryos excluded due to aneuploidy, leading to more IVF cycles in which there are no embryos eligible for transfer. Even though current protocols only assess nine or ten chromosomes, a small but significant number of PGS cycles end without embryo transfer due to all embryos receiving an unfavorable diagnosis. Fortunately, the small amount of data currently available indicate that the proportion of embryos diagnosed as normal may actually increase following CGH analysis. In a small study comparing FISH and CGH for the purposes of PGS, 40% of embryos analyzed by CGH were found to be chromosomally normal, while the proportion of embryos diagnosed normal by FISH was 33%.30 The most likely explanation for this counterintuitive observation is inaccuracy in the FISH methodology caused by the 

presence of micronuclei. Micronuclei, which are a relatively common phenomenon in human blastomeres, contain chromosomal material and are prone to loss during fixation and spreading prior to FISH. This may explain why an excess of monosomies (relative to trisomies) have been detected following FISH analysis of human embryos. CGH does not involve the spreading of the sample and consequently avoids this source of error.

RELATIONSHIP BETWEEN GENE EXPRESSION AND EMBRYO VIABILITY

The application of PGS has been shown to benefit several groups of IVF patients and proves that chromosome abnormality is an important factor affecting embryo implantation potential. However, when considering implantation failure it is also clear that chromosome anomaly is only part of the story. Many IVF-PGS cycles fail to provide a pregnancy, despite the transfer of embryos classified as chromosomally and morphologically normal. Furthermore, there are significant subgroups of IVF patients that show little or no improvement in embryo implantation following PGS (e.g. patients under 35 years of age). Although some of these failures may be attributed to uterine or endocrine problems, it is likely that many (perhaps most) are the consequence of nonchromosomal embryological issues. A host of vital, yet largely invisible, processes occur during the first few days of life. These include the transition away from a reliance on maternal protein and mRNA that accompanies activation of the embryo genome, and the first types of cellular differentiation. Given the complex and fundamental nature of events the embryo must successfully undertake at this time, it is perhaps unsurprising that developmental arrest during the preimplantation phase is common. One of the great challenges now facing scientists involved in translational IVF research is to characterize the cellular pathways perturbed in embryos destined for implantation failure. An improved understanding in this area may lead to new viability markers, revealing embryos that are under pressure and heading for arrest, and

HPE_Chapter23.qxp

7/13/2007

5:32 PM

Page 293


assist in the optimization of ovarian stimulation and embryo culture methods. An extremely useful approach for investigating the processes occurring in a biological sample (e.g. a cell, a tissue, an embryo, etc.) is to examine gene expression. The activity of individual genes is constantly changing, fluctuating in response to the changing needs of the cell. Variation in environmental conditions will induce changes in gene expression, as will factors such as altered metabolic requirements, the presence of aneuploidy or DNA damage, progress through the cell cycle, and processes such as differentiation and apoptosis. Quantification of the number of mRNA transcripts derived from a given gene provides an indication of how actively expressed it is. This in turn provides information on the activity of the cellular pathways that require the product of that gene. Embryonic gene expression at different preimplantation stages has been assessed using real-time polymerase chain reaction (PCR). This method involves extraction of RNA from the sample, reverse transcription in order to produce cDNA and PCR amplification of a cDNA fragment from the gene(s) of interest. The accumulation of amplified DNA is measured in each sample tube in real-time (i.e. once per PCR cycle) by monitoring the fluorescence emitted by sequence specific probes or generic DNA stains. All samples are assessed relative to standards containing known numbers of template molecules. Measurements taken while amplification is still proceeding in an exponential fashion, allow the number of templates in each sample at the beginning of the PCR to be calculated by reference to the standards.38 A recent gene expression study focused on genes with roles in cell cycle regulation, DNA repair, signaling pathways, and apoptosis, functions of great importance during early development.39 Although transcripts from every gene assessed were detectable at every stage, from mature oocyte to hatched blastocyst, the number of transcripts at different stages varied considerably. The quantity of mRNA transcripts was generally high in oocytes, but decreased dramatically after fertilization, in agreement with earlier observations.40–42 The data obtained clearly

indicate that transcripts from many genes reach extremely low levels in 2- and 3-cell embryos, in some cases scarcely above the threshold of detection.39 Depletion of maternal mRNA after fertilization appears to be a normal occurrence and yet may have great significance for the embryo. Most cellular pathways are controlled to some degree by the transcriptional activation or repression of specific genes. In other words, gene expression allows the subtle control of cellular mechanisms in response to changes in the intracellular and extracellular environment. It is likely that prior to the activation of the embryonic genome many important cellular pathways exist in a rigid form, controlled by a reservoir of proteins inherited from the oocyte. Consequently, the embryo may have limited ability to respond to environmental challenges during this period. This may have important implications for the practice of in vitro fertilization and embryo culture methods. No embryonic gene expression was detected until the 4-cell stage, consistent with previous data demonstrating the onset of gene activation.43,44 In most cases the increase in transcript number upon genome activation was modest, however, in a subset of embryos dramatic increases in expression were seen for the BRCA1 gene (several hundred fold).39 BRCA1 is a multifunctional protein with roles in cell cycle regulation and DNA repair. The increased expression of BRCA1 in cleavage stage embryos may indicate the presence of DNA damage. Such genetic damage could be derived from the sperm (as indicated by studies employing the sperm chromatin structure assay)45 or the oocyte. It is possible that DNA repair is relatively inefficient until the embryonic genome is activated and fresh gene expression permits the stimulation of DNA repair pathways. From the 10-cell to morula stages the activity of most genes examined stabilized or even underwent a small decline. Thereafter, expression levels rose proportionately with increasing cell number, and as a result, most genes displayed greater transcript numbers in blastocysts than at any other stage. The high gene activity in blastocysts is not surprising, as most of the genes assessed produce proteins that interact with DNA or chromosomes (i.e. the quantity of protein required is likely to be closely related 

HPE_Chapter23.qxp

7/13/2007

5:32 PM

Page 294


to the number of nuclei). Gene expression was clearly linked to developmental stage, with embryos at similar stages having very similar patterns of gene expression. The gene expression data were also analyzed using a hierarchical cluster algorithm, a mathematical method for identifying samples with similar characteristics. The results are displayed graphically as a dendogram, embryos with similar gene expression being placed on closely associated branches of a tree diagram. Cluster analysis grouped embryos of specific stages together, confirming statistically that patterns of gene expression and developmental stage are closely associated. Only 20% of embryos failed to cluster with counterparts of similar stage. Interestingly, more than half of these embryos displayed morphological abnormalities, suggesting that such aberrations may cause (or be caused by) perturbations of gene expression46 (Figure 23.2).

10

As well as separating embryos by developmental stage, cluster analysis was able to identify subgroups of identically staged embryos that had distinct expression profiles. Analysis of the morphological notes for the embryos within these subgroups revealed that one group was associated with the presence of granular cytoplasm, condensed organelles, and multiple nuclei – features that are negatively correlated with embryo implantation. A second group of 4–10-cell stage embryos was found to contain a preponderance of embryos with uneven cleavage divisions.46 Interestingly, not all the groups of embryos displaying a distinct pattern of gene activity were associated with a characteristic morphology. This demonstrates that differences in expression are frequently related to non-morphological factors, which could include factors such as the cell cycle, cellular stress, chromosomal status (e.g. aneuploid

10 1

8

2 8

6

6

4

4

2

2

0

0 A

B

C D

E

F

G H

I

J

A

10

B

C D

E

F

G H

I

J

10 3

8 6

6

4

4

2

2

0

4

8

0 A

B

C D

E

F

G H

I

J

A

B

C D

E

F

G H

I

J

Figure 23.2 Quantification of gene expression reveals that most embryos show an expression profile characteristic of their developmental stage. However, abnormalities can lead to disturbances in the normal pattern of gene expression. This figure shows the results of gene expression analysis for ten genes (A–J) in four blastocysts (1–4). Blastocysts 1, 2, and 3 displayed very similar patterns of gene expression and appeared to be morphologically identical (good morphology). However, analysis of blastocyst 4 revealed altered expression of several genes. This blastocyst had a poorly defined inner cell mass and minor growth retardation.



HPE_Chapter23.qxp

7/13/2007

5:32 PM

Page 295


versus euploid), or microenvironment of individual embryos. The potential of gene expression analysis to reveal ‘hidden’ information concerning processes occurring within the embryo may be of clinical value, as some of these processes are likely to have a bearing on viability. Further studies will continue to improve our understanding of the cellular processes occurring during the preimplantation phase. A recent advance has been the development of in vitro transcription techniques that permit the mRNA content of single oocytes to be amplified to levels sufficient for microarray analysis.47 Microarrays allow the simultaneous analysis of ⬎29 000 genes in a single experiment and have allowed us to catalogue over 9000 genes that are expressed in human oocytes (Wells et al, unpublished data). It is anticipated that the application of such methods to the study of human preimplantation development will greatly accelerate the progress of discovery in this area. Our analysis of human oocytes has already revealed several key cellular pathways that are differentially expressed when a chromosome abnormality is present. Not only does this open up new possibilities for the detection of aneuploidy, but also it reveals the cellular mechanisms underlying this problem. The knowledge gained from such studies may lead to the development of preventative interventions, such as improved ovarian stimulation regimens or optimized culture media.

PROTEOMIC ANALYSIS OF HUMAN EMBRYOS

Investigation of proteins may prove to be even more useful for assessing embryo viability than either cytogenetic or gene expression analysis. Although the gene expression data obtained provide an interesting insight into the genetic activity of the early embryo, a change in the number of mRNA transcripts derived from a given gene does not necessarily indicate altered utilization of the pathway in which it functions. Most genes experience some degree of regulation at the post-translational level, through protein modification, degradation, or sequestration. Consequently, there may be occasions when a

change in the concentration of active protein is not mirrored by an alteration in gene activity. Furthermore, individual genes usually produce more than one type of protein, accomplished by utilizing mechanisms such as alternative splicing and posttranslational modification. It is not possible to detect post-translational modifications by looking at gene expression. Advances in mass spectrometry have led to the development of techniques with sufficient sensitivity to permit the analysis of individual embryos. A recent study by Katz-Jaffe and colleagues utilized surface-enhanced laser desorption and ionization time-of-flight mass spectrometry (SELDI-TOF MS) to produce proteomic profiles for individual human blastocysts.48 The method involved the lysis of whole blastocysts and binding their proteins to chips. Bound proteins were released from the surface of the chip and ionized by laser activation and the mass : charge ratio of the liberated ions determined by time-of-flight mass spectrometry. The mass : charge ratios of the many different protein fragments produced during this process produce a unique proteomic fingerprint for each sample assessed. Analysis of individual blastocysts revealed differences in proteomic profiles related to morphology and developmental stage. For example, degenerating embryos displayed numerous alterations in protein expression when compared with developing blastocysts.48 Proteomic analysis also identified differences between embryos that were not correlated with either developmental stage or morphology. In some cases such changes may be related to altered utilization of specific cellular pathways that influence or reflect embryo viability. Analysis of the proteins involved may allow morphologically identical embryos to be distinguished in terms of their implantation potential.

ASSESSMENT OF OOCYTES AND EMBRYOS WITHOUT BLASTOMERE BIOPSY

Analyses of genes and proteins have demonstrated that dynamic fluctuations in expression occur throughout preimplantation development and that 

HPE_Chapter23.qxp

7/13/2007

5:32 PM

Page 296


in order to provide an indication of embryo viability. Such an approach has the potential to be infinitely more subtle than current PGS strategies, revealing a wealth of information concerning the cellular processes occurring in the embryo and, by inference, the health and well-being of the embryo. To date most gene and protein expression studies have focused on analysis of whole embryos and it is not yet clear whether similar results can be obtained

TP53 RB1 MAD2 BUB1 b-ACTIN APC BRCA2 ATM BRCA1

specific gene/protein expression profiles are characteristic of developmental stage and influenced by morphology. It is likely that embryos with patterns of expression appropriate for their developmental stage and without any perturbations associated with abnormal morphology have superior viability to those displaying atypical gene/protein profiles.46,48 In the future, it may be possible to assess patterns of gene or protein expression prior to embryo transfer

4–10 cells, disorganized, uneven cleavage

2–10 cell embroyos

Pre-genome activation 2-cells, 3-cells, early 4-cells

4–10-cell embryo with granular cytoplasm, condensed organelles and multinucleation

Morulae

Blastocysts (hatching)

Figure 23.3 Results of hierarchical cluster analysis, revealing embryos that have similar patterns of gene expression for the nine genes assessed during this study. Each line of nine colored squares represents the gene expression data from a single embryo. The color of each square represents the relative number of transcripts of a specific gene for a single embryo. Green ⫽ low expression; black ⫽ intermediate expression; red ⫽ high expression. Embryos of identical developmental stage have similar gene expression profiles and tend to cluster together. Additionally, embryos with certain abnormal morphologies are also grouped together, indicating that patterns of gene expression may be correlated with morphology. Reproduced from Wells et al., 2005 (Fertility & Sterility 84: 343–55) with the permission of The American Society for Reproductive Medicine.



HPE_Chapter23.qxp

7/13/2007

5:32 PM

Page 297


by analyzing single blastomeres. Unpublished data suggest that some genes display equal levels of expression in all blastomeres on day 3. Such genes could be suitable for preimplantation viability screening via blastomere biopsy; however, this is not the case for all genes, with many of the most interesting candidates for viability assessment showing wide variation in expression between different cells of the same embryo (Wells et al, unpublished data). For proteomic approaches a more useful strategy may prove to be detection of excreted or cell surface proteins, which can be analyzed without the need for blastomere biopsy. An attempt to move away from embryo biopsy should be considered when contemplating future methods of assessing embryo viability. Although blastomere biopsy has proven highly successful, permitting the development of PGD and PGS, indirect evidence suggests that the removal of a cell has a small but significant impact on embryo implantation potential.49 In order for biopsy-based methods such as PGS to be advantageous in terms of implantation rates, they must provide sufficient improvement in embryo selection to more than compensate for any reduction in viability caused by cell loss. The possibility that removal of a single blastomere may be detrimental increases the interest in alternative methods that avoid embryo biopsy. Assessment of cumulus cells in order to infer oocyte/embryo quality represents an attractive prospect for the future. Cumulus cells are intimately associated with the oocyte and have roles in oocyte support and maturation.50,51 The patterns of gene and protein expression in cumulus cells are likely to be influenced by processes occurring in the oocyte. Additionally, alterations in the follicular environment that affect oocyte competence are likely to induce characteristic changes in gene expression and cumulus cell biology. Thus, by studying RNA and protein from cumulus cells it may be possible to determine the quality of the oocyte and the follicle from which it was derived. Although this strategy only provides information on maternal factors affecting embryo viability, it has the advantage of utilizing cells that can be easily obtained without damage to the embryo.

Several studies have already demonstrated that alterations in the cumulus cell expression of specific genes can provide useful information, assisting the selection of oocytes and embryos for transfer.52–54 Data also exist indicating that the incidence of cumulus cell apoptosis influences fertilization rates, embryo development, and implantation.55,56 Studies aimed at using microarrays to provide a detailed characterization of gene expression in oocytes and cumulus cells are now beginning to yield results.57 The data produced are expected to greatly enhance the understanding of oocyte and follicle biology, shedding light on the cellular processes underlying maturation. Analyses of this type may also reveal the impact of different protocols of ovarian stimulation or in vitro maturation and assist in the optimization of such methods.

CONCLUSIONS

Thus far, few methods have been proven to assist in the identification of viable embryos. Among those techniques with convincing positive data are morphological assessment and, for a subset of patients, PGS. Undoubtedly, new and ever more complex schemes for assessing viability based upon morphological criteria will be proposed. However, advances in this area are likely to be subtle rather than dramatic. PGS methods will continue to be refined and the number of chromosomes examined will be expanded, encompassing the entire chromosome complement in the near future. Widespread availability of comprehensive chromosome screening is most likely to be achieved using a combination of CGH and microarray technology. With research progressing simultaneously on several fronts, the next few years are likely to witness a number of novel approaches for embryo viability assessment becoming widely available. These advances may involve quantification of gene expression, proteomic profiling, or analysis of cumulus cells. The insight provided by such approaches will prove invaluable as the number of embryos transferred per cycle is further reduced and we move closer to the goal of elective single embryo transfer. 

HPE_Chapter23.qxp

7/13/2007

5:32 PM

Page 298


REFERENCES 1. De Sutter P, Gerris J, Dhont M. A health-economic decision-analytic model comparing double with single embryo transfer in IVF/ICSI. Hum Reprod 2002; 15: 402–9. 2. Gerris J, De Sutter P, De Neubourg D et al. A real-life prospective health economic study of elective single embryo transfer versus twoembryo transfer in first IVF/ICSI cycles. Hum Reprod 2004; 19: 917–23. 3. Lukassen HG, Schonbeck Y, Adang EM et al. Cost analysis of singleton versus twin pregnancies after in vitro fertilization. Fertil Steril 2004; 81: 1240–6. 4. Nyboe Andersen A, Gianaroli L, Nygren KG. Assisted reproductive technology in Europe, 2000. Results generated from European registers by ESHRE. Hum Reprod 2004; 19: 490–503. 5. Giorgetti C, Terriou P, Auquier P et al. Embryo score to predict implantation after in-vitro fertilization: based on 957 single embryo transfers. Hum Reprod 1995; 10: 2427–31. 6. Ziebe S, Petersen K, Lindenberg S et al. Embryo morphology or cleavage stage: how to select the best embryos for transfer after in-vitro fertilization. Hum Reprod 1997; 12: 1545–9. 7. Rijinders PM, Jansen CAM. The predictive value of day 3 embryo morphology regarding blastocyst formation, pregnancy, and implantation rate after day 5 transfer following in vitro fertilization or intracytoplasmic sperm injection. Hum Reprod 1998; 13: 2869–73. 8. Van Royen E, Mangelschots K, De Neubourg D et al. Characterization of a top quality embryo, a step towards single-embryo transfer. Hum Reprod 1999; 14: 2345–9. 9. Milki AA, Hinckley MD, Gebhardt J et al. Accuracy of day 3 criteria for selecting the best embryos. Fertil Steril 2002; 77: 1191–5. 10. Scott L, Alvero R, Leondires M et al. The morphology of human pronuclear embryos is positively related to blastocyst development and implantation. Hum Reprod 2000; 15: 2394–403. 11. Neuber E, Mahutte NG, Arici A et al. Sequential embryo assessment outperforms investigator-driven morphological assessment at selecting a good quality blastocyst. Fertil Steril 2006; 85: 794–6. 12. Munne S, Lee A, Rosenwaks Z et al. Diagnosis of major chromosome aneuploidies in human preimplantation embryos. Hum Reprod 1993; 8: 2185–91. 13. Delhanty JD, Harper JC, Ao A et al. Multicolour FISH detects frequent chromosomal mosaicism and chaotic division in normal preimplantation embryos from fertile patients. Hum Genet 1997; 99: 755–60. 14. Laverge H, De Sutter P, Verschraegen-Spae MR et al. Triple colour fluorescent in-situ hybridization for chromosomes X, Y and 1 on spare human embryos. Hum Reprod 1997; 12: 809–14. 15. Iwarsson E, Lundqvist M, Inzunza J et al. A high degree of aneuploidy in frozen-thawed human preimplantation embryos. Hum Genet 1999; 104: 376–82. 16. Voullaire L, Slater H, Williamson R et al. Chromosome analysis of blastomeres from human embryos by using comparative genomic hybridization. Hum Genet 2000; 106: 210–7. 17. Wells D, Delhanty JD. Comprehensive chromosomal analysis of human preimplantation embryos using whole genome amplification and single cell comparative genomic hybridization. Mol Hum Reprod 2000; 6: 1055–62. 18. Sandalinas M, Sadowy S, Alikani M et al. Developmental ability of chromosomally abnormal human embryos to develop to the blastocyst stage. Hum Reprod 2001; 16: 1954–8. 19. Munne S. Preimplantation genetic diagnosis of numerical and structural chromosome abnormalities. Reprod Biomed Online 2002; 4: 183–96.



20. Boue J, Boue A, Lazar P. Retrospective and prospective epidemiological studies of 1500 karyotypes human spontaneous abortions. Teratology 1975; 12: 11–26. 21. Hassold T, Chen N, Funkhouser J et al. A cytogenetic study of 1000 spontaneous abortions. Ann Hum Genet 1980; 44: 151–78. 22. Menasha J, Levy B, Hirschhorn K et al. Incidence and spectrum of chromosome abnormalities in spontaneous abortions: new insights from a 12-year study. Genet Med 2005; 7: 251–63. 23. Gianaroli L, Magli MC, Ferraretti AP et al. Preimplantation diagnosis for aneuploidies in patients undergoing in vitro fertilization with a poor prognosis: identification of the categories for which it should be proposed. Fertil Steril 1999; 72: 837–44. 24. Munne S, Sandalinas M, Escudero T et al. Improved implantation after preimplantation genetic diagnosis of aneuploidy. Reprod Biomed Online 2003; 7: 91–7. 25. Platteau P, Staessen C, Michiels A et al. Preimplantation genetic diagnosis for aneuploidy screening in women older than 37 years. Fertil Steril 2005; 84: 319–24. 26. Kallioniemi A, Kallioniemi OP, Sudar D et al. Comparative genomic hybridization for molecular cytogenetic analysis of solid tumors. Science 1992; 258: 818–21. 27. Telenius H, Carter NP, Bebb CE et al. Degenerate oligonucleotideprimed PCR: general amplification of target DNA by a single degenerate primer. Genomics 1992; 13: 718–25. 28. Wells D, Sherlock JK, Handyside AH et al. Detailed chromosomal and molecular genetic analysis of single cells by whole genome amplification and comparative genomic hybridisation. Nucleic Acids Res 1999; 27: 1214–8. 29. Voullaire L, Wilton L, McBain J et al. Chromosome abnormalities identified by comparative genomic hybridization in embryos from women with repeated implantation failure. Mol Hum Reprod 2002; 8: 1035–41. 30. Wilton L, Voullaire L, Sargeant P et al. Preimplantation aneuploidy screening using comparative genomic hybridization or fluorescence in situ hybridization of embryos from patients with recurrent implantation failure. Fertil Steril 2003; 80: 860–8. 31. Bahce M, Cohen J, Munne S. Preimplantation genetic diagnosis of aneuploidy: were we looking at the wrong chromosomes? J Assist Reprod Genet 1999; 16: 176–81. 32. Munne S, Bahce M, Sandalinas M et al. Differences in chromosome susceptibility to aneuploidy and survival to first trimester. Reprod Biomed Online 2004; 8: 81–90. 33. Wilton L, Williamson R, McBain J et al. Birth of a healthy infant after preimplantation confirmation of euploidy by comparative genomic hybridization. N Engl J Med 2001; 345: 1537–41. 34. Wells D, Escudero T, Levy B et al. First clinical application of comparative genomic hybridization and polar body testing for preimplantation genetic diagnosis of aneuploidy. Fertil Steril 2002; 78: 543–9. 35. Wells D, Steuerwald N, Chu L et al. Microarrays for analysis and diagnosis of human embryos. In: Papp Z, Rodeck C, eds. Recent Advances in Prenatal Genetic Diagnosis. Bologna: Medimond, 2004: 9–17. 36. Hu DG, Webb G, Hussey N. Aneuploidy detection in single cells using DNA array-based comparative genomic hybridisation. Mol Hum Reprod 2004; 10: 283–9. 37. Le Caignec C, Spits C, Sermon K et al. Single-cell chromosomal imbalances detection by array CGH. Nucleic Acids Res 2006; 34: e68. 38. Wittwer CT, Herrmann MG, Moss AA et al. Continuous fluorescence monitoring of rapid cycle DNA amplification. BioTechniques 1997; 22: 130–8. 39. Wells D, Bermudez MG, Steuerwald N et al. Expression of genes regulating chromosome segregation, the cell cycle and apoptosis during

HPE_Chapter23.qxp

7/13/2007

5:32 PM

Page 299


40. 41. 42.

43.

44.

45.

46.

47.

48.

human preimplantation development. Hum Reprod 2005; 20: 1339–48. Bachvarova R, De Leon V. Polyadenylated RNA of mouse ova and loss of maternal RNA in early development. Dev Biol 1980; 74: 1–8. Clegg KB, Piko L. Poly(A) length, cytoplasmic adenylation and synthesis of poly(A)⫹ RNA in early mouse embryos. Dev Biol 1983; 95: 331–41. Telford NA, Watson AJ, Schultz GA. Transition from maternal to embryonic control in early mammalian development: a comparison of several species. Mol Reprod Dev 1990; 26: 90–100. Tesarik J, Kopecny V, Plachot M et al. Activation of nucleolar and extranucleolar RNA synthesis and changes in the ribosomal content of human embryos developing in vitro. J Reprod Fertil 1986; 78: 463–70. Braude P, Bolton V, Moore S. Human gene expression first occurs between the four- and eight-cell stages of preimplantation development. Nature 1988; 332: 459–61. Virro MR, Larson-Cook KL, Evenson DP. Sperm chromatin structure assay (SCSA) parameters are related to fertilization, blastocyst development, and ongoing pregnancy in in vitro fertilization and intracytoplasmic sperm injection cycles. Fertil Steril 2004; 81: 1289–95. Wells D, Bermudez MG, Steuerwald N et al. Association of abnormal morphology and altered gene expression in human preimplantation embryos. Fertil Steril 2005; 84: 343–55. Bermudez MG, Wells D, Malter H et al. Expression profiles of individual human oocytes using microarray technology. Reprod Biomed Online 2004; 8(3): 325–37. Katz-Jaffe MG, Gardner DK, Schoolcraft WB. Proteomic analysis of individual human embryos to identify novel biomarkers of development and viability. Fertil Steril 2006; 85: 101–7.

49. Cohen J, Wells D, Munne S. Removal of two cells from cleavage stage embryos is likely to reduce the efficacy of chromosomal tests employed to enhance implantation rates. Fertil Steril 2007; 87(3): 496–503. 50. Dekel N, Beers WH. Development of the rat oocyte in vitro: inhibition and induction of maturation in the presence or absence of the cumulus oophorus. Dev Biol 1980; 75: 247–54. 51. Larsen WJ, Wert SE, Brunner GD. A dramatic loss of cumulus cell gap junctions is correlated with germinal vesicle breakdown in rat oocytes. Dev Biol 1986; 113: 517–21. 52. Elvin JA, Clark AT, Wang P et al. Paracrine actions of growth differentiation factor-9 in the mammalian ovary. Mol Endocrinol 1999; 13: 1035–48. 53. Yan C, Wang P, DeMayo J et al. Synergistic roles of bone morphogenetic protein 15 and growth differentiation factor 9 in ovarian function. Mol Endocrinol 2001; 15: 854–66. 54. Zhang X, Jafari N, Barnes RB et al. Studies of gene expression in human cumulus cells indicate pentraxin 3 as a possible marker for oocyte quality. Fertil Steril 2005; 83(Suppl 1): 1169–79. 55. Corn CM, Hauser-Kronberger C, Moser M et al. Predictive value of cumulus cell apoptosis with regard to blastocyst development of corresponding gametes. Fertil Steril 2005; 84: 627–33. 56. Lee KS, Joo BS, Na YJ et al. Cumulus cells apoptosis as an indicator to predict the quality of oocytes and the outcome of IVF-ET. J Assist Reprod Genet 2001; 18: 490–8. 57. Assou S, Anahory T, Pantesco V et al. The human cumulus–oocyte complex gene-expression profile. Hum Reprod 2006; 21: 1705–19.



HPE_Chapter23.qxp

7/13/2007

5:32 PM

Page 300

HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 301

24. Oocyte selection in contemporary clinical IVF: do follicular markers of oocyte competence exist? Jonathan Van Blerkom and Susan W Trout

The adage ex ovo omni, ‘from the ovum all’, was first advanced by Harvey in the 17th century to explain the origin of life; this observation retains a measure of currency in the 21st century, as basic research findings and clinical outcomes indicate the extent to which the developmental competence of the embryo is already determined in the oocyte prior to fertilization.1–3 Two important issues related to the oocyte in contemporary infertility treatments, such as those using in vitro fertilization (IVF) are (1) whether noninvasive methods for follicular assessments can be sufficiently sensitive to distinguish between morphologically comparable oocytes with different developmental competence, and (2) when and how defects arise in the oocyte that have adverse effects on embryo development. The oocyte is unique among cells because it spends virtually all of its life in a ‘quiescent’ state, surrounded by an equally dormant layer of somatic cells (pregranulosa cells). The pregranulosa cells first associate with the oocyte at the so-called primordial stage, when oogenesis (primordial oocyte) and folliculogenesis (primordial follicle) begin in the human embryo around week 9 of gestation (for a review of the fine structural aspects of human oogenesis, see reference 4). Proteinaceous secretions from this somatic cell monolayer form a basal lamina that delineates each primordial follicle within the ovary. Indeed, when a woman ovulates, virtually all of the granulosa component (cumulus and mural granulosa) did not exist during the previous decades, but rather arose in two phases, from a very small number of morphologically indistinguishable progenitors: (1) after the spontaneous recruitment of a primordial follicle into a developmental pathway that

leads to ovulation, and, more significantly, (2) following stimulation by the pituitary gonadotropin, follicle stimulating hormone (FSH). The first phase can last several months, with the oocyte nucleus (germinal vesicle) in meiotic arrest from the primordial oocyte stage at the beginning of oogenesis. Over this period, the oocyte progressively grows to its terminal size (termed fully grown) in concert with several cell divisions of the pregranulosa monolayer, to form a multilayered structure in which the innermost layer (so-called corona radiata) resides on the surface of the zona pellucida. Bidirectional communication between the pregranulosa cells and the oocyte involves gap junctions that initially developed at the site of contact on the oocyte plasma membrane (oolemma), and after formation of the zona pellucida, is maintained by elongated cellular extensions (termed transzonal processes) of the coronal cells. In some transmission and scanning electron microscopic images of the early stages of human oogenesis, transzonal processes can be seen to penetrate or deeply invaginate into the oocyte cytoplasm and associate with mitochondria, Golgi bodies, and elements of the endoplasmic reticulum.4 It has been suggested that this complex interaction between germ and somatic cell(s) provides a structural basis for the maternal regulation of oogenesis.5 When the oocyte is fully grown and surrounded by multiple layers of granulosa cells, the follicle is designated ‘preantral’, and at this stage of folliculogenesis it is normally able to undergo rapid development in response to FSH. Over a relatively short period of exposure (days), FSH induces a profound numerical expansion (proliferation) of the granulosa 

HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 302


cell complement that occurs in concert with a progressive accumulation of fluid within the follicle, forming a space or antrum. Antrum formation physically separates the somatic cell component into two geographically distinct compartments: (1) the mural granulosa that lines the follicular wall and is largely involved in the production of the sex steroid, estrogen; and (2) the cumulus granulosa, that surrounds the oocyte and is largely responsible for the synthesis and secretion of a complex array of growth factors and other bioactive molecules (see below). The fully expanded antral follicle is termed a tertiary or Graafian follicle, and at this stage, is normally responsive to a second pituitary gonadotropin, luteinizing hormone (LH). LH is released rapidly (the so-called LH surge), and upon entrance into the follicular fluid initiates a co-ordinated series of cytoplasmic and nuclear activities in the oocyte. The term preovulatory maturation is generally used to describe these changes, and their timed expression is necessary for monospermic fertilization and the development of a competent diploid embryo.4 LH also induces changes in follicular biochemistry and morphology that lead to ovulation. In women, ovulation usually occurs some 36–40 hours after the initial surge of LH is detected. The terminal stages of folliculogenesis encompass the growth (follicular) and preovulatory (luteal) phases, which are regulated by FSH and LH, respectively. For infertility treatments that involve artificial (i.e. exogenous) management of the menstrual cycle (termed controlled ovarian hyperstimulation, COHS) to produce multiple Graafian follicles by timed administrations of gonadotropins, evaluation of the follicular phase usually involves both ultrasonography to measure intrafollicular fluid accumulation, indicated by a progressive increase in follicular diameter, and measurements of circulating levels of estradiol, which reflect the biosynthetic capacity of the mural granulosa in FSH-stimulated follicles. In spite of apparently normal follicular growth rates and estradiol levels, it has long been recognized that the meiotic maturity and developmental competence of each oocyte is unique when it is retrieved from a fully grown antral follicle. Meiotically mature (metaphase II stage, MII) oocytes are often burdened with chro

mosomal or cytoplasmic defects the adverse developmental influences of which become evident during the preimplantation stages by abnormal embryo development in vitro, or after uterine transfer, by implantation failure, or early post-implantation demise.1–3 Confirmation of this apparent ‘fact’ is one of the most significant and basic findings to come from human IVF, and one that continues to present a formidable barrier to the realization of the ‘holy grail’ of this enterprise: namely, the retrieval of a small number of MII oocytes with equally high developmental competence, such that the replacement of one or two fresh or thawed embryos has a high probability of producing a term pregnancy and normal offspring. Current notions of oocyte biology suggest that gametes largely begin life with a normal developmental potential. Yet, at the completion of preovulatory nuclear and cytoplasmic maturation, whether induced naturally by LH or by the administration of human chorionic gonadotropin (hCG) in cycles of COHS, developmental heterogeneity within cohorts is the rule. In clinical IVF, this is clearly evident by differences in embryo development in vitro during culture under identical conditions, and by negative or singleton outcomes after the transfer of multiple cleavage or expanded/hatched blastocyst stage embryos that were stage-appropriate and morphologically equivalent. The presence of certain chromosomal abnormalities in the oocyte can be particularly problematic for embryo selection because their development during the preimplantation stages is often indistinguishable from that of their normal siblings.6,7 The likelihood of a successful outcome in infertility treatments that use IVF is often considered in the context of oocyte quality with simple descriptors such as ‘good’ or ‘bad’ often applied to each oocyte, based on empirically derived cytoplasmic characteristics detectable at the light microscopic level. Although certain cytoplasmic features and defects are clearly and negatively related to outcome,8–11 others features suggested to indicate poor developmental competence for the oocyte may be more apparent than real9,12 if based on subtle differences in the transmission of light. In our experience, for example, the cytoplasm of oocytes with an apparently

HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 303

OOCYTE SELECTION IN CONTEMPORARY CLINICAL IVF

high lipid body content often appears dense, and owing to the refraction of light, to almost ‘sparkle’ when viewed under the dissecting microscope. While early blastomeres of embryos that develop from these oocytes have a similar appearance, normal births indicate that this phenotype is of no developmental significance. In this regard, these simple descriptors are not only ill-defined, but depending upon the empirical criteria used to characterize a cytoplasmic phenotype as defective or ‘dysmorphic’ (see below), they may also be potentially misleading if used to ‘deselect’ oocytes for insemination. By definition, a perfect oocyte is one that results in a developmentally viable embryo and normal offspring. While the ‘perfect oocyte’ may have a certain esthetic quality for the viewer that can bias operator selection for insemination, an ideal morphology indicative of normal competence seems unlikely to exist in the human. This notion is supported by clinical experience, where MII oocytes from the same cohort are subjected to close inspection during intracytoplasmic sperm injection (ICSI); despite evident morphological heterogeneity, equivalent competence is demonstrated by successful outcomes in the current cycle and in subsequent transfers involving cryopreserved embryos. Chromosomal (aneuploidy) and certain common cytoplasmic defects (commonly termed ‘dysmorphisms’,8,9 of developmental significance for the embryo may accumulate in the oocyte over time, or may develop de novo at a particular stage of oogenesis. For example, numerical chromosomal defects (aneuploidy) that are first detectable by conventional cytogenetic methods during the terminal stages of preovulatory maturation (MI, MII) have often been considered natural and unavoidable consequences of reproductive (i.e. maternal) aging in our species.13 This view can lead to an oocyte-centric interpretation of how normal competence is established: if biological and chronological aging of the female gamete are largely synonymous, elevated frequencies of ‘bad eggs’ (e.g. aneuploid) would be expected with age because the underlying defects have unavoidably developed with age and therefore already exist in the fully grown, albeit immature oocyte. Numerical and structural (e.g. premature chromatid

separation) defects that affect chromosomes would be expected to manifest during MI and MII of preovulatory meiotic maturation because it is at this stage that bivalent and monovalent chromosomes, respectively, first form. Given the complex cellular and molecular biology of the follicle during the follicular and preovulatory phases (see below), the developmental heterogeneity of MII oocytes expressed after fertilization by their differential performance in vitro and outcome in vivo may also reflect follicle-specific (i.e. intrafollicular) conditions that arose de novo during the follicular phase and negatively impacted the competence of oocytes that were developmentally competent prior to gonadotropin-induced follicular growth. This notion is especially relevant for the human, where the time-line of folliculogenesis, beginning during fetal life, can be separated by over four decades from its terminal stage, the preovulatory Graafian follicle in the adult. This prolonged period is one of the biological bases for an oocytecentric view of an age-related origin of developmental defects in the female gamete.13 However, effects of biological and chronological aging on oocyte competence can also be seen in young infertile women who show age-inappropriate biological aging of their gametes in the presence of raised FSH levels.14 Therefore, a question central to understanding the origins of normal competence for the oocyte is the extent to which defects that are first detectable during the preovulatory stage arise de novo, or whether they pre-exist and are age-related. If common defects that adversely affect competence pre-exist in the immature oocyte, it seems unlikely that different analytical measures of follicular biology will be universally informative at MII for selection in IVF. In contrast, if they arise during the follicular phase, then the identification of parameters of normalcy that can be detected non-invasively and validated by outcome would provide a means by which the relative competence of an oocyte could be assessed prior to retrieval. The availability of such information, prior to ovulation induction or follicular aspiration, would begin to approach the ‘holy grail’ of selection for IVF, especially when the biochemical or physiological bases of empirically measured characteristics are 

HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 304


identified. However, the selection of parameters must meet the following criteria: they need to be measurable before or after ovum retrieval, by methods that provide timely results, based on outcome, and which can be easily incorporated in the clinical IVF laboratory; and clearly shown to be informative in the prediction of competence. The studies discussed here generally support the notion that such parameters exist, are detectable by non-invasive methods, and their application could classify follicles as high or low with respect to the prediction of competence in the corresponding oocyte. Indeed, this line of investigation has become increasingly relevant in circumstances where the number of oocytes that can be inseminated or embryos transferred is restricted by protocol or law. Therefore, the intent of this chapter is to review certain developmental and biochemical characteristics of human ovarian follicles that currently meet these criteria.

CLINICAL PARAMETERS OF FOLLICULAR DEVELOPMENT

In clinical practice, follicular development is currently monitored by conventional ultrasonography, and the diameter of each follicle is measured at varying intervals, according to program-specific protocol. The timing of ovulation, which is usually induced by the administration of hCG, is most often determined by circulating levels of estradiol and the relative number of follicles within a cohort that have attained a threshold diameter, usually, ⱖ18 mm. However, based on oocyte meiotic maturity at retrieval, fertilizability by conventional IVF or ICSI, embryo development in vitro and outcome after transfer, clinical experience indicates that follicular size at ovulation induction or aspiration is not a definitive indicator of competence for the female gamete. Indeed, some of the earliest studies in clinical IVF correlated fertilization rates, embryo performance in vitro and outcome after transfer with follicular diameters measured at ovulation using different protocols of COHS,15 and scores of studies in this regard have been published to date.16–23 A consensus conclusion from this wealth of information 

indicates the following: (1) within cohorts of follicles stimulated to grow by COHS, oocytes from smaller follicles are more likely to be immature than their counterparts from larger follicles; (2) if mature at aspiration, ICSI is more likely to achieve fertilization than conventional IVF; (3) normal outcomes with mature oocytes from small follicles demonstrate their competence; (4) the completion of maturation in vitro may be beneficial but has an attendant risk of generating aneuploid embryos at high frequency; and (5) in clinical practice, delaying ovulation induction until several follicles are in the size range of 18–20 mm improves outcome. Our clinical experience is quite similar in this regard: MII oocytes aspirated from smaller follicles are pooled and inseminated separately from their larger counterparts, but they are not discarded, and often produce embryos that ultimately result in successful outcomes in both fresh and cryopreserved/thawed transfers. However, the demonstrable differential competence of MII oocytes aspirated from stage- and size-appropriate follicles that displayed similar growth characteristics during the follicular phase supports the longstanding assumption that, with respect to possible influences on oocyte competence, each follicle is unique. The biochemical composition of follicular fluid is a highly complex mix of potent growth factors, cytokines, steroid hormones, and other bioactive molecules of extrinsic (blood borne) and intrinsic origin (secreted by mural and cumulus granulosa cells). Against this complex and follicle-specific background, the identification, characterization and quantification of specific molecules the levels of which may be informative with respect to oocyte competence presents formidable logistic and analytical challenges in assisted reproductive medicine. For validation in the selection of oocytes or embryos in clinical IVF, for example, studies of follicular chemistry and physiology should satisfy the following conditions: (1) information needs to be derived from a large and diverse population of women of different ages and etiologies of infertility; (2) certain lifestyle factors known to adversely affect fertility need to be considered (see below); (3) individual follicles characterized ultrasonically during the follicular phase have to be identifiable at aspiration;

HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 305


and (4) the meiotic status (maturity) and cytoplasmic morphology of the corresponding oocyte must be evaluated by a standard light microscopic protocol. After insemination, the occurrence of fertilization by conventional IVF or ICSI must be recorded, and the timing and pattern of cell divisions, the nuclear status of each blastomere (mono-, multi-, micronucleated) and their cellular characteristics (e.g. fragmentation pattern and extent) must be systematically monitored for each embryo associated with a defined follicle. This is a formidable undertaking for any IVF program, and is often complicated by the occurrence of so many follicles, oocytes, and embryos as to be impractical under routine circumstances. However, certain follicular characteristics detectable by power Doppler ultrasonography (described in detail below) can facilitate the classification of follicles and permit the corresponding oocytes to be grouped accordingly. For most of the nearly 30 years in which clinical IVF has been used to treat infertility, a ‘let nature take its course’ approach has guided decision making in terms of the number of embryos transferred when a cohort is ‘mixed’ with respect to stage or morphology, or both. With improvements in culture methodology and the availability of highly experienced personnel in the IVF laboratory, high frequencies of multiple fetal gestations became commonplace, making the multiple embryo transfer approach untenable in general,24 or illegal in certain countries where by law, the number of oocytes that can be inseminated or embryos transferred is proscribed.25 In some countries, the imposition of this new paradigm renewed interest in the identification of follicular characteristics, physiology, and biochemistry that could be used to select oocytes for insemination. This has become particularly relevant where current restrictions prohibit the prolongation of embryo culture to the blastocyst stage (day 5 or 6), which eliminates the ability to select embryos at the terminal phase of the preimplantation period based on developmental performance and stage-appropriate morphology,25 and genetic diagnosis of single cells at any preimplantation stage.27,28 Correlations with outcome can be unreliable, if not uninformative when transfers involve multiple

embryos, and therefore tracking the fate of individual oocytes is less problematic when single (sET) or double (dET) embryo transfers are mandated. At present, a comparatively high degree of confidence in selection protocols can be derived from certain ultrasonic and biochemical characteristics that have been associated with positive outcomes. While still a work in progress, it is already evident that, when used in combination, the following methodologies can and do positively bias the decision-making process. In the following discussion, these methodologies are divided into two categories that involve non-invasive measurements of follicular characteristics that have been associated with oocyte and embryo developmental competence: perifollicular blood flow velocities and patterns; and follicular fluid levels of oxygen and certain growth factors.

DOPPLER ULTRASONOGRAPHY

For evaluations of follicular development under gonadotropin stimulation, conventional pulse-echo ultrasonography provides static, gray-scale images from which follicular dimensions can be calculated; however, it is largely uninformative with respect to other dynamic aspects that may be follicle specific and competence related. Color Doppler ultrasonography was first introduced over a decade ago in clinical IVF to assess degrees and spatial patterns of blood flow in cycles of COHS as related to the ovarian response to stimulation29 and optimal uterine conditions before embryo transfer.30 Initial studies of perifollicular blood flow characteristics showed that follicle-specific velocity was related to oocyte recovery rate, meiotic maturity, preimplantation embryo quality, and early suggestions of more positive outcomes when oocytes were derived from high flow follicles.31–33 Subsequent studies that examined Doppler indices in larger numbers of women undergoing IVF came to the following basic conclusions: perifollicular blood flow characteristics are follicle specific, and in cycles of COHS, fully grown follicles on the same ovary that exhibit similar rates of growth during the follicular phase frequently exhibit very different blood flow patterns and velocities; and 

HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 306


perifollicular blood flow and oocyte competence were related as indicated by embryo development in vitro and outcome after transfer.34–39,40,41 In two large-scale studies in 1997, Chiu et al34 and Van Blerkom et al35 characterized individual perifollicular blood flow rates and the proportion of the follicular circumference in which flow was detected in several hundred individual follicles. Both studies came to similar conclusions: perifollicular velocity was follicle specific; for the ‘typical’ infertile patient undergoing COHS for IVF, follicles with moderateto-high flow indices were often in the minority; and MII oocytes aspirated from comparatively high flow follicles had a significantly higher level of competence when compared with siblings derived from follicles with perifollicular flow that was low-toundetectable, or involved only a portion of the circumference of the follicle. Van Blerkom et al35 examined the ploidy of MII oocytes derived from high and low flow follicles and reported that a significantly higher frequency of aneuploidy (monosomy and trisomy) occurred in oocytes aspirated from fully grown follicles with low or no detectable perifollicular blood flow, and that after fertilization, oocytes recovered from such follicles showed poor performance in vitro, as indicated by high frequencies of asymmetric cleavages, toxic levels of fragmentation and premature arrest. Other studies have confirmed the follicle-specific nature of perifollicular velocity and the significant improvement in outcome when oocytes aspirated from high flow follicles are identified and selected for transfer as embryos.36–41 Independent support for IVF outcome findings was reported in a study of over 700 women undergoing COHS and intrauterine insemination.42 Two conclusions applicable for decision making in infertility treatment can be derived from these findings.40 First, for IVF, treatment should be halted when perifollicular flow characteristics indicate that few, if any, Graafian follicles have developed a perifollicular vascular network (during the follicular phase) that can support high velocity blood flow. Second, for intrauterine insemination (IUI), treatment should be halted when multiple fully grown follicles display high flow characteristics, owing to the increased probability that the induced 

ovulation will produce multiple high-competence oocytes and multifetal pregnancies. Measurements of perifollicular blood flow in cycles of COHS and IVF are made on each day during the final stages of the follicular phase, or in most programs, only once on the day of ovulation induction. This provides an overview of how many follicles in high, moderate, or low flow categories exist, and can be used in planning the retrieval such that the number and location of follicles in each class can be determined prior to aspiration. This protocol is of particular relevance when numerous follicles occur and the oocytes are aspirated under conventional ultrasound guidance, in a sequence that is based only on their proximity to each other. Even when there are large numbers of follicles, oocytes that are distinguished by their follicular Doppler ultrasound characteristics can be pooled separately, and their performance after IVF followed without adding significant complexity to the laboratory routine of embryo inspection. Figures 24.1A–D show representative color pulsed Doppler images taken at the time of ovulation induction (following COHS for IVF) and presented here in gray-scale. Typically, differential perifollicular blood flow rates are seen between follicles on the same ovary, and in these cross-sectional views, high flow at specific regions around the follicular wall appears in white (see arrows in Figure 24.1A–C). In contrast, follicles with low perifollicular blood flow show either focal regions of detectable flow (asterisk in Figure 24.1B) or no detectable signal (asterisks in Figure 24.1D). The presence, absence, or focal nature of a Doppler signal is confirmed for each follicle by rotating the ultrasound probe to survey the entire detectable perimeter of the follicular wall. Collectively, results from the above studies of hundreds of stimulated cycles are consistent with respect to oocyte selection in IVF, as follows: (1) the extent and velocity of perifollicular blood flow are follicle specific, and adjacent follicles of comparable size can display very different characteristics that are not predicted by conventional ultrasonography; (2) the number of high flow follicles can be ovary specific, with most or all residing on one ovary; (3) with the same protocol of COHS, the number of ‘high flow’

HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 307


-1 -2 CPA

*

-3

A

B

-4

* * C

D

C N N

C E

F

Figure 24.1 (A)–(D) Pulsed Doppler ultrasound images of fully grown human follicles in four women after controlled ovarian hyperstimulation for in vitro fertilization. Although the original follicular scans involve the entire circumference of the follicular wall, they are shown here in midline. The original color images that depict differential perifollicular blood flow velocity were transformed into gray-scale with the most intense flow shown in white and noted by arrows. Follicles characterized by a pattern of perifollicular blood flow that was of low velocity, or that was discontinuous along the follicular wall when viewed in three dimensions, are indicated by asterisks (see text for details). (E) and (F) Immunofluorescent images of a 24 hour human cumulus granulosa cell culture showing translocation of the hypoxia-inducible transcription factor HIF1␣ from the cytoplasm ‘C’, in (E) to the nucleus ‘N’, in (F), within 15 minutes of transfer from normal culture conditions (95% air, 5% CO2) to medium pre-equilibrated in an atmosphere containing 1% O2 (see text for details).

follicles is cycle specific and does not necessarily improve by altering stimulation regimen;41 and (4) some infertile patients and women of advanced reproductive age repeatedly fail to develop any moderate-to-high flow follicles, regardless of stimulation protocol. In cases where there is a reduced probability of competent gametes being aspirated or spontaneously ovulated, with a past history of repeated IUI or IVF failure, Doppler ultrasonographic evidence could be used to determine whether treat-

ment should be continued, or whether alternatives with higher probability outcomes should be considered, such as IVF with donor oocytes. This notion was recently supported by Shrestha et al,43 who conducted a retrospective longitudinal study of perifollicular blood flow rates during the follicular phase of COHS in IVF cycles. Patient responses could be classified into two groups based on sequential power Doppler ultrasonographic assessments: the so-called ‘good beginners’, defined by cycles in 

HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 308


which at least one small or medium size follicle(s) exhibited high grade blood flow rates in the early follicular phase; and the so-called ‘poor beginners’, distinguished by the absence of moderate-to-high grade flow in any small or medium size follicles during the early follicular phase. Although individual oocytes and transferred embryos were not correlated with specific follicles, the authors reported that perifollicular blood flow rates in the early follicular phase were associated with graded follicle-specific flow rates at ovulation induction and with outcome after embryo transfer in IVF. In both the early and late follicular phase, clinical pregnancy rates were significantly higher in the ‘good beginner’ group compared with those classified as ‘poor beginners’, and notably, this included relatively older patients (median age 40 years) who achieved a comparatively high pregnancy rate (37%) and low pregnancy loss rate (11%). The pattern of ‘good beginners’ was so distinct that Shrestha et al43 proposed that it may be possible to use Doppler analysis of perifollicular blood flow patterns to identify good and poor prognosis IVF cycles at an early stage of COHS; if the pattern is suboptimal, this could provide a biological justification for early cancellation. In contrast, Palomba et al44 reported that power Doppler assessments do not seem to be clinically useful for oocyte selection in IVF cycles for young non-obese infertile women with low basal FSH levels. In this study, oocytes were chosen for insemination based on perifollicular vascularity, and the outcome was compared with a control group where ‘standard’ morphological criteria alone were used to select oocytes for insemination and embryos for transfer. While this study confirmed previous findings that poor perifollicular flow characteristics and poor embryo performance in vitro were closely related,35 this method of follicular analysis was problematic for these investigators, due to the higher number of oocytes with optimal quality (and vascularization) that were available for insemination. As they noted, ‘in almost all cases’ the embryos transferred had developed from oocytes obtained from follicles with high-grade vascularization, which led to the suggestion that more refined ultrasound criteria may be required to distinguish 

between high-grade (i.e. high velocity) follicles if this is to be used for the assessment of oocyte and embryo competence. Merce et al45 offers a more sanguine conclusion concerning the clinical usefulness of follicular imaging, by demonstrating that three dimensional (3D) ultrasonography combined with power Doppler angiography on the day of ovulation induction predicted gestation in 76% of IVF patients. As they note, 3D Doppler angiography depicts all the ovarian and follicular blood vessels and can be used to derive a vascularization index and a blood flow index from the whole ovarian volume, as well as highresolution vascular and flow imaging of individual follicles. With this higher resolution approach to Doppler imaging of stimulated ovaries, Merce et al45 reported a pregnancy rate of 30% when no embryos from high-grade (i.e. flow) follicles were transferred, but the miscarriage rate was 67%. In contrast, transfer of a single embryo from a high-grade follicle was associated with a pregnancy rate of 35% and a miscarriage rate of zero. Whether the inclusion of 3D power Doppler angiography provides the ability to discriminate between follicles classified as high grade by 2- and 3D color pulsed Doppler ultrasonography, as described above, remains to be determined. However, the very high miscarriage rate observed by Merce et al45 with embryos derived from low grade follicles is consistent with the very high frequency of aneuploidy reported by Van Blerkom et al35 for oocytes aspirated from such (low flow) follicles. The preponderance of evidence from color, power, and most recently 3D Doppler imaging of ovary and follicle-specific perifollicular blood flow and vascularity demonstrates a clear association with developmental competence and outcome after IVF. The clinical usefulness of this protocol has been demonstrated with patients classified as ‘poor beginners’,43 or at the terminal stage of the follicular phase in IUI42 and IVF cycles,34–41,43–46 when only low-grade follicles (low-to-undetectable perifollicular flow) are detectable. Taken together, these outcome-based findings support a conclusion that was advanced by Chui et al34 and Van Blerkom et al35 nearly a decade ago; namely, that perifollicular blood flow rates are a very positive non-invasive indicator of oocyte and embryo

HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 309


competence. While not absolute, Doppler imaging should have a central role in the decision making process as to whether a treatment cycle should be continued or stopped, and for some patients, whether alternatives to assisted reproduction with patient oocytes need to be considered. At present, the utility of Doppler ultrasound measurements is controversial for the so-called ‘poor responder’ class of infertility patients, i.e. women who develop comparatively few antral follicles and have relatively low serum estradiol levels after COHS. Pan et al47 suggested that the reduced sensitivity of poor responders to gonadotropin stimulation may be associated with changes in ovarian stromal blood flow detectable by 3D power Doppler ultrasonography. These investigators reported marked differences in blood flow patterns and velocities between normoresponders and poor responders that were reflected by their outcome after IVF. Because estrogen is known to bind to its receptors on vessels and initiate the release of nitric oxide, which increases vasodilation and blood flow,48,49 Pan et al47 proposed that an estrogen deficit in poor responders may be one of the factors contributing to significantly reduced rates of perifollicular blood flow. In contrast, Kan et al50 reported no significant differences in implantation, clinical pregnancy, and live birth rates among poor responders, with and without high-grade perifollicular vascularity; patients in the former category tended to have higher multiple and live birth rates, and a lower frequency of miscarriage. However, too few women were included in this study to achieve statistical significance, and no correlations between specific follicles, oocytes, and embryos were made. The poor responder is not only common in assisted reproduction, but is also one that is often difficult to treat because of the high rate of recurrence despite different protocols of endocrine management. Additional Doppler studies of the poor responder are warranted, and, in particular, may determine if the presence of a single high-grade follicle could be grounds for a measure of optimism with respect to outcome. While the above evidence clearly shows the importance of ovarian and follicular imaging in contemporary assisted reproduction in general, and in

clinical IVF in particular, a question that remains to be answered is whether this technology should be extended to fertile women who may be at high risk for ovulating aneuploid MII oocytes. These patients include women of advanced reproductive age and those who chronically miscarry. It might also include women of any age with severe anxiety owing to a previous conception that was aneuploid. Doppler imaging of ovaries in natural menstrual cycles may be of significant value in this regard, if it can be shown to identify cycles in which conception should be avoided. Validation for its application in these instances will require confirmation that the occurrence of low-grade follicles is associated with an unacceptably high probability that the corresponding oocyte is chromosomally abnormal.

MECHANISMS OF REGULATION OF INTRAFOLLICULAR OXYGEN AND PERIFOLLICULAR BLOOD FLOW

Although Doppler ultrasonography and power angiography can identify perifollicular blood flow characteristics and vascularization, that are unique to each follicle, these imaging and analytical methods should not be considered to be absolute predictors of competence. Oocytes with chromosomal defects and embryos that prematurely arrest or develop abnormally in vitro can originate from follicle(s) judged as high grade,35,40,42 albeit at a much lower frequency than their low grade/flow counterparts. The relationship between high competence for the oocyte and perifollicular blood flow is also not understood. However, when the dissolved oxygen content of follicular fluid was measured at ovum retrieval, high frequencies of aneuploidy were detected in oocytes that originated from poorly oxygenated follicles that also showed poor flow characteristics during the follicular phase.35 A preliminary study of the average intracellular pH (pHi) in MII oocytes of normal appearance obtained from follicles on the same ovary showed that pHi levels (~6.8–6.9) in oocytes derived from low flow follicles with a dissolved oxygen content ⱕ1%, were below those measured in sibling oocytes (~7.2) that had been 

HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 310


aspirated from high flow follicles with dissolved oxygen contents ⱖ3%.35,51 Although these early findings require confirmation, they suggest that a slight reduction in pHi could result from inadequate cytoplasmic buffering capacity or inability to effectively metabolize lactic acid. Gaulden52 proposed that slight reductions in human oocyte pHi could retard or perturb the normal dynamics of microtubule polymerization/depolymerization, leading to asymmetric and malformed meiotic spindles with a reduced ability to capture, retain, or normally segregate chromosomes. In her morphological studies of Graafian follicles in women of advanced reproductive age (who were more prone to have children with Down’s syndrome), she observed that the perifollicular capillary bed in older women was underdeveloped compared with the situation in younger women, and that it tended to be further from the approximate center of the follicle. This too requires confirmation, but the suggestion is supported by coincident immunofluorescent studies of spindle microtubules and DNA fluorescent staining of MII chromosomes in MII oocytes obtained from fully grown follicles with poor blood flow characteristics and low intrafollicular oxygenation. These studies demonstrated abnormalities in spindle organization and morphology, and high frequencies of chromosomal malalignment and displacement, respectively.35,41 In a study of naturally cycling women ⱖ40 years of age, Battaglia et al53 reported that about 80% of MII oocytes of normal appearance after aspiration from single dominant follicles showed spindle abnormalities that would be likely to result in aneuploid embryos if fertilized. The apparent physiological relationship between oocyte pHi and levels of intrafollicular oxygen needs to be investigated in detail, especially with respect to metabolic pathways that may contribute to the acidification of the ooplasm in a low oxygen environment. In this regard, a recent mathematical modeling of oxygen concentrations in bovine and murine cumulus–oocyte complexes concluded that the cumulus cells probably remove little oxygen, suggesting that the concentration of dissolved oxygen measured in follicular fluids reflects what is experienced by the oocyte itself.54 If confirmed by experimental studies, follicle

specific differences in oxygen content measured in human follicular aspirates at ovum retrieval35,38 could reflect the unique capacity of each follicle to respond to hypoxia41 and upregulate angiogenesis in the perifollicular capillary bed, as discussed below. For the corresponding oocyte, these differences may have development consequences for competence at the molecular, cellular, and nuclear levels.35,51 The above findings suggest a physiological basis for compromised oocyte competence that correlates with follicle-specific information obtained by noninvasive Doppler studies that can be undertaken routinely during the follicular phase. The results obtained to date strongly suggest that MII oocytes from moderate-to-high flow follicles have a significantly higher bias for normal competence than do those obtained from follicles with low or no detectable perifollicular flow. Given the reported correlations between outcome and oocyte/embryo selection schemes that have incorporated perifollicular blood flow characteristics, it is curious that Doppler imaging has not become widespread in contemporary clinical IVF. For many programs, the added expense, time, and operator expertise associated with real-time Doppler ultrasound imaging could be prohibitory. In other instances, current laboratory paradigms of selection that include embryo morphology and performance are satisfactory, and the potential benefits of follicular monitoring are not deemed sufficient to warrant change. However, recent restrictions on the number of oocytes that can be inseminated or embryos transferred, without the option of cryopreservation or preimplantation genetic diagnosis, have placed greater importance on oocyte selection and the need to improve the sensitivity of the criteria used. As a result, the recent re-investigations of Doppler ultrasound assessments of perifollicular blood flow have largely validated previous findings with respect to the occurrence and distribution of high and low flow follicles, and more importantly, have been sufficiently related to outcome, on a per oocyte basis, to seriously warrant their inclusion in selection schemes for IVF. The use of this non-invasive technology in clinical IVF may be encouraged by enhancements in resolution and imaging capability incorporated into newer generation ultrasound instruments, including

HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 311


those specific for gynecological use that offer views of perifollicular blood flow and vascularity in three dimensions. While the above findings support the notion that rates of perifollicular blood flow reflect differences in the underlying biology of the follicle that appear to be of direct relevance for the developmental competence of the oocyte, several important questions remain. For example, if the velocity of perifollicular blood flow indicates normal follicular function, why do rates differ between follicles, and most importantly, what specific intrafollicular influences could have a positive or negative impact on this activity? Since the follicle is both a source of, and a ‘sink’ for a wide variety of bioactive molecules and growth factors, the rate of transfollicular trafficking between the systemic circulation and the intrafollicular milieu might be higher in follicles with a well-developed perifollicular microvasculature, consistent with a comparably high velocity flow. This could explain the association between higher levels of intrafollicular oxygenation and perifollicular blood flow rates. However, it should be emphasized that expansion of human perifollicular microvasculature under gonadotropin stimulation has not been demonstrated directly, but rather, inferred from upregulated blood flow rates54 and associations with the dissolved oxygen content of preovulatory follicular fluid. In this regard, increased blood flow resulting from the increased patency of pre-existing perifollicular vessels could be a direct effect of estrogen synthesized by the mural granulosa, a cellular layer that is usually separated from the capillary bed by the follicular wall, and a space that is measured in microns.41 Increased vascularization of the existing bed by angiogenic induction is another means by which perifollicular blood flow and capillary permeability may be significantly increased over a relatively short period of time during follicular stimulation, as discussed below.

ANGIOGENIC FACTORS IN FOLLICULAR FLUID

Human follicular fluid is a ‘factorologists’ dream owing to the wide range of growth and other regulatory molecules that are synthesized by follicular

granulosa cells or which enter the follicle from external sources (for review see references 56–58). With respect to perifollicular blood flow, however, the production of potent angiogenic factors by granulosa cells, such as vascular endothelial growth factor (VEGF) and leptin may be particularly relevant in understanding follicle-specific blood flow rates.58–63 Analysis of these angiogenic factors in human follicular fluid has shown that levels are follicle specific and has led to the notion that their export may induce a rapid expansion of the perifollicular vasculature by stimulating angiogenesis from existing endothelial cells.35,38 VEGF is also expressed by thecal cells, and this has a proximal source which could stimulate the expansion of perifollicular endothelial cells.64 VEGF expression is regulated by hypoxia in general65 and by the hypoxia inducible transcription factor (HIF-1) pathway in particular.66,67 This pathway can in turn be upregulated by certain reactive oxygen species normally produced by mitochondria (e.g. superoxide),68 which are the organelles involved in sensing oxygen at the cellular level.69 The possibility that the oxygen sensing ability of mural and cumulus granulosa cell mitochondria is upregulated during gonadotropin (FSH) stimulation can be tested.41 According to this hypothesis, diffusion of oxygen into the developing follicle from the existing perifollicular capillary bed is first ‘sensed’ by mitochondria in the gonadotropin-stimulated mural granulosa layer, which responds by increasing levels of superoxide that is converted to peroxide within the cytosol. Bell et al68 proposed that peroxide upregulates HIF-1 expression, which in turn upregulates the expression of a cohort of regulatory genes, including VEGF. Indeed, the expression of this transcription factor has been detected in mural and cumulus granulosa cells in human preovulatory follicles, and differences in mRNA levels between high and low flow follicles on the same ovary have been reported.41 The HIF-1 pathway is commonly used by cells that recognize and respond to hypoxia by upregulating levels of angiogenic factor production. It seems likely that such a mechanism may be involved in increasing perifollicular vascularity or the patency of the existing vasculature, or both, in order to increase intrafollicular oxygen. 

HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 312


The speed with which HIF-1␣ translocates from the cytoplasm to the nucleus in human cumulus cells cultured in a hypoxic environment is illustrated in Figure 24.1E and F. Figure 24.1E shows the typical pattern of diffused cytoplasmic HIF-1␣ immunostaining in cumulus cells cultured on glass coverslips in a standard atmosphere of 95% air 5% CO2, 24 hours after retrieval from a high flow follicle. Within 15 minutes following transfer to medium that had been pre-equilibrated in an atmosphere containing 1% O2 (Figure 24.1F), intense nuclear HIF-1␣ immunofluorescence is detected, with virtually no signal evident in the cytoplasm. A similar response is observed with granulosa cells derived from the mural layer, and from cells obtained from follicles with low or poor perifollicular circulation (Antczak and Van Blerkom, unpublished). When returned to normal atmospheric conditions, cytoplasmic HIF1␣ immunofluorescence becomes pronounced as the nuclear signal diminishes. These findings confirm previous molecular analyses demonstrating the presence of HIF-1 mRNA in human granulosa cells33 and provide direct evidence for the expression of a hypoxia-sensing mechanism in the human ovarian follicle.70 The association between follicle-specific differences in perifollicular blood flow, the dissolved oxygen content of follicular fluid, and levels of angiogenic factor biosynthesis that may be regulated by the HIF pathway warrants further investigation. At present, it is an intriguing possibility that failure to express HIF-1␣, or an inability to translocate it to the nucleus, may contribute to the occurrence of developmentally incompetent oocytes in women who rarely, if ever, develop follicles with high perifollicular blood flow characteristics. Whether age-associated mitochondrial dysfunctions compromise their ability to sense or respond to hypoxia also warrants further study. Ovarian pathologies, such as ovarian hyperstimulation syndrome (OHSS), are associated with intrafollicular levels of angiogenic promoters (e.g. VEGF) that are often significantly higher than levels measured in the follicular fluids of normal women undergoing COHS.71 The inability to downregulate angiogenic factor expression can lead to excessive perifollicular



vascular permeability, which results in capillary leakage and the third-spacing of fluids that is characteristic of this abnormal response to gonadotropin stimulation, especially in women with polycystic ovarian syndrome. Unusually high VEGF levels have been measured in the follicular fluids of young infertility patients (non-PCOS) and women of advanced reproductive age who showed no indications of OHSS during or after COHS for IVF.72 While the etiology of this atypical expression pattern is unknown, it may result from an angiogenic receptor defect that renders perifollicular endothelial cells refractory to this promoter(s) and consequently, the affected follicle(s) may be unable to upregulate oxygenation or downregulate HIF-1 activity (and the attendant expression of other growth factors regulated by this factor pathway).41,51 How each follicle recognizes and responds to hypoxia and the extent to which, if any, this has an impact on the viability of the oocyte remains to be investigated. However, many of the same factors are involved in the postovulatory development of a wellvascularized corpus luteum. In this respect, follicle competence is a periodic process that begins during folliculogenesis and continues after ovulation.56,59–61 At present, quantification of the levels of angiogenic factors such as VEGF or leptin on a follicle-specific basis to assess oocyte and embryo competence, is is not advisable. Van Blerkom et al35 reported that follicles with different perifollicular blood flow rates and intrafollicular dissolved oxygen contents tend to be associated with certain ranges of VEGF concentrations, but the interfollicular levels in each group (low-to-moderate-to-high flow) were not considered sufficiently significant or sensitive to use as an independent indicator of oocyte competence. A similar conclusion has been reported for normoresponders46 and poor responders.50 However, when taken together, these findings begin to offer a biological context for follicle-specific differences that may be related to fertility in general, and the maternal-age related reduction in fertility, in particular. They also provide a molecular basis for investigating the mechanisms by which the potential influences of certain intrafollicular signaling pathways may

HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 313


have adverse or beneficial effects on oocyte competence.

FOLLICULAR BIOCHEMISTRY – MINING FOLLICULAR FLUID FOR COMPETENCE RELATED DETERMINANTS

The expression levels of certain regulatory genes appears to be follicle specific,41 and this emphasizes a central issue in human follicular biology: the response of each follicle to stimulation can be unique, and is often unpredictable in terms of whether or how it may affect the developmental competence of its oocyte. However, the identification of specific follicular components that can be correlated with competence and combined with findings from noninvasive measures, such as Doppler ultrasound, would go a long way towards providing objective and standard parameters for oocyte/embryo selection in IVF. Follicular fluid is a complex substance, and investigation of the different growth factors, cytokines and other bioactive and regulatory molecules that may influence oocyte developmental competence adds its own layer of complexity to the clinical IVF laboratory: each assay must be validated, standardized, and quality controlled. If certain molecules are found to be competence related, the results need to be available in a timely manner for oocyte selection, i.e. before the consequences of the natural decay in competence defeat the purpose of establishing a biochemical basis for selection (e.g. spontaneous degradation of normal meiotic spindle organization and chromosomal segregation capacity).13 Thus, the choice of factors for analysis, either singly or in combination, must meet both stringent laboratory criteria and, more importantly, have a demonstrable association with outcome. It should also be noted that regulatory factors detected in follicular fluid may largely represent paracrine and autocrine processes that reflect the heterogeneous activities of the somatic cell component of the follicle; influences on oocyte competence, if any, may be indirect. Since the time that clinical IVF became an established treatment for infertility, a sizable literature

has developed on follicular biochemistry and its relationship to outcome. However, as is often the case, contradictory and equivocal results populate this landscape, and upon closer examination, many promising leads for oocyte and embryo selection are not confirmed, or their suggested sensitivity is not validated (for review see reference 57). However, there is growing evidence that quantification of certain bioactive molecules may provide a positive bias for selection with sensitivity levels that warrant incorporation into the clinical IVF laboratory. CORTISOL : CORTISONE RATIO

The ratio of cortisol to cortisone stands out as one of the more consistent and reproducible markers of oocyte and embryo competence, the expression of which is upregulated by LH and hCG,73 which upregulate expression of the enzyme 11␤-hydroxysteroid dehydrogenase (11␤-HSD), which converts the active glucocorticoid cortisol into cortisone, an inactive steroid. Keay et al74 reported that significantly higher pregnancy rates occurred at higher intrafollicular cortisol : cortisone ratios, and suggested that the presence of the active form in follicular fluid may influence the final stages of preovulatory oocyte maturation. The increased levels of cortisol in follicles from which competent oocytes were retrieved may reflect an early progesterone effect that results from the initial stages of transformation of granulosa cells to luteal cells. Progesterone inhibits 11␤HSD activity, and therefore inhibits the conversion of cortisol to cortisone.75 In a study of 80 IVF cycle outcomes,57 a cortisol : cortisone ratio of ⬎11.2 was associated with pregnancy, while unsuccessful procedures occurred with ratios of ⬍11.2. In support of a developmentally significant relationship between cortisol/cortisone levels and competence, Michael57 observed that other follicular factors suggested to predict outcome actually mediate levels of 11␤-HSD expression and activity. On a follicle-specific basis, these can indirectly determine the ratio of the active to inactive forms of this steroid. Higher cortisone levels appear to be associated with preovulatory luteinization of the mural granulosa,



HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 314


which may influence the production of regulatory factors within the follicle that promote oocyte competence during the terminal stages of oogenesis. Michael57 offered the following explanation as to how these steroid hormones could exert an influence on the oocyte: at the end of the follicular phase, higher cortisol levels resulting from the inhibition of 11␤-HSD activity by progesterone and other follicle-specific agents, may facilitate the closure of communicating (gap) junctions between the oocyte and cumulus/corona transzonal processes (for review see reference 4). After closure, the oocyte is no longer under direct maternal regulation, and the process of preovulatory nuclear (i.e. meiotic) and cytoplasmic maturation required to produce a competent gamete is initiated. This could explain why competent oocytes may be more likely to originate from follicles in which progesterone-mediated luteinization begins prior to ovulation. Whether the cortisol : cortisone ratio influences perifollicular blood flow rates and levels of angiogenic promoters synthesized by follicular granulosa cells warrants investigation. However, the 11␤-HSD-cortisol : cortisone pathway appears to be a common denominator in numerous interactions between important regulatory factors (inhibitors and promoters) in follicular fluid the levels of which have been associated with competence.57 If the superior outcomes observed at the high ratios are confirmed in large-scale studies, and the mechanisms by which oocyte competence is influenced clearly identified, the inclusion of this assay in the clinical IVF laboratory will merit very serious consideration. However, as with any assay that requires follicular specificity, including those described below, modifications in the protocol of aspiration would have to be incorporated such that the integrity of each aspirate is maintained in order to detect unique chemical characteristics. This may prove to be difficult to apply on a routine basis when many follicles exist, and, because of high patient load, oocyte retrievals are done rapidly and in a sequential manner, based on proximity and without intervening washes of the aspiration equipment. Thus, to warrant use in the clinical IVF program, the burden of proof will require unambiguous evidence that is based on outcome. 

PROTEIN GROWTH FACTORS

ANTI-MULLERIAN HORMONE

Protein markers of competence would seem to be a ‘laboratory-friendly’ approach to oocyte selection if they can be detected and quantified by immunologically based ‘kits’ that are currently available. Despite the diversity of protein growth factors, cytokines, and hormones identified in follicular fluid, including those known to be products of granulosa cells, few of the suggested protein markers of competence (e.g. VEGF), whether assessed individually or collectively, have proved to be sufficiently sensitive and specific to permit definitive determinations of competence. However, one regulatory protein, Mullerian inhibiting hormone (homologous to its male counterpart, anti-Mullerian hormone, or AMH), a dimeric glycoprotein member of the transforming growth factor (TGF)-␤ superfamily, currently shows promise in this regard. This factor is expressed by granulosa cells when the primordial follicle enters into the growth phase, and is upregulated at the transcriptional and translational levels in large pre-antral and small antral follicles.76 Several studies indicate that serum AMH is an independent indicator of the ovarian reserve (ovarian aging) that seems to effectively predict a patient’s unique response to COHS in general, and in IVF cycles, in particular.77–79 Whether AMH is an indicator of oocyte competence is controversial, since in some animal systems its expression is markedly downregulated in FSHsensitized follicles that are in the growing antral and preovulatory stages.80 However, AMH is detected in the serum and follicular fluids of women during COHS and in preovulatory follicles for at least 34 hours after the administration of an ovulatory dose of hCG.80,81 Results from several recent clinical studies indicate that serum AMH levels measured before,82 or during follicular stimulation,83 or on the day of ovulation induction with hCG,83 can be highly predictive of the ovarian reserve and follicular response to stimulation, measured by the number of preovulatory follicles that develop, and normal stage-appropriate embryo development in vitro. AMH levels

HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 315


measured in follicular fluid at ovum retrieval82 also seem to predict normal embryo development in vitro.83 These results strongly suggest that determinations of AMH may be a significant predictor of oocyte competence. Indeed, Eldar-Geva et al84 concluded that of all the protein factors currently suggested to be markers of competence, AMH may be the only reliable predictor of outcome that has currently been identified in follicular fluid. The finding that AMH levels are follicle specific and correlate with outcome when combined with levels measured in serum during the follicular and early luteal phase82 is an important one, because it strongly suggests that measurements of AMH in fluid aspirates from stimulated follicles may be an indication of normal intrafollicular conditions that produce a competent oocyte. Confirmation of this very promising lead will require following embryo development in vitro and fate after transfer on a follicle-specific basis. Follicle specific differences in cortisol : cortisone ratios, levels of HIF1 expression, and perifollicular blood flow patterns have each been suggested to indicate the development of a healthy follicle with a competent oocyte; whether these are related to AMH concentrations warrants detailed investigation. A strong association between perifollicular velocity characteristics, AMH levels measured in the corresponding follicular fluid, and outcome after embryo transfer, could lead to a common assessment for most infertility patients undergoing COHS and IVF, namely, Doppler ultrasonography as the primary non-invasive means of predicting, whether competent oocytes exist prior to retrieval, and in which follicles they are most likely to reside.

early follicular phase, peak at the early to mid-follicular phase, and then progressively decline until just after the LH surge. A transient spike in the level of inhibin B is detected 2 days after the LH surge, but then declines for the remainder of the luteal phase.85 Inhibin B is part of the hypothalamic– pitutary–ovarian feedback mechanism that inhibits FSH secretion as follicles develop during the follicular phase. Quantitative assays initially measured both inhibin A and B in the follicular fluid of spontaneously cycling women,86 and high levels were correlated with the fertilizing capacity of the corresponding oocytes. Later assays distinguished A from B, and serum inhibin B levels in cycles of COHS were reported to positively correlate with the number of follicles and oocytes obtained for IVF, as well as with the risk of developing ovarian hyperstimulation syndrome.87 Follicular fluid inhibin B levels were also correlated with oocyte quality and normal embryo development on days 2 and 3 of culture.87,88 Inhibin B is produced by the cumulus granulosa and its occurrence in serum at high levels on the day of follicular aspiration appears to positively correlate with outcome in cycles of COHS and IVF.89 Inhibin B assays tend to be difficult to perform on a routine basis and as a result, it remains to be determined whether the results are more predictive with respect to outcome than assays of AMH.90

OTHER CONSIDERATIONS OF FOLLICULAR BIOLOGY RELATED TO COMPETENCE EMPTY FOLLICLE SYNDROME

INHIBINS

Inhibins, specifically inhibin B, are another class of protein factors the follicle-specific levels of which have been related to competence. These heterodimeric glycoproteins in the TGF-␤ superfamily consist of a common ␣-subunit and one of two possible ␤-subunits (A and B) linked by disulfide bonds. Inhibins are secreted by granulosa cells, but inhibin B is expressed only in developing preantral and antral follicles. Inhibin B levels increase quickly during the

The current emphasis on the identification of bioactive molecules in follicular fluid is related to the need for unambiguous criteria for gamete selection in IVF.91 However, IVF cycles occur in which no oocytes are recovered. This phenomenon, termed ‘empty follicle syndrome’, is applied when no oocytes are found despite repeated, vigorous, and meticulous aspiration of follicles. Generally, the rate of follicular growth during the follicular phase is normal, and serum estradiol levels are stage-appropriate 

HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 316


with respect to the number of fully grown antral follicles.92 Failure to retrieve oocytes has often been attributed to patient error in the timing of ovulation induction by hCG administration. In these instances, the self-administration of hCG may be mistimed and so late that the subsequent oocyte retrieval was too premature with respect to the preovulatory development of the follicle. Failures in these retrievals may be due to oocytes that remain firmly adhered to the follicular wall during aspiration. In addition to timing errors, failure to administer hCG, or to inject it correctly (and entirely), as well as potential differences in hCG bioactivity between lots, have been suggested as iatrogenic rather than biological etiologies of this phenomenon. In these cycles, especially when the follicular aspirates are devoid of the usual masses of cumulus granulosa cells, it is prudent to assay serum ␤hCG level. In practice, when serum levels are ⬍10 mIU/ml, the procedure is suspended, a second hCG injection is given, and a second aspiration attempted some 36 hours later. The low circulating levels of hCG are indicative of an iatrogenic origin of the ‘empty’ follicles and a timely follow-up ovulation induction can produce retrievable oocytes that are viable, as evidence by successful outcomes.93 However, many cases do not have an iatrogenic origin, and in these instances, follicles that have adequate exposure to hCG do not appear to contain retrievable oocytes. Analysis of steroid levels in follicular fluid aspirates from ‘empty’ follicles shows high estradiol : progesterone ratios and high androstenedione levels, which indicates that they are not simply cysts, or prematurely luteinized, or atretic follicles.94 One biological explanation for their occurrence suggests an intrinsic defect in folliculogenesis that leads to early oocyte atresia.95 This notion is supported by the finding that empty follicles recur in subsequent IVF cycles for some affected women.96 However, case reports of successful outcomes with a second round of ovulation induction when circulating hCG levels are well above the 10 mIU/ml threshold noted above have been reported. In these instances, empty follicles may not reflect oocyte degeneration but rather an inability of the oocyte–cumulus complex to respond to hCG, or possibly, the need for a longer period to separate the complex from the 

follicular wall.97,98 Although a relatively rare phenomenon, additional research is needed to determine the exact cause(s) of this syndrome and whether factors that indicate a normal follicle such as perifollicular blood flow values, can predict the occurrence of ‘empty’ follicles.

THE INFLUENCE OF MATERNAL AGE AND LIFESTYLE FACTORS ON FOLLICULAR BIOLOGY AND OOCYTE COMPETENCE

Competence assessments at the follicular level may be particularly relevant in the evaluation of women of advanced reproductive (i.e. biological) age and those whose life-style factors may negatively contribute to fertility. Maternal age is the single most important biological determinant of a woman’s fertility potential, and has an enormous impact on the quantity of follicles and the quality of the oocytes she ovulates towards the end of her reproductive life. The total number of oocytes peaks at approximately 5 million around month 5 of fetal age and about 2 million remain at birth.99 This number is further reduced during a woman’s reproductive life; perhaps a thousand or fewer primordial follicles remain when menstrual cycles cease.100 The physical ‘expulsion’ of oocytes from the ovarian cortex and programmed cell death by apoptosis accounts for oocyte loss during the fetal and neonatal stages (for review see reference 4), while apoptosis and other mechanisms of cell death (e.g. atresia) are involved during reproductive life. In a woman’s normal reproductive lifespan, approximately 400–500 primordial follicles will progress to ovulation. Although there are usually sufficient primordial follicles remaining to allow menstrual cycles to continue until the sixth decade of life, fertility rates begin to fall much earlier. As shown in Figure 24.2, a gradual decline is first observed in the late 20s102 but the rate accelerates considerably around age 37.101,103,104 This progressive and natural reduction in ovarian reserve is one factor contributing to the age-related decline in fertility, but the quality of the corresponding oocytes also declines. Between the time of their formation in the fetus and the few hours

HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 317


30

Antral follicle count

15 10

12% 4.8%

5

2

1 25

30

35

40

45

Age (years) Figure 24.2 The relation between antral follicle counts and age shows a steeper drop after age 37 years. From Faddy et al101 with permission.

prior to ovulation, the cell cycle has been arrested at prophase of the first meiotic division. The resumption of arrested meiosis is initiated by the spontaneous LH surge (or hCG administration after COHS) and as a result, over 4 decades can elapse between the beginning and end of the first meiotic division in women. After the LH-induced completion of meiosis I, meiosis II progresses to and arrests at metaphase II. The second meiotic division is resumed after sperm penetration and completed with the extrusion of the second polar body. Chromatin condensation into bivalent MI and monovalent MII chromosomes permits cytogenetic analysis of ploidy at each metaphase during the preovulatory maturation of the oocyte. The availability of unfertilized oocytes from IVF procedures has provided an abundance of gametes for cytogenetic analysis, and the results consistently

demonstrate that the incidence of numerical (hypohaploidy, hyperhaploidy), structural (premature chromatid separation), and organizational abnormalities (scattering or malalignment on the spindle) increases with age, as do malformations in meiotic spindle organization.2,3,53 How these defects arise and their specific association with age remains controversial. In 1968, Henderson and Edwards105 proposed what has become known as the ‘first in, first out’ theory, whereby chromosomal defects common in oocytes ovulated late in life occur in those gametes which formed late during fetal oogenesis. According to this notion, the longer an oocyte resides in the ovary, the greater is the probability that it will have been exposed to developmentally toxic or adverse influences, such as oxidative free radicals or lytic enzymes released from proximal atretic oocytes.106,107 Others suggest that the prolonged interval of meiotic arrest compromises the capacity of the oocyte to resume meiosis108 or to form normally functioning metaphase spindles that can capture and retain all chromosomes, or once formed properly segregate chromosomes in a numerically equivalent manner between the oocyte and the polar bodies.13,53 The normal cycle checkpoints that prevent cell division when chromosomal capture by kinetochore microtubules is incomplete or chromosomal alignment on the metaphase spindle is aberrant do not occur in meiotic oocytes, and those which are operational may decline in efficiency with advanced maternal age.13 The occurrence of aneuploidies associated with spindle and segregation disorders (premature centromeric separation, anaphase lag, non-disjunction) not only lowers fertility potential and increases miscarriage rates, but also is clearly associated with a higher incidence of chromosomally abnormal offspring. Patterns of follicular growth change as women age; during the follicular phase, early growth is initially more rapid but then slows to produce terminally developed follicles the average diameter of which at ovulation is smaller than is typical for their younger counterparts. Older women also tend to have frequent cycles in which more than the usual single large (dominant) follicle develops. Both phenomena 

HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 318


are probably the consequence of unusually high FSH levels that produce a pattern of follicular stimulation not unlike the situation observed in cycles of COHS.109 While the capacity of follicles to make estradiol and inhibin A does not appear to be age related, against a background of equivalent FSH levels, ‘younger’ follicles tend to produce more inhibin B.110 Follicles from older women are less likely to transform into functional corpora lutea; despite ultrasonic and hormonal evidence of follicular development and ovulation, these follicles often fail to synthesize progesterone after ovulation. Because estradiol production is sufficient to prevent menses for as many as 12 days following ovulation, these aluteal cycles are clinically unrecognized.109 Rates of ovarian stromal blood flow are reduced in women over 40.111 Gaulden52 showed that older women experience inadequate expansion of the perifollicular vasculature during the follicular phase of the natural menstrual cycle, and proposed that perifollicular under-vascularization could be associated with intrafollicular under-oxygenation. According to her novel hypothesis on the relationship between intrafollicular hypoxia and trisomy (e.g. Down syndrome), increased frequencies of chromosomal aneuploidy may result from a modest acidification of the ooplasm such that even a slight reduction in intracellular pH could adversely influence the microtubular organization of the spindle and kinetic capacity of spindle microtubules to segregate chromosomes normally. Preliminary Doppler ultrasound findings indicate that in natural menstrual cycles, women ⬎40 years of age rarely produce follicles the perifollicular blood flow characteristics of which are consistent with higher levels of intrafollicular oxygenation and competent, euploid oocytes (Van Blerkom, unpublished). This possible mechanism of aneuploidy generation requires confirmation, but if validated, may provide one of the most important reasons to include Doppler ultrasonography in fertility treatments that require COHS in general, and in women of advanced reproductive age, in particular. Tobacco smoking is one of the most common and significant ‘lifestyle’ factors that can reduce a woman’s fertility potential. Polycyclic aromatic hydrocarbons, heavy metals such as cadmium, and nicotine 

metabolites are detected in follicular fluid and are currently thought to interfere with normal biosynthetic and regulatory activities during the follicular and preovulatory phases that may also involve the oocyte.112–115 Smoking causes an increase in intrafollicular levels of reactive oxygen species the pleiotropic effects of which on the oocyte and granulosa cell complement can include DNA fragmentation, chemical modification of proteins and certain nucleotides, and lipid peroxidation.116 Biochemical alterations of this type adversely influence cell function, and in severe cases, cause irreversible cell damage or death. Follicular aspirates obtained from smokers during ovum retrieval for IVF show higher levels of oxidation and lower levels of activity for normal follicular antioxidative mechanisms.117 Smokers also produce a high number of diploid oocytes, which may explain the unusually high frequency of triploidy reported for these women, despite a reduced fertilization capacity of their oocytes.118 Significantly, fewer follicles exist in the ovaries of current or previous smokers,119 and women who smoke go through an earlier menopause.120 Although other agents such as caffeine and alcohol have been shown to decrease fertility,121 it remains unclear whether the follicle or oocyte, or both, are directly affected. Weight is another lifestyle factor that can influence fertility, but the ovary is not directly involved; women who are significantly underweight or overweight are more likely to be anovulatory because of hormonal changes associated with altered hypothalamic and pituitary function. However, for ovulatory women undergoing COHS and IVF, weight is negatively associated with outcome. After ovulation induction, intrafollicular levels of hCG have been reported to be inversely related to body mass index (BMI). High BMI was not only associated with low intrafollicular hCG levels, but also with poor fertilizability, embryo quality, and a significant reduction in the likelihood of a positive outcome.122 For most women, however, smoking currently remains the single most important and ubiquitous lifestyle factor that negatively influences fertility. This factor becomes especially acute when it is considered that over the reproductive life of a woman, only 400–500 follicles will progress to ovulation,123

HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 319


and with age, the ability of follicles to produce a normal oocyte undergoes a natural and progressive decline. Studies of perifollicular blood flow that may correlate with levels of glucocorticoid and growth factor expression during the follicular phase (e.g. VEGF, leptin, AMH, inhibin B) may provide a molecular basis for reduced fertility in these women.

DO REPEATED CYCLES OF COHS NEGATIVELY INFLUENCE FERTILITY POTENTIAL?

It is common for infertile women to undergo multiple cycles of ovarian stimulation prior to initiating IVF, and to then have multiple IVF procedures performed before a viable pregnancy occurs or treatment is abandoned. Animal studies suggest that repeated ovarian stimulation can have adverse effects on follicular development and oocyte competence, especially with respect to the formation of a normal meiotic spindle with appropriately aligned chromosomes (for review see reference 124). However, at present there is no compelling evidence to suggest that a similar decline in oocyte competence accompanies multiple cycles of COHS and ovulation induction in the human. While women who undergo repeated cycles of stimulation are more likely to require progressively higher dosages of gonadotropins, when compared with levels used in the first cycle, the number of follicles produced and peak estradiol levels detected are similar.125,126 Repeated IVF cycles involve multiple follicular punctures and aspirations, and therefore can theoretically damage the ovary. There is no microscopic evidence to date to support potential adverse downstream consequences for human oocyte competence associated with multiple COHS/IVF procedures, although this issue does require further investigation.

SUMMARY

While the complex cellular and molecular biology of the ovarian follicle has long been recognized as a factor in the establishment of normal developmental

competence for the oocyte, understanding its relationship to the success of assisted reproduction in the human has increased in importance where restrictions on the number of embryos that can be replaced, or the stage of preimplantation embryogenesis permissible for transfer, are now legally mandated. These limits make the identification of folliclespecific biochemical properties and physiological characteristics that may be associated with oocyte competence necessary in clinical practice, as well as prerequisites for the aforementioned ‘holy grail’ of clinical IVF: accurate gamete selection. This is all the more important where restrictions extend to the number of oocytes that can be inseminated for IVF. Ultimately, the most practical means of follicular evaluation for oocyte selection are those that are non-invasive and can be readily introduced in assisted reproduction programs. The types of biochemical analyses and Doppler ultrasound characterizations described here meet these criteria, and remain promising leads in this regard. However, much remains to be understood about intrafollicular conditions that may influence the human oocyte, including the origin of follicle-specific differences and whether their developmental effects on the oocyte are direct or indirect. While the tone of this review is sanguine, the well-used caveat ‘more research is required’ applies because unambiguous associations with outcome will continually be required to validate this approach in the assessment of oocyte competence. Ideally, as investigations progress and clinical correlations with outcome are obtained, a more complete understanding of the molecular mechanisms that operate within the follicle, and how they may influence oocyte competence, will emerge. REFERENCES 1. Edwards R. Causes of early embryonic loss in human pregnancy. Hum Reprod 1986; 1: 185–98. 2. Van Blerkom J. Developmental failure in human reproduction associated with chromosomal abnormalities and cytoplasmic pathologies in meiotically mature oocyes. In. Van Blerkom J, ed. The Biological Basis of Early Human Reproductive Failure. Oxford: Oxford University Press, 1994: 283–326. 3. Plachot M. Genetic analysis of the oocyte – a review. Placenta 2003; 24 (Suppl 2): S66–72.



HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 320


4. Makabe S, Van Blerkom J. An Atlas of Human Female Reproductive Function: Ovarian Development to Early Embryogenesis after in vitro Fertilization. New York: Taylor and Francis, 2006. 5. Motta PM, Nottola SA, Familiari G et al. Morphodynamics of the follicular-luteal complex during early ovarian development and reproductive life. Int Rev Cytol 2003; 223: 177–288. 6. Sandalinas M, Sadowy S, Alikani M et al. Developmental ability. of chromosomally abnormal human embryos to develop to the blastocyst stage. Hum Reprod 2001; 16: 1954–8. 7. Verlinsky Y, Kuliev A. An Atlas of Preimplantation Genetic Diagnosis. New York; Parthenon Publishing, 2000. 8. Van Blerkom J, Henry G. Oocyte dysmorphism and aneuploidy in meiotically mature human oocytes after controlled ovarian stimulation. Hum Reprod 1992; 7: 379–90. 9. Meriano J, Alexis J, Visram-Zaver S et al. Tracking of oocyte dysmorphisms for ICSI patients may prove relevant to the outcome in subsequent patient cycles. Hum Reprod 2001; 16: 2118–23. 10. Mikkelsen A, Lindenberg S. Morphology of in-vitro matured oocytes: impact on fertility potential and embryo quality. Hum Reprod 2001; 16: 1714–18. 11. Otsuki J, Okada K, Morimoto Y et al. The relationship between pregnancy outcome and smooth endoplasmic reticulum clusters in MII human oocytes. Hum Reprod 2004; 19: 1591–7. 12. Ciotti P, Notarangelo A, Morselli-Labate V et al. First polar body morphology before ICSI is not related to embryo quality or pregnancy rate. Hum Reprod 2004; 19: 2334–9. 13. Eichenlaub-Ritter U, Sun F. Maternal age and oocyte competence. In Van Blerkom J, Gregory L, eds. Essential IVF: Basic Research and Clinical Applications. Boston: Kluwer Academic Publishers, 2004: 201–30. 14. Akande V, Flemming C, Hunt L, Keay S, Jenkins J. Biological versus chronological ageing of oocytes, distinguishable by raised FSH levels in relation to the success of IVF treatment. Hum Reprod 2002; 17: 2002–8. 15. Simonetti S, Veeck L, Jones H, Jr. Correlation of follicular fluid volume with oocyte morphology from follicles stimulated by human menopausal gonadotropin. Fertil Steril 1985; 44: 177–80. 16. Wittmaack F, Kreger D, Blasco L et al. Effect of follicular size on oocyte retrieval, fertilization, cleavage, embryo quality in in vitro fertilization cycles: a 6-year data collection. Fertil Steril 1994; 62: 1205–10. 17. Dubey A, Wang H, Duffy P, Penzias A. The correlation between follicular measurements, oocyte morphology and fertilization rates in an in vitro fertilization program. Fertil Steril 1995; 64: 787–90. 18. Arnot A, Vanderckhove P, DeBono M, Rutherford A. Follicular volume and number during in vitro fertilization: association with oocyte developmental capacity and pregnancy rate. Hum Reprod 1995; 10: 256–61. 19. Miller K, Goldberg J, Falcone T. Follicle size and implantation of embryos from in vitro fertilization. Obstet Gynecol 1996; 88: 583–6. 20. Ectors F, Vanderzwalmen P, Van Hoeck J et al. Relationship of human follicular diameter with oocyte fertilization and development after in-vitro fertilization or intracytoplasmic sperm injection. Hum Reprod 1997; 12: 2002–5. 21. Bergh C, Broden H, Lundin K, Hamberger L. Comparison of fertilization, cleavage and pregnancy rates of oocytes from large and small follicles. Hum Reprod 1998; 13: 1912–5. 22. Salha O, Nugent D, Dada T et al. The relationship between follicular fluid aspirate volume and oocyte maturity in in-vitro fertilization cycles. Hum Reprod 1998; 13: 1901–6. 23. Triwitayakorn A, Suwajanakorn S, Pruksananonda K et al. Correlation between human follicular diameter and oocyte outcome in an ICSI program. J Assist Reprod Genet 2003; 20: 143–7.



24. Gerris J. Reducing the number of embryos to transfer after IVF/ICSI In Van Blerkom J, Gregory L, eds. Essential IVF: Basic Research and Clinical Applications. Boston: Kluwer Academic Publishers, 2004: 505–75. 25. Benagiano G, Gianaroli L. The new Italian IVF legislation. Reprod Biomed Online 2004: 9: 117–25. 26. Papanikolaou E, Camus M, Kolibiankakis E et al. In vitro fertilization with single blastocyst-stage versus single cleavage-stage embryos. N Engl J Med 2006; 354: 1139–46. 27. Borini A. Predictive factors for embryo implantation potential. Reprod Biomed Online 2005; 10: 653–68. 28. Manna C, Nardo N. Italian law on assisted conception: clinical and research implications. Reprod Biomed Online 2005;11:532–4. 29. Strohmer H, Herczeg C, Plockinger B et al. Prognostic appraisal of success and failure of in vitro fertilization program by transvaginal Doppler ultrasound at the time of ovulation induction. Ultrasound Obstet Gynecol 1991; 1: 272–4. 30. Steer C, Campbell T, Tan S et al. The use of color flow imaging after in vitro fertilization to identify optimum uterine conditions before embryo transfer. Fertil Steril 1992; 57: 372–6. 31. Oyesanya O, Parsons J, Collins W et al. Prediction of oocyte recovery rate by transvaginal ultrasonography and color Doppler imaging before human chorionic gonadotropin administration in in vitro fertilization cycles. Fertil Steril 1996; 65: 806–9. 32. Nargund G, Doyle P, Bourne T et al. Association between ultrasound indices of follicular blood flow, oocyte recovery, and preimplantation embryo quality. Hum Reprod 1996; 11: 109–12. 33. Nargund G, Doyle P, Bourne T et al. Ultrasound derived indices of follicular blood flow before HCG administration and the prediction of oocyte recovery and preimplantation embryo quality. Hum Reprod 1996; 11: 2515–7. 34. Chui D, Jurovic D, Bourne T et al. Follicular vascularity – the predictive value of transvaginal power Doppler ultrasonography in an in vitro fertilization programme. A preliminary study. Hum Reprod 1997; 12: 191–6. 35. Van Blerkom J, Antczak M, Schrader R. The developmental potential of the human oocyte is related to the dissolved oxygen content of follicular fluid: association with vascular endothelial growth factor levels and perifollicular blood flow characteristics. Hum Reprod 1997; 12: 1947–55. 36. Bahl P, Pugh N, Chui D et al. The use of transvaginal power Doppler ultrasonography to evaluate the relationship between perifollicular vascularity and outcome in in-vitro fertilization treatment cycles. Hum Reprod 1999; 14: 939–45. 37. Coulam C, Goodman C, Rinehart J. Color Doppler indices of follicular blood flow as predictors of pregnancy after in-vitro fertilization and embryo transfer. Hum Reprod 1999; 14: 1979–82. 38. Huey S, Abuhamad A, Barrosa G et al. Perifollicular blood flow Doppler but not follicular pO2, pCo2, or pH, predict oocyte developmental competence in in vitro fertilization. Fertil Steril 1999; 72: 707–12. 39. Borini A, Maccolini A, Tallarini A et al. Perifollicular vascularity and its relationship with oocyte maturity and IVF outcome. Ann NY Acad Sci 2001; 943: 64–7. 40. Gregory L. Perifollicular vascularity: a marker of follicular heterogeneity and oocyte competence and a predictor of implantation in assisted conception cycles. In Van Blerkom J, Gregory L, eds. Essential IVF: Basic Research and Clinical Applications. Boston: Kluwer Academic Publishers, 2004: 59–80. 41. Van Blerkom J. Follicular influences on oocyte and embryo competence In DeJonge C, Barratt C, Eds. Assisted Reproductive Technology, Cambridge: Cambridge University Press, 2002: 81–105.

HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 321


42. Bahl P, Pugh N, Gregory L et al. Perifollicular vascularity as a potential variable affecting outcome in stimulated intrauterine insemination treatment cycles: a study using transvaginal power Doppler. Hum Reprod 2001: 16; 1682–9. 43. Shrestha S, Costello M, Sjoblom P et al. Power Doppler ultrasound assessment of follicular vascularity in the early follicular phase and its relationship with outcome in in vitro fertilization. J Assist Reprod Genet 2006; 23: 161–9. 44. Palomba S, Russo T, Falbo A et al. Clinical use of the perifollicular vascularity assessment in IVF cycles: a pilot study. Hum Reprod 2006; 21: 1055–61. 45. Merce L, Bau S, Barco M et al. Assessment of the ovarian volume, number and volume of follicles and ovarian vascularity by threedimensional ultrasonography and power Doppler angiography on the HCG day to predict the outcome in IVF/ICSI cycles. Hum Reprod 2006; 21: 1218–26. 46. Kim K, Oh D, Jeong J et al. Follicular blood flow is a better predictor of the outcome of in vitro fertilization-embryo transfer than follicular fluid vascular endothelial growth factor and nitric oxide concentrations. Fertil Steril 2004; 82: 586–92. 47. Pan H-A, Wu H-H, Cheng Y-C et al. Quantification of ovarian stromal Doppler signals in poor responders undergoing in vitro fertilization with three-dimensional power Doppler ultrasonography. Am J Obstet Gynecol 2004; 190: 338–44. 48. Bergqvist A, Bergqvist D, Ferno M. Estrogen and progesterone receptors in vessel walls: biochemical and immunochemical assays. Acta Obstet Gynecol Scand 1993; 72: 10–6. 49. Gislard V, Miller V, Van Houte P. Effects of 17␤-oestradiol on endothelium-dependent responses in the rabbit. J Pharmacol Exp Ther 1998; 244: 19–22. 50. Kan A, Ng EH, Yeung WS, Ho PC. Perifollicular vascularity in poor ovarian responders during IVF. Hum Reprod 2006; 21: 1539–44. 51. Van Blerkom J. Epigenetic influences on oocyte developmental competence: Follicular oxygenation and perifollicular vascularity. J Assist Reprod Genet 1998; 15: 226–34. 52. Gaulden M. The enigma of Down syndrome and other trisomic conditions. Mutat Res 1992; 269: 68–88. 53. Battaglia D, Goodwin P, Klein N et al. Influence of maternal age on meiotic spindle assembly in oocytes from naturally cycling women. Mol Hum Reprod 1996; 11: 2217–22. 54. Clark A, Stokes Y, Lane M, Thompson J. Mathematical modelling of oxygen concentration in bovine and murine cumulus-oocyte complexes. Reproduction 2006; 131: 999–1006. 55. Dickey R, Matulich E. Ultrasound imaging at the beginning of the second millennium. In DeJonge C, Barratt C, eds. Assisted Reproductive Technology, Cambridge: Cambridge University Press, 2002: 282–301. 56. Antczak M. The synthetic and secretory behaviors (nonsteroidal) of ovarian follicular granulosa cells: parallels to cells of the endothelial cell lineage. In Van Blerkom J, Gregory L, Eds. Essential IVF: Basic Research and Clinical Applications. Boston: Kluwer Academic Publishers, 2004: 1–41. 57. Michael A. Do biochemical predictors of outcome exist? In Van Blerkom J, Gregory L, Eds. Essential IVF: Basic Research and Clinical Applications. Boston: Kluwer Academic Publishers, 2004: 81–110. 58. Malamitsi-Puchner A. Novel follicular fluid factors influencing oocyte developmental potential in IVF: a review. Reprod Biomed Online 2006; 12: 500–6. 59. Otani N, Minami S, Yamoto M et al. The vascular endothelial growth factor/fms-like tyrosine kinase system in human ovary during the menstrual cycle and early pregnancy. J Clin Endocrinol Metabol 1999; 84: 3845–51.

60. Yan Z, Neulen J, Raczek S et al. Vascular endothelial growth factor (VEGF)/vascularity permeability factor (VPF) production by luteinized human granulosa cells in vitro; a paracrine signal in corpus luteum formation. Gynecol Endocrinol 1993; 12: 145–53. 61. Antczak M, Van Blerkom J, Clark A. A novel mechanism of vascular endothelial growth factor, leptin and transforming growth factor beta2 sequestration in a subpopulation of human ovarian follicle cells. Hum Reprod 1997; 12: 2226–34. 62. Cioffi J, Van Blerkom J, Antczak M et al. The expression of leptin and its receptors in pre-ovulatory human follicles. Mol Hum Reprod 1997: 3: 467–72. 63. Bouloumie A, Drexler H, Lafontan M, Busse R. Leptin, the product of the ob gene, promotes angiogenesis. Circ Res 1998; 83: 1059–66. 64. Kamut B, Brown L, Mannseau E et al. Expression of vascular endothelial growth factor by human granulosa and theca lutein cells. Role in corpus luteum development. Am J Pathol 1995; 146: 157–65. 65. Shweiki D, Itin A, Soffer D, Keshet E. Vascular endothelial growth factor induced by hypoxia may mediate hypoxia-initiated angiogenesis. Nature 1992; 359: 843–5. 66. Smeneza G. Regulation of mammalian O2 homeostasis by hypoxiainducible factor 1. Ann Rev Cell Dev Biol 1999; 15: 551–78. 67. Bruick RK. Oxygen sensing in the hypoxic response pathway: regulation of the hypoxia-inducible transcription factor. Genes Dev 2003; 17: 2614–23. 68. Bell E, Brooke M, Emerling B, Navdeep S. Mitochondrial regulation of oxygen sensing. Mitochondrion 2005; 5: 322–32. 69. Bunn H, Poynton R. Oxygen sensing and molecular adaptations to hypoxia. Physiol Rev 1996; 76: 839–85. 70. Antczak M. Possible ramifications of the identification of ovarian follicular granulosa cells as specialized endothelial-like cells: a speculative treatise. Reprod Biomed Online 2001; 2: 188–97. 71. Artini P, Monti M, Fasciani A et al. Vascular endothelial cell growth factor, interleukin-6 and interleukin-2 in serum and follicular fluid of patients with ovarian hyperstimulation syndrome. Eur J Obset Gynecol Reprod Biol. 2002; 101: 169–74. 72. Friedman C, Seifer D, Kennard E et al. Elevated levels of follicular fluid vascular endothelial growth factor is a marker of diminished pregnancy 73. Tetsuka M, Haines L, Milne M et al. Regulation of 11␤-hydroxysteroid dehydrogenase type 1 gene expression by LH and interleukin-1␤ in cultured rat granulosa cells. J Endocrinol 1999; 163: 417–23. 74. Keay S, Harlow C, Wood P et al. Higher cortisol : cortisone ratios in the preovulatory follicle of unstimulated IVF cycles indicate oocytes with increased pregnancy potential. Hum Reprod 2002; 17: 2003–8. 75. Michael A, Gregory L, Thaventhiran L et al. Follicular variation in ovarian 11␤-hydroxysteroid dehydrogenase (11␤HSD) activities: evidence for the paracrine inhibition of 11␤HSD in human granulosa-lutein cells. J Endocrinol 1996;148: 419–25. 76. Durlinger A, Gruijeters M, Kramer P et al. Anti-Mullerian hormone inhibits initiation of primordial follicle growth in the mouse ovary. Endocrinology 2002; 143: 1976–84. 77. van Rooij I, Broekmans F, te Velde E et al. Serum anti-Mullerian hormone levels: a novel measure of ovarian reserve. Hum Reprod 2002; 17: 3065–71. 78. Hazout A, Bouchard P, Seifer D et al. Serum antimullerian hormone/mullerian-inhibiting substance appears to be a more discriminatory marker of assisted reproductive technology outcome than follicle-stimulating hormone, inhibin B, or estradiol. Fertil Steril 2004; 82: 1323–9. 79. Seifer D, MacLaughlin D, Christian B et al. Early follicular serum mullerian-inhibiting substance levels are associated with ovarian



HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 322


80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

90.

91. 92. 93. 94. 95.

96. 97.

98.



response during assisted reproductive technology cycles. Fertil Steril 2002; 77: 468–71. Hirobe S, He W, Gustafson M, MacLaughlin D, Donahoe P. Mullerian inhibiting substance gene expression in the cycling rat ovary correlates with recruited or graafian follicle selection. Biol Reprod 1994; 50: 1238–43. Fallat M, Siow Y, Marra M et al. Mullerian-inhibiting substance in follicular fluid and serum: a comparison of patients with tubal factor infertility, polycystic ovary syndrome, and endometriosis. Fertil Steril 1997; 67: 962–5. Franchin R, Mendez Lozano D, Louafi N et al. Dynamics of serum anti-mullerian hormones levels during the luteal phase of controlled ovarian hyperstimulation. Hum Reprod 2005; 20: 747–51. Silberstein T, MacLaughlin D, Shai I et al, Mullerian inhibiting substance levels at the time of HCG administration in IVF cycles predicts both ovarian reserve and embryo morphology. Hum Reprod 2006: 21: 159–63. Eldar-Geva T, Ben-Chetrit A, Spitz I et al. Dynamic assays of inhibin B, anti-mullerian hormone and estradiol following FSH stimulation and ovarian ultrasonography as predictors of IVF outcome. Hum Reprod 2005; 20: 3178–83. Groome N, Illingworth P, O’Brien M et al. Measurement of dimeric inhibin B throughout the human menstrual cycle. J Clin Endocrinol Metab 1996; 814: 1401–5. Fowler P, Fahy U, Culler M et al. Gonadotrophin surge-attenuating factor bioactivity is present in follicular fluid from naturally cycling women. Hum Reprod 1995; 10: 68–74. Enskog A, Nilsoon L, Brannstrom M. Peripheral blood concentrations of inhibin B are elevated during gonadotrophin stimulation in patients who later develop ovarian OHSS and inhibin A concentrations are elevated after OHSS onset. Hum Reprod 2000; 15: 532–8. Lau C, Ledger W, Groome NP et al. Dimeric inhibins and activin A in human follicular fluid and oocyte-cumulus culture media. Hum Reprod 1999; 14: 2525–30. Chang C, Wang T, Horng S et al. The concentration of inhibin B in follicular fluid: relation to oocyte maturation and embryo development. Hum Reprod 2002; 17: 1724–8. Urbancsek J, Hauzman E, Klinga K et al. Use of serum inhibin B levels at the start of ovarian stimulation and at oocyte pickup in the prediction of assisted reproductive treatment outcome. Fertil Steril 2005; 83: 341–8. Mendoza C, Ruiz-Requena E, Ortega E et al. Follicular fluid markers of oocyte developmental potential. Hum Reprod 2002; 17: 1017–22. Coulam C, Bustillo M, Schulman J. Empty follicle syndrome. Fertil Steril 1986; 46: 1153–5. Ndukew G, Thornton S, Fishel S et al. Curing empty follicle syndrome. Hum Reprod 1997; 12: 21–3. Tsuiki A, Rose B, Hung T. Steroid profiles of follicular fluids from a patient with empty follicle syndrome. Fertil Steril 1988; 49: 104–7. Bustillo M. Unsuccessful oocyte retrieval: technical artifact or genuine ‘empty follicle syndrome’. Reprod Biomed Online 2004; 8: 59–67. Zreik TG, Garcia-Velasco J, Vergara T et al. Empty follicle syndrome: evidence for recurrence. Hum Reprod 2000; 15: 999–1002. Hassan H, Saleh H, Khalil O et al. Double oocyte aspiration may be a solution for empty follicle syndrome. Case reports. Fertil Steril 1998; 69: 138–9. Awoniyi A, Govindbhai J, Zierke S et al. Continuing the debate on empty follicle syndrome: can it be associated with normal bioavailability of ␤-human chorionic gonadotropin on the day of oocyte recovery? Hum Reprod 1998: 13: 1281–4.

99. Yen S, Jaffe R, Barbieri R. Reproductive Endocrinology, 4th edn. WB Saunders, 1999, Philadelphia. 100. Faddy M, Gosden R. A model conforming the decline in follicle numbers to the age of menopause in women. Hum Reprod 1996; 11: 1484–6. 101. Scheffer G, Broekmans F, Dorland M et al. Antral follicle counts by transvaginal ultrasonography are related to age in women with proven natural fertility. Fertil Steril 1999; 72: 845–51. 102. Dunson D, Colombo B, Baird D. Changes with age in the level and duration of fertility in the menstrual cycle. Hum Reprod 2002; 17: 1399–403. 103. Piette C, de Mouzon J, Bachelot A et al. In-vitro fertilization: influence of women’s age on pregnancy rates. Hum Reprod 1990; 5: 56–9. 104. Faddy M, Gosden R, Gougeon A et al. Accelerated disappearance of ovarian follicles in mid-life-implications for forecasting menopause. Hum Reprod 1992; 7: 1342–6. 105. Henderson D, Edwards R. Chiasma frequency and maternal age in mammals. Nature 1968; 218: 22–8. 106. Volarcik K, Sheean L, Goldfarb J et al. The meiotic competence of in vitro matured human oocytes is influenced by donor age: evidence that folliculogenesis is compromised in the reproductively aged ovary. Hum Reprod 1998; 13: 154–60. 107. Tarin J. Aetiology of age-associated aneuploidy: a mechanism based of the ‘free radical theory of aging’. Mol Hum Reprod 1995; 1: 1563–5. 108. Crowley P, Gulah D, Hayden T et al. A chiasma-hormonal hypothesis relating Down’s syndrome and maternal age. Nature 1979; 280: 417–8. 109. Santoro N, Isaac B, Neal-Perry G et al. Impaired folliculogenesis and ovulation in older reproductive age women. J Clin Endocrinol Metab 2003: 88: 5502–9. 110. Hansen K, Thyer A, Sluss P et al. Reproductive ageing and ovarian function: is the early follicular phase FSH rise necessary to maintain adequate secretory function in older ovulatory women? Hum Reprod 2005; 20: 89–95. 111. Ng E, Chan C, Yeung W et al. Effect of age on ovarian stromal flow measured by three-dimensional ultrasound with power Doppler in Chinese women with proven fertility. Hum Reprod 2004; 19: 2123–37. 112. Zenzes M, Wang P, Casper R. Cigarette smoking may affect meiotic maturation of human oocytes. Hum Reprod 1995; 10: 3213–7. 113. Zenzes M, Krishnan S, Krishnan B et al. Cadmium accumulation in follicular fluid of women in in vitro fertilization-embryo transfers is higher in smokers. Fertil Steril 1995; 64: 599–603. 114. Zenzes M, Reed T, Wang P et al. Cotinine, a major metabolite of nicotine, is detectable in follicular fluids of passive smokers in in vitro fertilization therapy. Fertil Steril 1996; 66: 614–19. 115. Zenzes M, Puy L, Bielecki R. Immunodetection of benzo[a]pyrene adducts in ovarian cells of women exposed to cigarette smoke. Mol Hum Reprod 1998; 4: 159–65. 116. Finkel T, Holbrook N. Oxidants, oxidative stress and the biology of ageing. Nature. 2000; 408: 239–47. 117. Paszkowski T, Clarke RN, Hornstein MD. Smoking induces oxidative stress inside the Graafian follicle. Hum Reprod 2002; 17: 921–5. 118. Rosevear S, Holt D, Lee T et al. Smoking and decreased fertilisation rates in vitro. Lancet 1992; 340: 1195–6. 119. Westhoff C, Murphy P, Heller D. Predictors of ovarian follicle number. Fertil Steril 2000; 74: 624–8. 120. Midgette A, Baron J. Cigarette smoking and the risk of natural menopause. Epidemiology 1990; 6: 474–80. 121. Hassan M, Killick S. Negative lifestyle is associated with a significant reduction in fecundity. Fertil Steril 2004; 81: 384–92.

HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 323


122. Carrell D. Body mass index is inversely related to intra-follicular HCG concentrations, embryo quality and IVF outcome. Reprod BioMed Online 2001; 3: 109–11. 123. Peters H. Intrauterine gonadal development. Fertil Steril 1976; 27: 493–500. 124. Van Blerkom J, Davis P. Differential effects of repeated ovarian stimulation on cytoplasmic and spindle organization in metaphase II

mouse oocytes matured in vivo and in vitro. Hum Reprod 2001; 16: 757–64. 125. Doldi N, Persico P, De Santis L et al. Consecutive cycles in in vitro fertilization–embryo transfer. Gynecol Endocrinol 2005; 20: 132–6. 126. Hoveyda F, Engmann L, Steele J et al. Ovarian response in three consecutive in vitro fertilization cycles. Fertil Steril 2002; 77: 706–10.



HPE_Chapter24.qxp

7/13/2007

4:54 PM

Page 324

HPE_Chapter25.qxp

7/14/2007

5:52 PM

Page 325

25. Sperm DNA and embryo development Denny Sakkas and Emre Seli

INTRODUCTION

The potential impact of an abnormal paternal genome on reproductive outcome is unquestionably less than that of its female counterpart. The importance of the egg is best illustrated by the plethora of studies relating female age to pregnancy outcome.1,2 Furthermore, the impact of egg quality on reproductive success has been powerfully established by the continued success of donor oocyte programs. In the USA, the Assisted Reproductive Technology Report for 2003 indicated that the national data from donor oocytes gave rise to a 50% live birth rate, with some clinics reporting live birth rates of over 70%.3 In contrast, the influence of the human sperm on reproductive outcome has not yet been well characterized. In animal studies, evidence of a paternal effect on reproductive outcome is clear. Many animal studies have centered around examining the effect of radiation or toxic chemicals on spermatozoa and their subsequent effects on reproductive outcome. An example of some of these intriguing studies includes data from Robaire’s group, who have shown clear evidence that cyclophosphamide treatment affects rodent sperm in such a way that it leads to dysregulation of gene expression in the preimplantation embryo, increased postimplantation losses, fetal malformations, and transmission of fetal abnormalities to the subsequent generation.4–6 One of the most worrying traits observed in these studies is the inheritance of a paternal component over generations. This observation has recently been dramatically highlighted in a study examining the transient exposure of a gestating female rat to endocrine disruptors during the period of gonadal sex determination. The endocrine disruptors induced

an adult male phenotype of decreased spermatogenic capacity (cell number and viability) and increased incidence of male infertility in the F1 generation. These effects were transferred through the male germ line to nearly all males of all subsequent generations examined (that is, F1 to F4).7 In the human, paternal effects on reproductive outcome are not as clear cut. One study by Parker et al8 suggested an association between the risk of stillbirth and the father’s exposure to external ionizing radiation before conception. However this epidemiological study has also been criticized.9 A number of studies have also linked advanced paternal age with the inheritance of adverse traits, including birth defects, mental retardation, Apert’s syndrome, autism, and schizophrenia.10–14 All of these studies further suggest that biological mechanisms exist in sperm, and their disturbance will disrupt the delicate equilibrium needed for successful fertilization and embryo growth, with various adverse reproductive outcomes as a result. The factors present in the paternal genome that may have an impact on poor reproductive outcome are still hypothetical. However, in the past decade the quality of the sperm nuclear DNA has been more rigorously examined. This has also coincided with an increased use of intracytoplasmic sperm injection (ICSI), for which the quality of spermatozoa is generally accepted to be poorer. An abundance of articles have been published indicating that DNA strand breaks are clearly detectable in ejaculated spermatozoa, and their presence is increased in the ejaculates of men with poor semen parameters.15–21 This chapter therefore examines increasing evidence that paternal factors exist which may impact on reproductive outcome, with particular emphasis on abnormal sperm nuclear DNA.



HPE_Chapter25.qxp

7/14/2007

5:52 PM

Page 326


SPERMATOZOAL COMPONENTS AND THEIR EFFECT ON REPRODUCTIVE OUTCOME

Various components of the fertilizing spermatozoon have been studied for their impact on the reproductive process. Anomalies of either the sperm cytoplasmic membrane or cytoskeleton are most likely to affect the postfertilization processes. Indeed, centriolar abnormalities have largely been shown to impede fertilization,22,23 although an effect on later stages cannot be excluded. The mitochondria are believed to be maternally inherited, although some argue that this inheritance could be skewed by micromanipulation techniques such as those utilized in assisted reproductive technologies.24,25 The nucleus therefore remains as the greatest contributor to the potential success of a reproductive outcome.

SPERM NUCLEAR DETERMINANTS IMPACTING ON REPRODUCTIVE OUTCOME

A multitude of acquired spermatozoal nuclear factors may affect reproductive outcome. These include numerical abnormalities in spermatozoal chromosome content, Y chromosome microdeletions, alterations in the epigenetic regulation of paternal genome, and, finally, non-specific DNA strand breaks (reviewed in reference 16). The impact of aneuploidy of paternal origin and Y-chromosome microdeletions are clear and have both been extensively investigated. Animal studies reveal that the male germ cell has a role in epigenetic inheritance,7 and suggest that epigenetic marks, especially DNA methylation, are unstable, and may be altered by culture conditions.26,27 Assisted reproductive technologies (ART) rely on manipulation and culture of gametes and embryos at times when epigenetic programs are being acquired and modified. An initial study of samples from 92 children conceived using ICSI found no evidence of altered methylation at 15q11-13, the locus linked to the pathogenesis of the imprinting disorders Angelman and Prader-Willi syndromes.28 

However, six studies have recently reported two imprinting disorders, Beckwith-Wiedemann syndrome29–32 and Angelman syndrome33,34 in association with the use of ART. The report by Halliday et al32 was the first case–control study to suggest that a child with Beckwith-Wiedemann syndrome had a probability of having been conceived using IVF that is 18 times greater than for a child without the syndrome. Although these data are worrying, they may also be exposing particular genetic and developmental traits in infertile couples that influence imprinting.35 A number of studies have examined imprinting in spermatozoa of men with abnormal semen parameters. An initial study by Manning et al36 used polymerase chain reaction (PCR)-based techniques to analyze DNA extracted from spermatozoa of men with normal semen analysis (n ⫽ 30) and from men with medium (n ⫽ 30, 5–20 million spermatozoa/ mL) and high grade semen pathology (n ⫽ 30, ⬍5 million spermatozoa/mL) undergoing ICSI. They failed to detect a difference in methylation status between the groups; however, this does not confirm that differences may not exist. More recently, Marques et al37 studied two oppositely imprinted genes in spermatozoal DNA from normozoospermic and oligozoospermic patients. In the mesodermal specific transcript gene (MEST), bisulfite genomic sequencing showed that maternal imprinting was correctly erased in all 123 patients. However, methylation of the H19 gene did not change in any of 27 normozoospermic individuals (0%) compared with methylation changes in eight moderate (17%, p ⫽ 0.026) and 15 severe (30%, p ⫽ 0.002) oligozoospermic patients. Their findings suggest an association between abnormal genomic imprinting and hypospermatogenesis, and that spermatozoa from oligozoospermic patients carry a raised risk of transmitting imprinting errors. Testicular sperm extraction and intracytoplasmic sperm injection (TESE-ICSI) is a frequently used therapeutic option in azoospermic males. Manning et al36 used hemi-nested methylationspecific PCR for the target region 15q11-13 to analyze imprinting at the single-cell level in cells from different stages of spermatogenesis. They analyzed

HPE_Chapter25.qxp

7/14/2007

5:52 PM

Page 327

SPERM DNA AND EMBRYO DEVELOPMENT

spermatozoa, elongated spermatids, and round spermatids, and found completed establishment of the correct paternal imprint in these three developmental stages. However, there was a high rate of amplification failure in round spermatids in this study, raising a factor of uncertainty. It should be emphasized that the mechanism that establishes paternal epigenetic marks which contribute to the future genome are still major issues that remain to be explored.38

now showing DNA damage in ejaculated sperm. Although some of these defects are clearly environmentally related, with smoking as a good (or bad) example,48,49 many are thought to be intrinsic to the patients’ clinical makeup; whether they are related to a failure in the patients’ DNA repair mechanisms is as yet unknown.

NUCLEAR DNA STRAND BREAKS IN SPERMATOZOA

Apoptosis represents an additional mechanism for removing cells with DNA lesions, and operates in relation to DNA repair. Spontaneous apoptosis does occur in all cell types of the testis during normal spermatogenesis in man,50 and testicular cells are very sensitive to apoptotic stimuli such as high-dose chemotherapy.51 It is likely that the total level of DNA lesions is of decisive importance for the fate of the spermatogenic cell. On the other hand, there is evidence that benzo(a)pyrene (B(a)P) adducts can indeed accumulate without eliciting apoptosis in spermatogenic cells, since ejaculated sperm from smokers contain more B(a)P adducts than from non-smokers.49 Furthermore, such adducts derived from the father do not prevent fertilization and are persistent even at the blastocyst level,47 also indicating that they are not always completely repaired in the fertilized oocyte. A recent publication52 showed that assisted fertilization was 40% less likely to produce a live offspring if the father was a smoker, compared with if he was a non-smoker. Four possible functional roles have been proposed for apoptosis during normal spermatogenesis.

The presence of nuclear DNA strand breaks in ejaculated spermatozoa as an aspect of sperm quality has been examined more closely in the past decade. Their presence was initially reported in the early 1990s,39–41 and their origin and subsequent impact is still not completely understood.21,42,43 The complex process of spermatogenesis involves an intricate series of mitoses and a meiotic division followed by marked changes in cell structure. In addition to proliferation and differentiation, the outcome of spermatogenesis is affected by the extent of germ cell death. During spermatogenesis, germ cell death occurs normally and continuously, and is estimated to result in the loss of up to 75% of the potential number of spermatozoa.44 It is believed that nuclear DNA strand breaks may arise due to the failure or inadequacy of the DNA policing and/or processing mechanisms that take place during spermatogenesis. The three main mechanisms responsible for the quality of DNA in the spermatozoa are DNA repair, apoptosis, and chromatin remodeling. All of these steps have been implicated in the generation of DNA strand breaks in the ejaculated spermatozoa. DNA REPAIR

A failure in effective DNA repair in human testicular cells is reflected in an accumulation of DNA damage adducts such as 8-oxoG and benzo(a)pyrene in human sperm15,45–47 and by the plethora of papers

APOPTOSIS

(1) Maintenance of an optimal germ cell/Sertoli cell ratio. It has been established that each Sertoli cell can support only a finite number of germ cells throughout their development into spermatozoa. Therefore, supraoptimal numbers of spermatogonia may undergo apoptosis to maintain an optimal ratio.53 (2) Elimination of abnormal germ cells. There may be a selective process in which abnormal germ 

HPE_Chapter25.qxp

7/14/2007

5:52 PM

Page 328


cells, especially chromosomally abnormal germ cells, are eliminated from the population by apoptosis.54 (3) Formation of the blood–testis barrier by tight junctions between Sertoli cells. Elimination of excessive germ cells is necessary for this process. Suppression of germ cell apoptosis by inactivation of the apoptosis-promoting factor bax prevents the formation of these tight junctions.55 (4) Creation of a prepubertal apoptotic wave. A massive wave of germ cell apoptosis normally takes place as mammalian species approach puberty. This wave serves as a regulator of the ratio between germinal cells in various stages and Sertoli cells, and facilitates the establishment of mature spermatogenesis. There is evidence that preventing this wave of apoptosis by expression of apoptosis-inhibitory proteins, such as Bcl-xL or Bcl-2, results in highly abnormal adult spermatogenesis accompanied by sterility.56 Whereas apoptosis appears to be a component of normal spermatogenesis, one puzzling occurrence in ejaculated human spermatozoa is the presence of apoptotic marker proteins. In the human, we originally found that men with abnormal sperm parameters display higher levels of the apoptotic protein Fas on their ejaculated spermatozoa.57 The presence of Fas on ejaculated spermatozoa correlated strongly with a decreased sperm concentration and sperm with abnormal morphology. More recently, we and others also found that other apoptotic markers such as Bcl-x, p53, caspase, and Annexin V are also present on ejaculated human spermatozoa and show distinct relationships with abnormal semen parameters.58–61 Double labeling experiments have shown that ejaculated human sperm expressing apoptotic marker proteins can also display chromatin damage and show signs of immaturity, such as cytoplasmic retention.62 Interestingly, exclusion of sperm displaying these apoptotic markers can also limit the number of sperm possessing DNA strand breaks, but this is not exclusive.60 Said et al63 have shown that annexin V-negative sperm displayed superior quality in terms of high motility, low caspase 3 acti

vation, MMP integrity, and small extent of DNA fragmentation. In the same study it was shown that these sperm also had a higher fertilization potential.

DNA REMODELING

DNA strand breaks detected in male germ cells may also be due to factors unrelated to apoptosis. Indeed, McPherson and Longo64,65 demonstrated the presence of endogenous DNA strand breaks in elongating rat spermatids, when chromatin structure and nucleoproteins are modified. They proposed that the presence of endogenous nicks in ejaculated spermatozoa might be indicative of incomplete maturation during spermiogenesis. They also postulated64,65 that chromatin packaging might involve endogenous nuclease activity in order to create and ligate nicks during the replacement of histones by protamines. The transient presence of DNA breaks has been reported in both mouse and human.66–68

THE IMPACT OF SPERM NUCLEAR DNA STRAND BREAKS ON REPRODUCTIVE OUTCOME

An increased presence of DNA strand breaks in ejaculated spermatozoa could affect fertilization, blastocyst development, and pregnancy rates. Investigation of the possible association between DNA strand breaks in spermatozoa and fertilization rates in patients undergoing ART have in general found no strong correlation between DNA integrity of ejaculated spermatozoa and IVF and ICSI fertilization rates.69–72 In contrast to these reports, a negative correlation between sperm DNA strand breaks and IVF and ICSI fertilization rates has been reported using the terminal deoxynucleotidyl transferase mediated dUTP nick end labeling (TUNEL) assay.45,73 A recent meta-analysis examining eight articles using the TUNEL and sperm chromatin structure assay (SCSA) found that sperm DNA damage did not affect either IVF or ICSI fertilization.74 Unfortunately, many of the human studies examine sperm populations and

HPE_Chapter25.qxp

7/14/2007

5:52 PM

Page 329


are unable to examine individual spermatozoa which actually fertilize the egg, as in animal studies. Activation of embryonic genome expression occurs at the 4–8-cell stage in human embryos,75 suggesting that the paternal genome may not have an influence on development until that stage. Therefore, an elevated incidence of DNA strand breaks in spermatozoa may not have an impact on early embryo development. Indeed, a number of studies have failed to find an association between sperm DNA damage and early embryo development.45,76 However, evaluating the outcomes of ART cycles that involved egg sharing, we did observe a paternal effect on embryo development prior to embryonic genome activation.77 During egg-sharing cycles, the eggs from one woman are shared between herself and a recipient, and spermatozoa from two different partners are used to fertilize the eggs. When comparing embryo development between egg-sharing pairs, we found that approximately 30% of patients showed a difference in mean embryo score of ⱖ5 in all embryo development, and 14% showed a difference in the quality of embryos available for transfer on day 2 and 3. While sperm DNA integrity was not evaluated in this study, its results indicate that a spermatozoal component may impede embryo development as early as the 4–8-cell stage in human. A number of early studies suggested a paternal effect on blastocyst development in relation to sperm quality78 and when comparing IVF and ICSI.79 An interesting observation in a number of these reports was that embryo quality was similar until day 2–3 and then the differences in blastocyst development were observed in the day 3 to day 5 stage.79,80 This again indicates that a paternal effect may manifest itself postembryonic genome activation. The possible consequences of fertilisation with sperm possessing abnormal DNA are shown in Figure 25.1. Whether the extent of nuclear DNA damage in ejaculated spermatozoa affects blastocyst development after IVF and ICSI has also been examined. A negative correlation using both the TUNEL assay to evaluate spermatozoa processed for IVF81 and SCSA to evaluate unprocessed spermatozoa82 has been observed. In our study we found that when the cut-

(a) No DNA Repair Fertilisation Failure Partial DNA Repair Fertilisation

DNA Repair Fertilisation

Normal offspring

?

Abnormal offspring

Development and offspring

(d)

(b)

(c)

Developmental failure post - embryonic genome activation

Figure 25.1 Consequences of fertilization with sperm containing damaged DNA (red). (A) The spermatozoon may fail to fertilize. (B) The spermatozoon containing damaged DNA may be repaired upon fertilization and a normal blastocyst develops with no paternal abnormalities (green). (C) The spermatozoon containing damaged DNA may be partially repaired upon fertilization and a normal blastocyst develops with paternal abnormalities (red). (D) The spermatozoon containing damaged DNA may fertilize, however, the embryo fragments postembryonic genome activation and fails to form a blastocyst.

off score for sperm DNA damage (as measured by TUNEL) was ⬎50%, four out of eight patients had no blastocysts developing in vitro. In contrast, only two out of the 41 remaining patients did develop blastocysts, all of whom had sperm DNA damage of ⬍50%. In the study by Virro et al,82 males with ⱖ30% DNA fragmentation index (DFI) had low blastocyst rates (⬍30%) and no ongoing pregnancies. These findings illustrate not only the importance of sperm DNA integrity and its association with the embryo’s ability to develop postembryonic genome 

HPE_Chapter25.qxp

7/14/2007

5:52 PM

Page 330


activation, but also indicate that it is not an all-ornone phenomenon, as sperm that can support embryo development are present in the population. Pregnancy rates after IVF are also reduced in couples who have higher percentages of spermatozoa with DNA strand breaks detected by in situ nick translation.70 Similarly, there is strong evidence for a relationship between sperm nuclear DNA integrity, as assessed with the SCSA, and fertility after normal intercourse83,84 or ART.85 Interestingly, Evenson et al83 found cases in which the classical criteria (concentration, motility, and morphology) were within the normal ranges, but the SCSA values were poor and not compatible with good fertility after intercourse. Based on these results, Evenson has proposed that SCSA parameters may be independent predictors of reproductive outcome, beyond World Health Organization (WHO) parameters. On the other hand, in a recent study, Gardner et al did not find a difference in the implantation and pregnancy rates between two groups that had 16% vs 40% fragmentation rates detected by SCSA.86 This result contradicts previous findings using this technique. The SCSA measures a number of parameters:46 the DNA fragmentation index (DFI), that is, the sperm fraction with detectable denaturable singlestranded DNA, mainly due to DNA breaks; and the highly DNA stainable cells (HDSs), the sperm fraction showing increased double-stranded DNA accessibility to acridine orange, mainly because of defects in the histone-to-protamine transition process. As these parameters are not correlated to each other, they represent independent aberrations of the human mature male gamete in the ejaculate. DFI has been shown to influence normally initiated pregnancy:83,84 increasing levels of DFI (⬎30%), independent of WHO standard semen parameters, were associated with a decreased probability of fathering a child. Initially, SCSA was thought to be a powerful new tool to predict ART outcome. An initial small pilot study (24 men) showed that when DFI was over 27%, no pregnancies could occur with IVF or ICSI.85 A number of subsequent studies reinforced this finding. Larson-Cook et al71 examined 89 couples



undergoing IVF/ICSI; the endpoint was clinical pregnancy as assessed by serum human chorionic gonadotropin (hCG) and ultrasonographic detection of a gestational sac. They showed that all patients who achieved a pregnancy had a DFI under 27%; however, on the other hand HDS was not correlated to pregnancy. Saleh et al87 evaluated 19 couples undergoing IUI, ten couples undergoing IVF, and four couples undergoing ICSI. In this study, levels of DFI (but not of HDS) were negatively correlated with biochemical pregnancy. The highest DFI value in biological fathers was 28%. Although the findings were quite consistent, some discrepancies emerged from these two studies: sperm concentration, percentage motility, and percentage morphology were significantly lower in patients who failed to initiate a clinical pregnancy in the Saleh et al87 study but not in the study by Larson-Cook et al.71 The fertilization rate was related to DFI in the Saleh et al87 study but not in the Larson-Cook et al71 investigation. It is noteworthy that Benchaib et al76 also showed that higher (⬎10%) sperm DNA fragmentation levels (this time evaluated by TUNEL assay on the discontinuous gradient centrifugation selected sperm) were a negative predictor of pregnancy via ICSI (but not via IVF) and no pregnancies occurred if DNA fragmentation was over 20%. This study assessed the clinical pregnancy rate in a cohort of 50 IVF patients and 54 ICSI patients by positive serum hCG and ultrasound detection of a fetal heartbeat. The enthusiasm generated by these original studies regarding the existence of an upper DNA fragmentation threshold above which no pregnancy can be obtained after ART, has cooled down as more investigations have been published. Firstly, Gandini et al,88 in a study involving 34 couples (12 IVF and 22 ICSI), did not note any difference between patients initiating pregnancies or not, and, above all, they reported healthy full-term pregnancies even with high levels of DFI (up to 66.3%). Pregnancy rates were 25% for IVF and 40.9% for ICSI. HDS did not correlate with pregnancy or live birth rate, and no association was found between the SCSA parameters and the fertility rate. Secondly, Bungum et al89 investigated 306 consecutive couples undergoing

HPE_Chapter25.qxp

7/14/2007

5:52 PM

Page 331


ART (131 IUI, 109 IVF, and 66 ICSI), taking into account biochemical pregnancy (positive serum hCG), clinical pregnancy (intrauterine gestational sac with a heartbeat 3 weeks after a positive hCG test), and delivery. The delivery rate was 15.3% after IUI, 28.4% after IVF, and 37.9% after ICSI. They reported that the chance of pregnancy/delivery after IUI was significantly higher in the group with DFI under 27% (and HDS under 10%): only one delivery resulted among the 23 men with a DFI over 27%. The combination of DFI and HDS gave a higher predictive value regarding the outcome of IUI. On the other hand, no statistically significant difference in the outcome after IVF/ICSI was noted by dividing patients according to a DFI level of 27%. However, the results after ICSI were significantly better than those after IVF: for example, as far as the group with DFI over 27% was concerned, comparing ICSI with IVF performances the authors reported higher clinical pregnancy (52.9% vs 22.2%), implantation (37.5% vs 19.4%), and delivery (47.1% vs 22.2%) rates. In addition, by restricting the analysis to IVF patients only, the group with a low DFI level under 27% consistently showed better clinical pregnancy (36.6% vs 22.2%), implantation (33.3% vs 19.4%), and delivery (29.7% vs 22.2%) rates as compared with the group of men with DFI over 27%. A more recent analysis by the same group90 assessed the previous 988 cycles of IUI, IVF, and ICSI showed that for IUI, the odds ratios (ORs) for biochemical, clinical pregnancy, and delivery were significantly lower for couples with DFI ⬎30% as compared with those with DFI ⱕ30%. No statistical difference between the outcomes of ICSI versus IVF in the group with DFI ⱕ30% was seen. In the DFI ⬎30% group, the results of ICSI were significantly better than those of IVF. Virro et al82 studied 249 couples undergoing IVF/ICSI and noted that men with DFI under 33% had a significantly greater chance of initiating a clinical pregnancy (positive hCG), lower rate of spontaneous abortions, and an increased rate of ongoing pregnancies at 12 weeks (47% vs 28%). HDS and standard WHO parameters were not related to pregnancy outcomes. Finally, a recent meta-analysis by Li et al74 showed that sperm DNA

damage, as assessed by the TUNEL assay, significantly decreases only the chance of IVF clinical pregnancy, but not that of ICSI clinical pregnancy. In addition they revealed that sperm DNA damage, when assessed by the SCSA, had no significant effect on the chance of clinical pregnancy after IVF or ICSI treatment. All of these studies demonstrate that high levels of DNA damage are still compatible with pregnancy and delivery after IVF/ICSI. The increasing number of publications in this field indicates that the relevance of sperm nuclear DNA tests is not completely black and white.91 From the ever-increasing wealth of data collected so far about the SCSA and TUNEL techniques and their predictive power in ART, we can speculate that an increased fraction of sperm showing DNA damage is certainly a negative trait that reduces the chance of fathering a child; however, a magic number or percentage of DNA damaged sperm in the population that is not compatible with pregnancy is far from being established. In addition, the studies of Bungum et al89,90 indicate that the predictive power of the SCSA seems to lose its strength from natural conception to ICSI, passing through IUI and IVF. The contradictory results for ICSI can also be explained on the basis that SCSA parameters refer to the DNA status of a whole sperm population, but the selection of just one single good-looking sperm for ICSI is biased by the intervention of a human operator on the basis of his own criteria. We know that SCSA parameters are only weakly associated with WHO standard semen parameters. The technician who performs ICSI attempts to select a sperm with normal morphology, in order to reduce the risk of introducing a sperm with DNA breaks. The selection of sperm from a subpopulation with no significant DNA damage, as assessed by the SCSA, cannot be excluded. This indicates that the sample tested by the sperm DNA testing techniques should also reflect the way in which they will be used. Therefore, while assessing DNA damage levels in the unprocessed semen sample may be predictive of natural fertility and IUI outcome, it may be more diagnostically relevant to use the prepared sample for assessment in order to predict IVF or ICSI outcome.



HPE_Chapter25.qxp

7/14/2007

5:52 PM

Page 332


NOVEL TECHNIQUES TO REMOVE SPERMATOZOA WITH DAMAGED DNA

Cleansing the ejaculate of spermatozoa with abnormal DNA would therefore overcome the above problems. A number of studies have now examined the possibility of improving the efficiency of preparation techniques in order to eliminate spermatozoa with nuclear anomalies. The simplest of these is a method that is routinely used in most ART laboratories, i.e. density gradient centrifugation. This may explain why there is a better correlation between sperm DNA tests when the final prepared sperm sample is used for testing. A number of studies, including our own, have shown that using spermatozoa prepared with a density gradient centrifugation technique significantly improves the quality of spermatozoa in the preparation. These studies indicate that density gradient centrifugation can enrich the sperm population by separating out those with nicked DNA and with poorly condensed chromatin.70,92–94 A number of novel techniques have now been proposed that may assist in limiting the chance of selecting an abnormal spermatozoon prior to ICSI. Bartoov et al95 reported that they were able to achieve a pregnancy rate of 58% in a small group of patients who had previously failed at least five consecutive routine cycles of IVF and ICSI, by selecting spermatozoa under high magnification. Aitken’s group96,97 report a novel electrophoretic sperm isolation technique for preparing functional human spermatozoa free of significant DNA damage. Briefly, the separation system consists of a cassette comprising two chambers. Semen is introduced into one chamber and a current applied that within seconds leads to a purified suspension of spermatozoa collecting on the other side of the chamber. Suspensions generated by the electrophoretic separation technique contain motile, viable, morphologically normal spermatozoa that exhibit lower levels of DNA damage. Another recently reported technique is one which sorts spermatozoa that possess the apoptotic marker protein annexin V. Said et al63 used a magnetic cell sorting 

technique to sort spermatozoa on the basis of annexin V labeling, producing two sperm fractions: an annexin V-negative (non-apoptotic) and annexin V-positive (apoptotic).63 They postulate that this technique may benefit the success of assisted reproduction techniques. A final selection technique has recently been reported by Huszar et al. They had previously reported that sperm that are able to bind to hyaluronic acid (HA) are mature and have completed the spermiogenetic process of sperm plasma membrane remodeling, cytoplasmic extrusion, and nuclear histone-protamine replacement.98 Testing this technique has shown that the HA bound sperm have lower rates of cytoplasmic, nuclear, and chromosomal abnormalities.98,99 They concluded that using this method to select sperm for ICSI might reduce the potential genetic complications and adverse public health effects of ICSI.99

CONCLUSION

There is accumulating evidence linking sperm nuclear DNA anomalies to poor reproductive outcome in relation to ART (Table 25.1). The concern as to the significance of this relationship is definitely greater in relation to ICSI. The tests currently available only provide an inkling of the impact of sperm nuclear DNA abnormalities on outcomes. More research is needed to improve our current knowledge in relation to the DNA anomalies in

Table 25.1 Selected studies indicating whether abnormalities in sperm DNA impact on reproductive outcome in animals and human Animal

Fertilization Embryo development Miscarriages and pregnancy Future generations

Human

Yes

No

Yes

No

— 103–105

100–102 —

73 81,82

74 —

102,106

—

83,84,107

—

7,106

—

?

—

HPE_Chapter25.qxp

7/14/2007

5:52 PM

Page 333


spermatozoa, how to detect them more accurately and how they may relate to failed reproductive outcomes. Additionally, more standardized, large-scale trials are needed to assess the predictive value of sperm DNA fragmentation techniques as useful pregnancy predictors in ART. REFERENCES 1. Wood JW. Fecundity and natural fertility in humans. Oxf Rev Reprod Biol 1989; 11: 61–109. 2. Spradling AC. Stem cells: more like a man. Nature 2004; 428(6979): 133–4. 3. SART 2003 National Report. 2003. https://www.Sartcorsonline.com/ rptCSR_PublicMultYear.aspx?ClinicPKID=() 4. Hales BF, Barton TS, Robaire B. Impact of paternal exposure to chemotherapy on offspring in the rat. J Natl Cancer Inst Monogr 2005; (34): 28–31. 5. Robaire B, Hales BF. Mechanisms of action of cyclophosphamide as a male-mediated developmental toxicant. Adv Exp Med Biol 2003; 518: 169–80. 6. Hales BF, Robaire B. Paternal exposure to drugs and environmental chemicals: effects on progeny outcome. J Androl 2001; 22(6): 927–36. 7. Anway MD, Cupp AS, Uzumcu M, Skinner MK. Epigenetic transgenerational actions of endocrine disruptors and male fertility. Science 2005; 308(5727): 1466–9. 8. Parker L, Pearce MS, Dickinson HO, Aitkin M, Craft AW. Stillbirths among offspring of male radiation workers at Sellafield nuclear reprocessing plant [Comments]. 1999; 354(9188): 1407–14. 9. Abrahamson S, Tawn EJ. Risk of stillbirth in offspring of men exposed to ionising radiation. J Radiol Prot 2001; 21(2): 133–44. 10. Reichenberg A, Gross R, Weiser M et al. Advancing paternal age and autism. Arch Gen Psychiatry 2006; 63(9): 1026–32. 11. Polednak AP. Paternal age in relation to selected birth defects. Hum Biol 1976; 48(4): 727–39. 12. Malaspina D, Reichenberg A, Weiser M et al. Paternal age and intelligence: implications for age-related genomic changes in male germ cells. Psychiatr Genet 2005; 15(2): 117–25. 13. Malaspina D, Brown A, Goetz D et al. Schizophrenia risk and paternal age: a potential role for de novo mutations in schizophrenia vulnerability genes. CNS Spectr 2002; 7(1): 26–9. 14. Moloney DM, Slaney SF, Oldridge M et al. Exclusive paternal origin of new mutations in Apert syndrome. Nat Genet 1996; 13(1): 48–53. 15. Irvine DS, Twigg JP, Gordon EL et al. DNA integrity in human spermatozoa: relationships with semen quality. J Androl 2000; 21(1): 33–44. 16. Seli E, Sakkas D. Spermatozoal nuclear determinants of reproductive outcome: implications for ART. Hum Reprod Update 2005; 11(4): 337–49. 17. Acharyya S, Kanjilal S, Bhattacharyya AK. Does human sperm nuclear DNA integrity affect embryo quality? Indian J Exp Biol 2005; 43(11): 1016–22. 18. Aitken RJ, Sawyer D. The human spermatozoon – not waving but drowning. Adv Exp Med Biol 2003; 518: 85–98. 19. Erenpreiss J, Spano M, Erenpreisa J, Bungum M, Giwercman A. Sperm chromatin structure and male fertility: biological and clinical aspects. Asian J Androl 2006; 8(1): 11–29.

20. Agarwal A, Said TM. Role of sperm chromatin abnormalities and DNA damage in male infertility. Hum Reprod Update 2003; 9(4): 331–45. 21. Sakkas D, Seli E, Manicardi GC, Nijs M, Ombelet W, Bizzaro D. The presence of abnormal spermatozoa in the ejaculate: did apoptosis fail? Hum Fertil (Camb) 2004; 7(2): 99–103. 22. Hewitson L, Simerly C, Schatten G. Cytoskeletal aspects of assisted fertilization. Semin Reprod Med 2000; 18(2): 151–9. 23. Hewitson L, Dominko T, Takahashi D et al. Unique checkpoints during the first cell cycle of fertilization after intracytoplasmic sperm injection in rhesus monkeys. Nat Med 1999; 5(4): 431–3. 24. St John J, Sakkas D, Dimitriadi K et al. Failure of elimination of paternal mitochondrial DNA in abnormal embryos [Letter]. Lancet 2000; 355(9199): 200. 25. St John JC, Lloyd R, El Shourbagy S. The potential risks of abnormal transmission of mtDNA through assisted reproductive technologies. Reprod Biomed Online 2004; 8(1): 34–44. 26. Doherty A, Mann M, Tremblay K, Bartolomei M, Schultz R. Differential effects of culture on imprinted H19 expression in the preimplantation mouse embryo. Biol Reprod 2000; 62(6): 1526–35. 27. Young L, Fernandes K, McEvoy T et al. Epigenetic change in IGF2R is associated with fetal overgrowth after sheep embryo culture. Nat Genet 2001; 27: 153–4. 28. Manning M, Lissens W, Bonduelle M et al. Study of DNA-methylation patterns at chromosome 15q11-q13 in children born after ICSI reveals no imprinting defects. Mol Hum Reprod 2000; 6(11): 1049–53. 29. DeBaun MR, Niemitz EL, Feinberg AP. Association of in vitro fertilization with Beckwith-Wiedemann syndrome and epigenetic alterations of LIT1 and H19. Am J Hum Genet 2003; 72(1): 156–60. 30. Gicquel C, Gaston V, Mandelbaum J, Siffroi JP, Flahault A, Le Bouc Y. In vitro fertilization may increase the risk of Beckwith-Wiedemann syndrome related to the abnormal imprinting of the KCN1OT gene. Am J Hum Genet 2003; 72(5): 1338–41. 31. Maher ER, Brueton LA, Bowdin SC et al. Beckwith-Wiedemann syndrome and assisted reproduction technology (ART). J Med Genet 2003; 40(1): 62–4. 32. Halliday J, Oke K, Breheny S, Algar E, J Amor D. BeckwithWiedemann syndrome and IVF: a case-control study. Am J Hum Genet 2004; 75(3): 526–8. 33. Cox GF, Burger J, Lip V et al. Intracytoplasmic sperm injection may increase the risk of imprinting defects. Am J Hum Genet 2002; 71(1): 162–4. 34. Orstavik KH, Eiklid K, van der Hagen CB et al. Another case of imprinting defect in a girl with Angelman syndrome who was conceived by intracytoplasmic semen injection. Am J Hum Genet 2003; 72(1): 218–9. 35. Edwards RG, Ludwig M. Are major defects in children conceived in vitro due to innate problems in patients or to induced genetic damage? Reprod Biomed Online 2003; 7(2): 131–8. 36. Manning M, Lissens W, Weidner W, Liebaers I. DNA methylation analysis in immature testicular sperm cells at different developmental stages. Urol Int 2001; 67(2): 151–5. 37. Marques CJ, Carvalho F, Sousa M, Barros A. Genomic imprinting in disruptive spermatogenesis. Lancet 2004; 363(9422): 1700–2. 38. Rousseaux S, Caron C, Govin J, Lestrat C, Faure AK, Khochbin S. Establishment of male-specific epigenetic information. Gene 2005; 345(2): 139–53. 39. Bianchi PG, Manicardi GC, Bizzaro D, Bianchi U, Sakkas D. Effect of deoxyribonucleic acid protamination on fluorochrome staining and in situ nick-translation of murine and human mature spermatozoa. Biol Reprod 1993; 49(5): 1083–8.



HPE_Chapter25.qxp

7/14/2007

5:52 PM

Page 334


40. Gorczyca W, Traganos F, Jesionowska H, Darzynkiewicz Z. Presence of DNA strand breaks and increased sensitivity of DNA in situ to denaturation in abnormal human sperm cells: analogy to apoptosis of somatic cells. Exp Cell Res 1993; 207(1): 202–5. 41. Manicardi GC, Bianchi PG, Pantano S et al. Presence of endogenous nicks in DNA of ejaculated human spermatozoa and its relationship to chromomycin A3 accessibility. Biol Reprod 1995; 52(4): 864–7. 42. Lewis SE, Aitken RJ. DNA damage to spermatozoa has impacts on fertilization and pregnancy. Cell Tissue Res 2005; 322(1): 33–41. 43. Sakkas D, Seli E, Bizzaro D, Tarozzi N, Manicardi GC. Abnormal spermatozoa in the ejaculate: abortive apoptosis and faulty nuclear remodelling during spermatogenesis. Reprod Biomed Online 2003; 7(4): 428–32. 44. Huckins C. The morphology and kinetics of spermatogonial degeneration in normal adult rats: an analysis using a simplified classification of the germinal epithelium. Anat Rec 1978; 190(4): 905–26. 45. Sun JG, Jurisicova A, Casper RF. Detection of deoxyribonucleic acid fragmentation in human sperm: correlation with fertilization in vitro. Biol Reprod 1997; 56(3): 602–7. 46. Evenson DP, Larson KL, Jost LK. Sperm chromatin structure assay: its clinical use for detecting sperm DNA fragmentation in male infertility and comparisons with other techniques. J Androl 2002; 23(1): 25–43. 47. Zenzes MT, Puy LA, Bielecki R, Reed TE. Detection of benzo[a] pyrene diol epoxide-DNA adducts in embryos from smoking couples: evidence for transmission by spermatozoa. Mol Hum Reprod 1999; 5(2): 125–31. 48. Shen HM, Chia SE, Ni ZY, New AL, Lee BL, Ong CN. Detection of oxidative DNA damage in human sperm and the association with cigarette smoking. Reprod Toxicol 1997; 11(5): 675–80. 49. Zenzes MT, Bielecki R, Reed TE. Detection of benzo(a)pyrene diol epoxide-DNA adducts in sperm of men exposed to cigarette smoke. Fertil Steril 1999; 72(2): 330–5. 50. Oldereid NB, Angelis PD, Wiger R, Clausen OP. Expression of Bcl-2 family proteins and spontaneous apoptosis in normal human testis. Mol Hum Reprod 2001; 7(5): 403–8. 51. Spierings DC, de Vries EG, Vellenga E, de Jong S. The attractive Achilles heel of germ cell tumours: an inherent sensitivity to apoptosis-inducing stimuli. J Pathol 2003; 200(2): 137–48. 52. Zitzmann M, Rolf C, Nordhoff V et al. Male smokers have a decreased success rate for in vitro fertilization and intracytoplasmic sperm injection. Fertil Steril 2003; 79(Suppl 3): 1550–4. 53. Orth JM, Gunsalus GL, Lamperti AA. Evidence from Sertoli celldepleted rats indicates that spermatid number in adults depends on numbers of Sertoli cells produced during perinatal development. Endocrinology 1988; 122(3): 787–94. 54. Sharpe RM. Regulation of spermatogenesis. In: Knobil E, Neill JD, eds. The Physiology of Reproduction. New York: Raven Press, 2006: 1363–434. 55. Knudson CM, Tung KS, Tourtellotte WG, Brown GA, Korsmeyer SJ. Bax-deficient mice with lymphoid hyperplasia and male germ cell death. Science 1995; 270(5233): 96–9. 56. Rodriguez I, Ody C, Araki K, Garcia I, Vassalli P. An early and massive wave of germinal cell apoptosis is required for the development of functional spermatogenesis. EMBO J 1997; 16(9): 2262–70. 57. Sakkas D, Mariethoz E, St John JC. Abnormal sperm parameters in humans are indicative of an abortive apoptotic mechanism linked to the Fas-mediated pathway. Exp Cell Res 1999; 251(2): 350–5. 58. Barroso G, Morshedi M, Oehninger S. Analysis of DNA fragmentation, plasma membrane translocation of phosphatidylserine and oxidative stress in human spermatozoa [In Process Citation]. Hum Reprod 2000; 15(6): 1338–44.



59. Grunewald S, Paasch U, Wuendrich K, Glander HJ. Sperm caspases become more activated in infertility patients than in healthy donors during cryopreservation. Arch Androl 2005; 51(6): 449–60. 60. Sakkas D, Moffatt O, Manicardi GC, Mariethoz E, Tarozzi N, Bizzaro D. Nature of DNA damage in ejaculated human spermatozoa and the possible involvement of apoptosis. Biol Reprod 2002; 66(4): 1061–7. 61. Oehninger S, Morshedi M, Weng SL, Taylor S, Duran H, Beebe S. Presence and significance of somatic cell apoptosis markers in human ejaculated spermatozoa. Reprod Biomed Online 2003; 7(4): 469–76. 62. Cayli S, Sakkas D, Vigue L, Demir R, Huszar G. Cellular maturity and apoptosis in human sperm: creatine kinase, caspase-3 and Bcl-XL levels in mature and diminished maturity sperm. Mol Hum Reprod 2004; 10(5): 365–72. 63. Said T, Agarwal A, Grunewald S et al. Selection of non-apoptotic spermatozoa as a new tool for enhancing assisted reproduction outcomes: an in vitro model. Biol Reprod 2005; 74: 530–7. 64. McPherson S, Longo FJ. Chromatin structure-function alterations during mammalian spermatogenesis: DNA nicking and repair in elongating spermatids. Eur J Histochem 1993; 37(2): 109–28. 65. McPherson SM, Longo FJ. Nicking of rat spermatid and spermatozoa DNA: possible involvement of DNA topoisomerase II. Dev Biol 1993; 158(1): 122–30. 66. Bizzaro D, Manicardi G, Bianchi PG, Sakkas D. Sperm decondensation during fertilisation in the mouse: presence of DNase I hypersensitive sites in situ and a putative role for topoisomerase II. Zygote 2000; 8(3): 197–202. 67. McPherson SM, Longo FJ. Localization of DNase I-hypersensitive regions during rat spermatogenesis: stage-dependent patterns and unique sensitivity of elongating spermatids. Mol Reprod Dev 1992; 31(4): 268–79. 68. Marcon L, Boissonneault G. Transient DNA strand breaks during mouse and human spermiogenesis new insights in stage specificity and link to chromatin remodeling. Biol Reprod 2004; 70: 910–8. 69. Morris ID, Ilott S, Dixon L, Brison DR. The spectrum of DNA damage in human sperm assessed by single cell electrophoresis (COMET assay) and its relationship to fertilization and embryo development. Hum Reprod 2002; 17: 990–8. 70. Tomlinson MJ, Moffatt O, Manicardi GC, Bizzaro D, Afnan M, Sakkas D. Interrelationships between seminal parameters and sperm nuclear DNA damage before and after density gradient centrifugation: implications for assisted conception. Hum Reprod 2001; 16(10): 2160–5. 71. Larson-Cook KL, Brannian JD, Hansen KA, Kasperson KM, Aamold ET, Evenson DP. Relationship between the outcomes of assisted reproductive techniques and sperm DNA fragmentation as measured by the sperm chromatin structure assay. Fertil Steril 2003; 80(4): 895–902. 72. Larson KL, DeJonge CJ, Barnes AM, Jost LK, Evenson DP. Sperm chromatin structure assay parameters as predictors of failed pregnancy following assisted reproductive techniques. Hum Reprod 2000; 15(8): 1717–22. 73. Lopes S, Sun JG, Jurisicova A, Meriano J, Casper RF. Sperm deoxyribonucleic acid fragmentation is increased in poor-quality semen samples and correlates with failed fertilization in intracytoplasmic sperm injection. Fertil Steril 1998; 69(3): 528–32. 74. Li Z, Wang L, Cai J, Huang H. Correlation of sperm DNA damage with IVF and ICSI outcomes: a systematic review and meta-analysis. J Assist Reprod Genet 2006. 75. Braude P, Bolton V, Moore S. Human gene expression first occurs between the four- and eight-cell stages of preimplantation development. Nature 1988; 332: 459–61.

HPE_Chapter25.qxp

7/14/2007

5:52 PM

Page 335


76. Benchaib M, Braun V, Lornage J et al. Sperm DNA fragmentation decreases the pregnancy rate in an assisted reproductive technique. Hum Reprod 2003; 18(5): 1023–8. 77. Sakkas D, D’Arcy Y, Percival G, Sinclair L, Afnan M, Sharif K. Use of the egg-share model to investigate the paternal influence on fertilization and embryo development after in vitro fertilization and intracytoplasmic sperm injection. Fertil Steril 2004; 82(1): 74–9. 78. Janny L, Menezo YJ. Evidence for a strong paternal effect on human preimplantation embryo development and blastocyst formation. Mol Reprod Dev 1994; 38(1): 36–42. 79. Shoukir Y, Chardonnens D, Campana A, Sakkas D. Blastocyst development from supernumerary embryos after intracytoplasmic sperm injection: a paternal influence? Hum Reprod 1998; 13(6): 1632–7. 80. Dumoulin JC, Coonen E, Bras M et al. Comparison of in-vitro development of embryos originating from either conventional in-vitro fertilization or intracytoplasmic sperm injection. Hum Reprod 2000; 15(2): 402–9. 81. Seli E, Gardner DK, Schoolcraft WB, Moffatt O, Sakkas D. Extent of nuclear DNA damage in ejaculated spermatozoa impacts on blastocyst development after in vitro fertilization. Fertil Steril 2004; 82(2): 378–83. 82. Virro MR, Larson-Cook KL, Evenson DP. Sperm chromatin structure assay (SCSA) parameters are related to fertilization, blastocyst development, and ongoing pregnancy in in vitro fertilization and intracytoplasmic sperm injection cycles. Fertil Steril 2004; 81(5): 1289–95. 83. Evenson DP, Jost LK, Marshall D et al. Utility of the sperm chromatin structure assay as a diagnostic and prognostic tool in the human fertility clinic. Hum Reprod 1999; 14(4): 1039–49. 84. Spano M, Bonde JP, Hjollund HI, Kolstad HA, Cordelli E, Leter G. Sperm chromatin damage impairs human fertility. The Danish First Pregnancy Planner Study Team. Fertil Steril 2000; 73(1): 43–50. 85. Larson KL, DeJonge CJ, Barnes AM, Jost LK, Evenson DP. Sperm chromatin structure assay parameters as predictors of failed pregnancy following assisted reproductive techniques. Hum Reprod 2000; 15(8): 1717–22. 86. Gardner DK, Schoolcraft WB, Surrey ES et al. The sperm chromatin structure assay (SCSA) and its relationship to IVF outcome. Philadelphia, PA: Fertility and Sterility, 82: S192–S192 P166 Suppl. 2 SEP 2004. 87. Saleh RA, Agarwal A, Nada EA et al. Negative effects of increased sperm DNA damage in relation to seminal oxidative stress in men with idiopathic and male factor infertility. Fertil Steril 2003; 79 (Suppl 3): 1597–605. 88. Gandini L, Lombardo F, Paoli D et al. Full-term pregnancies achieved with ICSI despite high levels of sperm chromatin damage. Hum Reprod 2004; 19(6): 1409–7. 89. Bungum M, Humaidan P, Spano M, Jepson K, Bungum L, Giwercman A. The predictive value of sperm chromatin structure assay (SCSA) parameters for the outcome of intrauterine insemination, IVF and ICSI. Hum Reprod 2004; 19(6): 1401–8. 90. Bungum M, Humaidan P, Axmon A et al. Sperm DNA integrity assessment in prediction of assisted reproduction technology outcome. Hum Reprod 2007; 22(1): 174–9. 91. Spano M, Seli E, Bizzaro D, Manicardi GC, Sakkas D. The significance of sperm nuclear DNA strand breaks on reproductive outcome. Curr Opin Obstet Gynecol 2005; 17(3): 255–60.

92. Larson KL, Brannian JD, Timm BK, Jost LK, Evenson DP. Density gradient centrifugation and glass wool filtration of semen remove spermatozoa with damaged chromatin structure. Hum Reprod 1999; 14(8): 2015–9. 93. Pasteur X, Maubon I, Sabido O, Cottier M, Laurent JL. Comparison of the chromatin stainability of human spermatozoa separated by discontinuous Percoll gradient centrifugation. A flow cytometric contribution. Anal Quant Cytol Histol 1992; 14(2): 96–104. 94. Sakkas D, Manicardi GC, Tomlinson M et al. The use of two density gradient centrifugation techniques and the swim-up method to separate spermatozoa with chromatin and nuclear DNA anomalies. Hum Reprod 2000; 15(5): 1112–6. 95. Bartoov B, Berkovitz A, Eltes F. Selection of spermatozoa with normal nuclei to improve the pregnancy rate with intracytoplasmic sperm injection. N Engl J Med 2001; 345(14): 1067–8. 96. Ainsworth C, Nixon B, Aitken RJ. Development of a novel electrophoretic system for the isolation of human spermatozoa. Hum Reprod 2005; 20(8): 2261–70. 97. Ainsworth C, Nixon B, Jansen RP, Aitken RJ. First recorded pregnancy and normal birth after ICSI using electrophoretically isolated spermatozoa. Hum Reprod 2006; 22(1): 197–200. 98. Huszar G, Ozenci CC, Cayli S, Zavaczki Z, Hansch E, Vigue L. Hyaluronic acid binding by human sperm indicates cellular maturity, viability, and unreacted acrosomal status. Fertil Steril 2003; 79 (Suppl 3): 1616–24. 99. Jakab A, Sakkas D, Delpiano E et al. Intracytoplasmic sperm injection: a novel selection method for sperm with normal frequency of chromosomal aneuploidies. Fertil Steril 2005; 84(6): 1665–73. 100. Bordignon V, Smith LC. Ultraviolet-irradiated spermatozoa activate oocytes but arrest preimplantation development after fertilization and nuclear transplantation in cattle. Biol Reprod 1999; 61(6): 1513–20. 101. Harrouk W, Codrington A, Vinson R, Robaire B, Hales BF. Paternal exposure to cyclophosphamide induces DNA damage and alters the expression of DNA repair genes in the rat preimplantation embryo. Mutat Res 2000; 461(3): 229–41. 102. Ahmadi A, Ng SC. Fertilizing ability of DNA-damaged spermatozoa. J Exp Zool 1999; 284(6): 696–704. 103. Austin SM, Robaire B, Hales BF, Kelly SM. Paternal cyclophosphamide exposure causes decreased cell proliferation in cleavagestage embryos. Biol Reprod 1994; 50(1): 55–64. Erratum in: Biol Reprod 1994; 50(3): 711. 104. Harrouk W, Khatabaksh S, Robaire B, Hales BF. Paternal exposure to cyclophosphamide dysregulates the gene activation program in rat preimplantation embryos. Mol Reprod Dev 2000; 57(3): 214–23. 105. Harrouk W, Robaire B, Hales BF. Paternal exposure to cyclophosphamide alters cell–cell contacts and activation of embryonic transcription in the preimplantation rat embryo. Biol Reprod 2000; 63(1): 74–81. 106. Hales BF, Crosman K, Robaire B. Increased postimplantation loss and malformations among the F2 progeny of male rats chronically treated with cyclophosphamide. Teratology 1992; 45(6): 671–8. 107. Carrell DT, Liu L, Peterson CM et al. Sperm DNA fragmentation is increased in couples with unexplained recurrent pregnancy loss. Arch Androl 2003; 49(1): 49–55.



HPE_Chapter25.qxp

7/14/2007

5:52 PM

Page 336

HPE_Chapter26.qxp

7/13/2007

5:32 PM

Page 337

26. The sperm centriole: its effect on the developing embryo Calvin R Simerly and Christopher S Navara Centrosomes are classically defined as a pair of centrioles, structures of nine triplet microtubules without a central microtubule pair, surrounded by the pericentrosomal matrix (PCM). The PCM is the amorphous material that nucleates the microtubules and defines their intrinsic polarity (i.e. minus ends (slower turnover) anchored at each centrosome; plus ends (faster turnover) radiating outward).1,2 The past 25 years have seen an explosion in understanding basic centrosomal biology with regards to molecular constituents and the minimal structure required to promote microtubule nucleation from this structure. Rapid advances in understanding centrosomal inheritance, assembly, duplication, and segregation in a variety of cell types provide crucial clues for understanding how the centrosome mediates intracellular motility, cytoplasmic organization, and the many other cellular processes linked to centrosomal activities.3–5 These advances are beginning to translate to modern molecular medicine, where clinical challenges including infertility treatments and contraception require a greater understanding of this crucial cellular organelle. Within the fields of reproductive and developmental biology, the cellular and molecular events during fertilization are intrinsically grounded in understanding the role of the centrosome. Boveri6 recognized more than a century ago that at fertilization the sperm contributes the centrosome, the cell’s major microtubule organizing center (MTOC) and the structure that organizes the mitotic spindle poles (cleavage centers). However, the process of centrosome reduction and restoration during meiosis, fertilization, and mitosis has remained perplexing7 (Figure 26.1). In somatic cell cycles, the chromosomes, cytoplasm, and centrosome double during interphase, with each daughter cell receiving one diploid chromosome set, one centrosome, and half of the cell volume following division (Figure 26.1).

During fertilization, however, there is ambiguity about the inheritance and reconstitution of the functional zygotic centrosome (Figure 26.1). Although each gamete contributes a haploid genome at insemination, the egg provides most of the cytoplasmic contribution. Boveri first recognized that the egg typically loses the centrosome during oogenesis and that the sperm typically introduces this structure at fertilization. Yet, understanding centrosome behavior has remained a persistent problem in cell biology and an especially enigmatic one during mammalian development. The foundations for the fields of cell, molecular, developmental, and reproductive biology, provided by the study of fertilization in lower animal species (i.e. sea urchins, frogs) is being greatly extended by enhanced methods for performing in vitro fertilization (IVF) in many mammals, including humans. A century after the discovery of the centrosome, human IVF was achieved8 – more than one million IVF babies have now been born.9 The otherwise discarded gametes and embryos from IVF clinics are providing a precious and unique research resource for the modern-day centrosomal physician/scientist. As a result, reproductive errors like polyspermy (fertilization by more than one sperm) and parthenogenesis (development beginning in an activated egg without any sperm) have provided insights that subtly challenge aspects of the unipaternal centrosome inheritance theory. Promising investigations in non-human primates appear to align closely to humans, and avoid the many complexities in working with fertilized human oocytes for research.10 Such advancements have direct implications in clinical reproductive medicine, which in many ways has pioneered amazing clinical achievements in assisted reproductive technologies (ART) such as intracytoplasmic sperm injection (ICSI) that have produced thousands of human babies.11 Although fundamental 

HPE_Chapter26.qxp

7/13/2007

5:32 PM

Page 338


Somatic cells cycles Chromosome Cytoplasm Centrosomes

Gametes at fertilization Chromosome Cytoplasm Centrosomes

100%

2N

1N

>99% ?

egg 2(2N)

1N 200% >1% ?

sperm

?

1N + 1N 100% 2N 100% zygote 2N 100%

A

B

Figure 26.1 The problem of centrosome inheritance. During each cycle in somatic cells (A), the chromosomes (2N1→2N2), cytoplasm (100→200%), as well as the centrosomes (·→··), duplicate during interphase. During cell division, the chromosomes, cytoplasm, and centrosomes all split in two. Each of the gametes during fertilization (B) contributes a haploid chromosome set to the zygote (1N1 ⫹ 1N1→ 2N1), and while the egg contributes the vast bulk of the cytoplasm (⬎99%), the relative contributions of the sperm and the egg to the zygotic centrosome are not yet understood (*). Reprinted with permission from Schatten.7

research still lags in efforts to understand these clinical accomplishments, biomedical researchers are now poised to begin investigations on the cellular and molecular events that underlie ART advances. This review focuses on the role of the centrosome during fertilization, with special attention to human reproduction and development. We also consider the centrosome in devastating disease disorders that arise in human development resulting from aberrant genomic imprints. These new research 

frontiers represent unique challenges for the next generation of centrosome biologists, owing in large part to the ethical, moral, political, and religious hurdles that accompany working with human gametes. Consequently, experimental results are obtained and/or corroborated by studying non-human primate development. This is essential, ironically, because fertilization in outbred mice, hamsters, and rats – while vital for fundamental research investigations – represent rare exceptions to Boveri’s theory on centrosome inheritance.7

HPE_Chapter26.qxp

7/13/2007

5:32 PM

Page 339

THE SPERM CENTRIOLE: ITS EFFECT ON THE DEVELOPING EMBRYO

CENTROSOMES DURING HUMAN FERTILIZATION F M

A

B M M

F F

C

D F M

E

F

G

H

Figure 26.2 Microtubule and DNA organization in normal inseminated human oocytes. The meiotic spindle in mature, unfertilized human oocytes is anastral, oriented radially to the cell surface, and asymmetric, with a focused pole abutting the cortex and a broader pole facing the cytoplasm (A). No other microtubules are detected in the cytoplasm of the unfertilized human oocyte. Shortly after sperm incorporation (3–6.5 hours postinsemination), sperm astral microtubules assemble around the base of the sperm head, as the inseminated oocytes complete second meiosis and extrude the second polar body (B)–(E). M, male pronucleus; F, female midbody. The close association of the meiotic midbody identifies the female pronucleus (B)–(E). Short, sparse, disarrayed cytoplasmic microtubules can also be observed in the cytoplasm following confocal microscopic observations of these early-activated oocytes (C). As the male pronucleus continues to decondense in the cytoplasm, the microtubules of the sperm aster enlarge, circumscribing the male pronucleus (D) and (E). By 15 hours postinsemination, the centrosome splits and organizes a bipolar microtubule array that emanates from the tightly apposed pronuclei (F). The sperm tail is associated with an aster (arrow) (F). At first mitotic prophase (16.5 hours postinsemination), the male and female

Centrosome activity in spermatogenic cells centers around the generation of the sperm axoneme required for sperm motility. With the exception of rodents, where the paternal centrioles are completely dismantled, mammalian sperm reduce the pair of centrioles to a single inactive structure during spermatogenesis, the proximal centriole.12 The vast majority of the proteins of the pericentriolar material are shed in the cytoplasmic droplet during sperm maturation during migration through the epididymal tract. Conversely, oocytes lose both centrioles during oogenesis as investigated by electron microscopy, but retain a significant pool of pericentriolar proteins as observed by immunocytochemistry and Western blotting.13 Centrosomal inheritance during human fertilization14 is depicted from the organization of microtubules and DNA observed in discarded human oocytes (Figure 26.2) and mirrors the inheritance pathway found in most animals, except rodents.7,15,16 Microtubules are only observed in the metaphasearrested second meiotic spindle in the unfertilized human oocyte (Figure 26.2A), unlike the rodent that retains multiple cytoplasmic microtubule asters (‘cytasters’). Within 6 hours postinsemination, a small microtubule aster emanates from the introduced sperm proximal centriole (Figure 26.2B and C). The activated oocyte extrudes the second polar body, observed here attached to the developing female pronucleus by the microtubule based midbody structure (Figure 26.2B and C). The enlarged sperm centrosome (the ‘zygote centrosome’) nucleates microtubules that assemble the first microtubulebased structure in the fertilized egg – the sperm aster. The sperm aster in fertilized human oocytes is

chromosomes condense separately as a bipolar array of microtubules marks the developing first mitotic spindle poles (G). By prometaphase, when the chromosomes begin to align on the metaphase equator, a barrel-shaped, anastral spindle forms in the cytoplasm (H). The sperm axoneme remains associated with one small aster found at one of the spindle poles (arrow) (H). Bar ⫽ 10 ␮m. Reprinted with permission from Simerly et al.14



HPE_Chapter26.qxp

7/13/2007

5:32 PM

Page 340


the typical radially arrayed monaster juxtaposed to the sperm nucleus (Figure 26.2D and E) (which is called the ‘male pronucleus’) after the sperm chromatin has decondensed within the egg cytoplasm. These microtubules elongate throughout the cytoplasm during early development, some contacting the female pronucleus and initiating pronuclear migration (Figure 26.2E and F). As in most animal eggs, the sperm tail enters the egg (Figure 26.2F). Often, one or two punctate foci are found at the center of the sperm aster exactly at the junction between the sperm axoneme and the male pronuclear surface which correspond to the sperm centriole(s).15 The reconstituted zygotic centrosome duplicates and splits during late interphase, presumably under the control of cell cyle regulatory machinery17,18 (Figure 26.2F). Mitotic prophase commences with the separate chromosomal condensation of the male and female pronuclei, as the zygotic centrosomes nucleate the microtubules of the bipolar mitotic spindle apparatus (Figure 26.2G). By late prometaphase, the condensing parental chromosomes align along the equator of the bipolar, anastral mitotic spindle (Figure 26.2H), completing the fertilization process in humans. These data confirm that humans inherit their centrosomes from their fathers. Evidence supporting the sperm centriolar complex as the foundation for zygotic centrosome formation in humans comes from studies on polyspermic fertilization and parthenogenesis.16,19 As shown in Figure 26.3, when two sperm enter the oocyte, each paternal centriolar complex organizes a sperm aster at the base of the sperm head (Figure 26.3A and B). Conversely, parthenogenetic (artificial) activation of oocytes, in which no contribution of the paternal centrosome is provided, causes random, disarrayed interphase microtubule patterns (Figure 26.3C and D). Collectively, these data reinforce the observation that the centrosome is paternally inherited in humans.7,14

CENTROSOME DYSFUNCTION AND HUMAN INFERTILITY

Human gametes discarded from infertility clinics performing ART and donated for research provide 

Mt

DNA

M

F M A

B M

F

C

D

Figure 26.3 Microtubules (Mt) and DNA organization during human polyspermic fertilization and parthenogenesis. (A) and (B) During dispermic fertilization, a microtubule-based sperm aster assembles at the base of each incorporated sperm head (M) (B) from the site of the paternal proximal centriole (arrows) (A). (C) and (D) Activation of the human oocyte without sperm penetration (parthenogenesis) results in random, disarrayed cortical microtubule assembly during interphase. F, female pronucleus. Bar ⫽ 10 ␮m. Reprinted images with permission from Simerly et al.14 The white bar dividing A and B indicates the two focal planes.

important insights into the diagnosis of novel forms of human infertility, especially male factor infertility.14,20,21 Centrosome dysfunction after the sperm enters the oocyte is being recognized as a new cause of male factor infertility during human reproduction (Figure 26.4).14,21–26 For instance, when the assembled sperm centrosome after incorporation is dysfunctional in microtubule assembly or organization, the sperm aster may fail to form or be so poorly organized as to be inconsistent with the ability to conduct pronuclear apposition. Sperm that fail to activate the oocyte and initiate exit from metaphase arrest often have underdeveloped microtubule asters adjacent to incorporated sperm heads (Figure 26.4A and B). Other sperm may nucleate multiple asters (Figure 26.4C and D), or can fail to correctly organize the sperm astral microtubules

HPE_Chapter26.qxp

7/13/2007

5:32 PM

Page 341


after incorporation resulting in oocytes that fail the fertilization process and arrest in early development (Figure 26.4E and F). About 25% of the human oocytes classified as ‘failed to fertilize’ demonstrated successful sperm incorporation, but with the oocytes failing to properly assemble sperm astral microtubules and/or undergo pronuclear formation.14,25 These data show that the formation and functioning of the sperm aster is essential during human fertilization, and that naturally occurring defects are causes of fertilization failures. A study on the phenotypic expression of bull centrosomes further illustrates the clinical, agricultural, and fundamental importance of the centrosome during fertilization.27 Sperm from bulls were selected and classified as superb, average, or subfertile based on estrous non-return rates of 2500 cows who underwent artificial insemination in field studies, as well as blastocyst production rates

M

A

B

M

C

D

M E

F 120

A B

B

a 80

c

60 40

2

b

n = 81

n = 91

n = 100

Bull A

Bull B

Bull C

1

Sperm aster quality

Sperm aster size (µm)

100

20

G

0

0

Figure 26.4 Microtubule (A), (C) and (E) and DNA (B), (D) and (F) organization in ‘fertilization failures’ and phenotypic variations among paternal centrosomes. Common examples of the stages at which human fertilization arrests after sperm incorporation include incomplete sperm aster assembly (A) and (B); inappropriate detachment of the sperm aster and tail from the male pronucleus (M), as well as androgenesis (C) and (D); and

disarrayed sperm astral organization (E) and (F). (A) and (B) A truncated aster at the base of the sperm head has formed, but not enlarged, by 24 hours postinsemination and the oocyte remains arrested at a late meiotic phase. The incorporated sperm tail is observed (arrow) (A). (C) and (D) The inseminated oocyte displays a single pronucleus separated from the nearby sperm tail (arrow) (C), indicating the presence of a male, but not female, pronucleus. This example of androgenesis may have occurred after the loss of the maternal chromosomes during first or second polar body extrusion. Two small asters are visible; one in association with the incorporated sperm tail (arrow), while the other is detected at the male pronucleus (arrowhead) (C). (E) and (F) A defect in microtubule aster organization around the decondensing male pronucleus (E). (G) Comparison of sperm aster size and quality in three bulls of known fertility. Diameter of the sperm aster at its largest plane was measured using the confocal microscope. A quality score was also given to each aster. The bull with the highest field fertility and in vitro fertility (bull A) also had the largest and bestorganized sperm asters with averages of 101.4 ␮m and 1.8, respectively. Bull B had an average sperm aster diameter of 78.2 ␮m and an aster quality score of 1.4. The bull with the worst in vitro fertility (bull C) had the smallest (77.9 ␮m) and most poorly organized sperm asters (1.2). These data represent five repetitions: comparisons were made using the protected means of the least squares method. Different letters indicate significant differences (p ⫽ 0.025); bars indicate standard error. Bar ⫽ 10 ␮m. Images (A)–(F) reprinted with permission from Simerly et al.14 Graph reprinted from Navara et al.27



HPE_Chapter26.qxp

7/13/2007

5:32 PM

Page 342


following IVF. Using randomized bovine oocytes, each bull sperm set was used for IVF and zygotes fixed at a selected timepoint. As shown in Figure 26.4G, the organization and size of the sperm aster varies according to the bull sperm, suggesting that the quality or quantity of the sperm centrosome directly affects the success and speed of fertilization, and is correlated with the frequency of live births. Perhaps variations in centrosomal vigor occur as is found in other inherited components, a view strengthened by recent observation of impaired cat embryonic development in vitro correlated with poor centrosomal function of cat testicular spermatozoa following ICSI.28 Collectively, these investigations have the potential to be developed into novel screens for male fertility.7,19,22 Most sperm assays examine parameters – motility, morphology, and counts – that are factors more geared to successful fusion of the sperm and egg plasma membranes. Yet, fertilization is not successfully concluded until the sperm and egg genomes align at metaphase of first mitosis, and this requires the proper formation and functioning of the zygotic centrosome and sperm aster.29

DIAGNOSING MALE INFERTILITY BY CENTROSOME FUNCTION ASSAYS

Zygotic centrosome formation as an early critical step towards the accurate completion of the fertilization process is a multistep pathway occurring between the end of second meiosis and the transition into interphase of the first cell cycle. Central to this process is assembling microtubules in the proper organization to form the sperm aster that can quickly direct proper pronuclear migration. Later, the zygotic centrosome, duplicated under cell cycle control, will define the site of first bipolar mitotic spindle assembly within the activated cytoplasm and participates in spindle organization by serving as a dominant MTOC at the spindle poles.7 Understanding centrosome reconstitution during fertilization is inherently important for exploring the molecular components necessary for determining centrosome parental origin and function.7,19 

Cell-free cytoplasmic extracts obtained from cytostatic factor (CSF)-arrested Xenopus laevis oocytes have effectively explored centriole formation and microtubule assembly in vitro.30–32 These pioneering studies demonstrated that the crucial constituents necessary for centrosome construction and reproduction reside in the oocyte. With regards to the relative parental contributions to the zygote’s centrosome, the Xenopus studies demonstrate that the sperm centrosome contains conserved centrosomal constituent proteins like centrin (a ubiquitous Ca2⫹-binding protein of centrosomes) and pericentrin (a 220 kDa component of the centrosomal matrix),30,33 but undetectable amounts of ␥-tubulin, a rare, invariant constituent of the centrosome required for microtubule nucleation and for defining the intrinsic polarity of assembled microtubules from the centrosome.5 After exposure to Xenopus egg extracts, both ␥-tubulin and phosphorylated epitopes are detected on the sperm centrosome. Careful experimental analysis of sperm in egg extracts suggests that the Xenopus sperm contributes a structure capable of binding maternal ␥-tubulin, does not require assembled microtubules or microfilaments, but does require egg extract and is ATP-dependent.30–32 The sperm centrosome thus becomes competent for nucleating microtubule growth into sperm asters in vitro. Analysis of human and bovine sperm in X. laevis CSF-arrested extracts provides a basis for studying the assembly of a zygotic centrosome capable of nucleating and organizing microtubules in vitro.19,34 Mammalian sperm exposed to increased calcium levels, plasma membrane destabilization, and disulfide bond reduction unveils paternal ␥-tubulin and other centrosomal protein binding sites, concomitant with the onset of pronuclear decondensation. This ‘procentrosome’ structure is thus primed to attract and bind maternal ␥-tubulin from the egg’s cytoplasmic pool. Conversely, other paternal centrosomal proteins predicted to be critical for the reorganization of the sperm centrosomal complex following insemination (i.e. centrin) are modified following exposure to the egg’s cytoplasm.33,35 Exposure to an elevated kinase activity within the meiotic cytoplasm then shifts the microtubule

HPE_Chapter26.qxp

7/13/2007

5:32 PM

Page 343


dynamics to a state conducive to nucleation and polymerization. A model for mammalian sperm reconstitution is presented in Figure 26.5.7 In the mature human sperm, centrin is found on the proximal centriole (Figure 26.5A), with the doublet microtubules of the sperm tail anchored to the triplet microtubules of the centriole(s). The centrosome is probably not phosphorylated, but extensively crosslinked by disulfide bonds that mask the presence of paternal ␥-tubulin. After sperm incorporation, a functional zygotic centrosome is quickly assembled (Figure 26.5B). Paternal centrin is lost, perhaps caused by the binding of calcium ions released during oocyte activation, and this triggers a calcium-induced transformation of the proximal centriole, with its axial microtubules, into a structure resembling a functioning centrosome that can participate in the organization of radial arrays of microtubules post-insemination. The sperm centriolar complex is predicted to be a coiled-coil structure,3,7 and the lattice structure is unravelled by the reduction of disulfide bonds through the reducing environment of the oocyte’s cytoplasm. This exposes minute amounts of paternal ␥-tubulin and probably additional binding sites onto which maternal ␥-tubulin complexes can attach. These events, along with centrosome phosphorylation, lead to the nucleation of the microtubules that can assemble the sperm aster. Clinical assays for the initial molecular characterization of the human sperm centrosome have been proposed (Figure 26.5).7 Examining microtubule assembly in vitro from the disulfide-reduced (i.e. ‘primed’) sperm centrosome can be assayed using Xenopus egg extracts in combination with polymerization-competent rhodamine-tagged tubulin protein.19 This prospective assay may measure the ability of a population of a patient’s sperm to nucleate microtubules after sperm incorporation, identifying men who are poor candidates for ART procedures, prior to couples’ undergoing arduous procedures of ART such as ICSI. Examining microtubule assembly and centrosome function after microinjecting human sperm into mature bovine or rabbit oocytes is also a potentially

useful prospective centrosomal assay system.22 This test permits observation of pronuclear apposition mediated by a functional human sperm centrosome within a living egg, as opposed to just microtubule assembly in egg extracts. Research has shown that the recipient oocyte must be from a species other than rodents, i.e. species that follow the paternal inheritance of the centrosome.36 The zona-free hamster oocyte sperm penetration test, for assaying human male infertility, is a uniquely inappropriate model for the investigation and diagnosis of impaired sperm centrosome function of human sperm, since hamster oocytes retain their maternal centrosomes from oogenesis.23 Instead, oocytes from animals like the rabbit or cow, which support paternal centrosomal functioning, are more relevant models to investigate human centrosome reconstitution, sperm aster formation, and sperm-mediated pronuclear apposition. Centrosome microinjection therapy could theoretically overcome defective centrosomes responsible for fertilization arrest in certain types of male infertility.7,19 However, research suggests that only centrosomes introduced from intact sperm will complete the entire fertilization process and correctly segregate their chromosomes at first cell division,37 indicating that centrosome numbers and positioning at the pronuclear surface is a critical parameter for completion of fertilization. While still speculative, if successful as a therapeutic treatment, the resulting embryo would have two fathers: a genomic father and a centrosomal one.

POLYSPERMY AND THE ‘DISPERMY HYPOTHESIS’ FOR THE ORIGINS OF GENOMIC IMPRINTED DISORDERS IN HUMANS

Polyspermy is invaluable for testing the relative parental contributions of the centrosome, since the paternal contribution is multiplied. In most animals, dispermic insemination introduces two centrosomes that duplicate and separate at mitosis to form tetrapolar spindles. The resultant embryos, however, are aneuploid because the triploid chromosome 

HPE_Chapter26.qxp

7/13/2007

5:32 PM

Page 344


Sperm centrosome

Zygote centrosome

Centrin γ-Tubulin Pericentrin, etc. Phosphorylation Disulfide bonds Sulthydryl groups

A

B

Human

Human + Frog extract

C

No extract

D

Figure 26.5 Molecular dissection of the human sperm centrosome and its reconstruction in the zygote leading to assembly of sperm astral microtubules. (A) The human sperm centrosome has centrin concentrated in one or two focal sites, corresponding to the centrioles. ␥-Tubulin is not apparent in mature human sperm, but becomes detectable after ‘centrosomal priming’ of the sperm with disulfide reducing agents; this is a novel type of cytoplasmic capacitation. ␥-Tubulin is also detectable on Western blots with intact or sonicated human sperm. The centrosome is not phosphorylated and the sperm tail microtubules extend from a centriole. The coiledcoil infrastructure of the centrosome probably anchors the centrosome to the sperm nucleus and regulates the exposure of, and binding sites for, ␥-tubulin. In the zygote centrosome (B), after permeabilization and incubation in extracts from Xenopus oocytes, the human sperm becomes phosphorylated and heavily immunoreactive with antibodies to ␥-tubulin. The ␥-tubulin found in the human sperm is probably a combination of some paternal (light) and largely maternal protein (dark). The binding of calcium ions, released during the transient increase during egg activation, to centrin is predicted to result later in a centrin-induced severing of the doublet sperm tail microtubules from the triplet microtubules of the centriole. Perhaps the severing of the tail microtubules from the basal body frees the basal body complex so that it can bind additional ␥-tubulin and undergo transformation into a centriole. The coiled-coil domains of the centrosome are drawn as unraveling, expanding, and averting in the zygote; this exposes paternal ␥-tubulin and also exposes binding sites for maternal ␥-tubulin. The halo of ␥-tubulin nucleates the microtubules, which assemble into the sperm aster. (C) and (D). Human sperm exposed to 5 ␮mol/l ionomycin and primed with 5 ␮mol/l dithiothreitol, (DTT) demonstrate assembly of microtubules in vitro from the centrosomal region after 40–60 minutes of incubation in CSF-arrested Xenopus



HPE_Chapter26.qxp

7/13/2007

5:32 PM

Page 345


complement is abnormally divided into the resultant four blastomeres at the end of cell division. Such experimental evidence clearly supports the hypothesis that only one centrosome – duplicated under cell cycle control – is required for correctly forming a bipolar spindle that can accurately segregate the chromosomes.2,17,18 In mammals, however, polyspermy experiments argue against this central dogma on the universal contribution of the sperm centrosome. Rodents violate the notion of the paternal centrosome contribution at fertilization, as both the distal and proximal sperm centrioles degenerate during spermatogenesis12 and no sperm aster assembles at the base of the sperm head in the cytoplasm following incorporation. Furthermore, di- or trispermic mouse zygotes divide from one to two,38 indicating no dominant sperm centrosomal contribution during mouse fertilization. In most other mammals, including marsupials, cows, sheep, pigs, rabbits, monkeys, and humans, supernumerary sperm asters form after polyspermy.7,39–42 However, as shown in human fertilization, these dispermic zygotes can divide from one-cell into two-, three-, or even four-cell embryos.42,43 As shown in Figure 26.6,19 dispermic human zygotes at mitosis assemble bipolar metaphase spindles in the presence of supernumerary centrosomes. At prophase, a disorganized multipolar spindle with four foci of ␥-tubulin assembles first (Figure 26.6A–C), which surprisingly resolves into a bipolar spindle with aligned chromosomes at mitotic metaphase (Figure 26.6D–F) and divides one to two cells. These observations in humans may reflect requirements of non-centrosomal components (i.e. molecular motors and spindle matrix proteins) necessary for bipolar spindle assembly in somatic cells.1 Analysis of parthenogenetic development in primate oocytes supports this view (discussed below).

Dispermic fertilizations in humans are observed frequently.42,43 Diandric triploidy is one of the consequences, although other mis-segregations leading to chromosome mosaicism can be found. 2N/3N mixoploidies can give rise to live births but with a variety of developmental disorders.44,45 Diploid sperm, the result of male meiotic error, can also produce triploidy.46 Studies of postzygotic diploidization in triploid embryos47 have suggested the origins of disorders arising from genomic imprinting errors. Mammals have the paternal and maternal chromosomes specifically modified by DNA-methylation of ‘imprinted’ genes, so that certain maternal genes are silenced, as are complementary paternal genes. Uniparental disomy disorders (UPD) like Angelman (AS) and Prader-Willi syndromes (PWS)47,48 result from the loss of either two maternally (AS) or two paternally (PWS) expressed genes. Since centrosomes in human zygotes form bipolar spindles even after dispermic insemination, the possibility exists that daughter blastomeres after cell division inherit two or more paternal or maternal chromosome sets. While the exact origins for uniparental disomies remain elusive, understanding the molecular consequences of these imprinting errors in humans would be enormously useful in resolving developmental, neurological, and behavioral consequences of these devastating diseases.

MATERNAL CENTROSOME ANOMALIES AND BIRTH DEFECTS

A primary cause of reproductive failure in older women may reflect aging of crucial maternal factors relevant to genetic fidelity and/or assembling a functional zygotic centrosome.49 Aneuploidy rates as high as 52–61% have been reported in preimplantation

cytoplasmic extract and subsequent exposure to rhodamine-conjugated bovine brain tubulin (C). Primed sperm not exposed to CSF extract do not nucleate microtubules after incubation in rhodamine-conjugated bovine brain tubulin (D). Bars ⫽ 1 ␮m. (A)–(C) Reprinted with permission from Schatten.7 (C) and (D) reprinted with permission from Simerly et al.19



HPE_Chapter26.qxp

7/13/2007

5:32 PM

Page 346


Mt

γ –Tub

A

B

Mt

γ –Tub

D

E

DNA

C DNA

F

Figure 26.6 Mitotic spindle formation in dispermic human zygotes. Dispermy in human zygotes results in a disorganized prophase mitotic spindle (A) with four foci of ␥-tubulin ((B) location denoted by asterisks in (A)). This disorganized prophase spindle resolves at metaphase resulting in a bipolar spindle (D) with well-organized chromosomes (F). Four foci of ␥-tubulin are still observed on the spindle ((E) location denoted by asterisks in (D)). Mt, microtubules; ␥-Tub, ␥-tubulin. Arrows denote incorporated sperm axonemes and asterisks denote ␥-tubulin foci. Bar ⫽ 10 ␮m. Reprinted with permission from Simerly et al.19

human embryos,50 the majority caused by chromosomal errors arising during meiosis I. For instance, chromosome non-disjunction in females that results in tragic birth defects occurs during first meiosis, probably because the stringency of metaphaseanaphase checkpoint control is very low in the maturing oocyte.51,52 These findings have produced clinically useful preconception genetic diagnostic tests to predict chromosome errors, such as fluorescent in situ hybridization (FISH) of the first polar body, a product of first meiosis,53 as opposed to the more typical preimplantation genetic diagnosis (PGD) after embryo blastomere biopsy.54,55 With the exception of rodents, all mammalian oocytes reduce and lose their traditional, replicating centrosomes, although the exact timing and nature of this reduction is not well known. Extrapolation from invertebrates56 suggests that the maternal centrosome is reduced before the completion of the 

meiotic divisions.57 In humans, all the mitotic divisions of the oogonia are completed by mid-gestation within female fetuses, i.e. 4–5 months prior to birth. As reproduction cannot start for more than a decade and the potential of eggs from women older than 35 years declines swiftly, the nature of meiotic spindle assembly and stability in oocytes aged over decades is of both fundamental and clinical importance. With the advent of oocyte donation, women beyond menopause can deliver healthy children if the oocyte is obtained from a younger woman.58 For reproductive biologists, questions persist regarding the nature of meiotic spindle assembly without the benefit of a maternal centrosome inside a maturing oocyte.59 As yet, it is unclear what distinguishes an oocyte that is reproductively viable in the third decade of life from another oocyte in the fifth decade that often produces aberrant meiotic spindles and aneuploid embryos incapable of either

HPE_Chapter26.qxp

7/13/2007

5:32 PM

Page 347


implantation or embryogenesis. Furthermore, female meiosis is fundamentally distinct from that of male meiosis for reasons not yet clear. Mammalian female meiosis begins in utero, arrests until puberty, and is then reinitiated once per month in a small pool of immature oocytes until they are depleted and menopause ensues. In contrast, mitosis in the male germ line restarts with the onset of puberty, and continuous waves of male meiosis as well as on-going proliferation of male germ cells occur through the majority of the life of the individual.60 Clearly, understanding these distinctions will be required to fully characterize gametogenesis and the special restrictions on centrosome restorative properties in male and female gametes so vital to ensure reproductive success.

THE SPECIAL PROBLEM OF PARTHENOGENESIS: ROLES OF MICROTUBULE MOTORS AND SPINDLE STRUCTURAL PROTEINS

In the classical sense, the centrosomes direct mitotic bipolar spindle assembly by acting as MTOCs at their poles.1,2 In rodents, the maternally derived centrosomes inherited from oogenesis remain active for directing the motility events crucial to the completion of the fertilization process during meiosis as well as after insemination.7 For other mammals, the uniparental contribution of the centrosome is strictly observed, as evidenced by multiple sperm asters during polyspermic insemination as well as by the lack of cytoplasmic asters after parthenogenetic activation. Most species that follow a paternal method of centrosome inheritance also support parthenogenic activation,7 and development of cleavage stage embryos is observed.61–64 Analyses of meiotic spindles in a variety of mammalian oocytes suggest that there is no maternal centrosome at their poles, although functional bipolar meiotic spindles assemble during meiosis.20,65 Centrioles are absent in mammalian oocytes, but can arise de novo in parthenogenetic embryos after a few cleavage divisions.13,61 Such observations pose a dilemma regarding the restoration of the previously lost maternal centro-

some after parthenogenetic activation.7 After primate and bovine parthenogenetic activation,14,66 a bipolar spindle with aligned chromosomes forms at first mitotic metaphase (Figure 26.7A), but without identifiable centrosomal proteins (i.e. ␥-tubulin or pericentrin) at the poles.7 These spindles accurately segregate the chromosomes (Figure 26.7B) and can complete preimplantation development to the blastocyst stage.62 Exactly how meiotic and parthenogenetic mitotic spindles can assemble and direct cell division without functional MTOCs is not well understood.68 Some evidence suggests that acentrosomal systems assemble bipolar spindles by an ‘inside-out’ mechanism where random microtubules bound to chromatin are sorted and assembled into a bipolar structure using specific molecular motor proteins related to the bim C and nod class of kinesin-related proteins. After sorting these microtubules, the spindle poles are then shaped by a host of other proteins, including the minus-end directed motors dynein and the Kin C kinesins, the plus-end directed motors of the bim C family (i.e. Eg5) (Figure 26.7), and the structural proteins dynactin and NuMA (nuclear mitotic apparatus protein).69 In mouse oocytes, Eg5 (a Kin N plus end-directed kinesin) and HSET (Kin C minus-end directed kinesin) are present in meiotic spindles; perturbing their activity following microinjection of function-blocking antibodies demonstrates that they are required for bipolar spindle assembly.70 Similar crucial roles for minus-end directed kinesin motors in organizing female meiotic spindles have been described for Drosophila mutants deficient in the minus-end kinesin nonclaret disjunctional (ncd).71 Thus, the centrosome may be dispensable for meiotic and first mitotic spindle formation after parthenogenetic activation in mammalian eggs. Rather, microtubule crosslinking and oppositely oriented motor activity, as well as spindle matrix proteins like NuMA and dynactin, are critical components required for assembling these spindles.69 However, the centrosome is required to complete male and female pronuclear apposition and serves as the dominant MTOC and spindle pole organizing entity when present. 

HPE_Chapter26.qxp

7/13/2007

5:32 PM

Page 348


Eg5/Mt/DNA

Eg5/Mt/DNA

A

B

Figure 26.7 Mitotic spindle assembly after parthenogenetic activation of monkey oocytes. (A) Parthenogenetically activated rhesus oocytes at mitosis assemble anastral, bipolar spindles with aligned chromosomes at metaphase. Maternal centrosomes are absent at the spindle poles based on the lack of foci containing conserved centrosomal constituent proteins like ␥-tubulin, pericentrin, or centrin. Oppositely oriented motor proteins like Eg5 (green) and HSET along with spindle pole organizing components like NuMA probably direct bipolar spindle assembly after artificial activation. (B) Telophase stage parthenogenetically activated monkey oocyte demonstrating organized chromosome separation during cell division. All images triple-labeled for Eg5 kinesin (green), microtubules (red), and DNA (blue). Bars ⫽ 10 ␮m. Reprinted with permission from Simerly et al.67

CENTROSOMES DURING NUCLEAR TRANSFER IN NON-HUMAN PRIMATES

Nuclear transfer (NT) in mammals challenges fundamental tenets of developmental biology, including requirements for exactly two parents of opposite sexes during natural reproduction.72 Whereas two haploid genomes unite within the activated egg’s cytoplasm during fertilization, as directed by the introduced sperm centrosome, the maternal spindle–chromosome complex is removed during ‘enucleation’ and a diploid nucleus is inserted. After a period of reprogramming, the NT construct is activated as during parthenogenesis and ultimately transferred to a pseudopregnant recipient for production 

of live young. NT is sporadically successful in a few mammals (mice, domestic species), but its mechanisms still remain largely unexplained. In non-human primates, however, application of NT protocols proven successful in other mammals have identified unexpected problems, including the extranuclear inheritance of the centrosome.67,73 Imaged live, non-human primate NT constructs appear normal, yet no pregnancies resulted from 140 transfers at the 2–8-cell stage into 25 surrogates.67 Microtubule and DNA imaging of interphase NT constructs showed abnormal microtubule patterns, most likely the result of dysfunctional somatic centrosomes, and inappropriate nuclear reconstitution (Figure 26.8A–C) compared with fertilized controls, androgenotes (ICSI fertilization of enucleated oocytes), and bovine clones produced by similar NT techniques (Figure 26.8D–F).66,67,73 Furthermore, somatic centrosomal dysfunction in the non-human primate NT constructs appears correlated to the primate ooplasm, as evident by a focused microtubule astral array when a monkey fibroblast nucleus is transferred into a bovine enucleated cytoplast (Figure 26.8G). At first mitosis, multipolar spindles with misaligned chromosomes assembled after somatic or embryonic donor cell nuclear transfer (Figure 26.9A),67 are distinctly different from NT bipolar mitotic spindles in cows produced by similar methods (compare with Figure 26.8F).66,67 Many resultant NT embryos that underwent cleavage were aneuploid with limited developmental potential, though strict adherence to the timing of enucleation and method of artificial activation suggested improved cloned blastocyst development rates.67 The high rate of incorrectly assembled bipolar mitotic spindles in non-human primate constructs is surprising, given the routine ability of polyspermic and parthenogenetic eggs to assemble bipolar spindles with aligned chromosomes (Figure 26.7). Without the sperm centrosome, unanticipated obstacles prevent proper non-human primate NT construct development, including chromosome imbalances, inaccurate spindle pole numbers, and the dysfunctional acquired somatic cell centrosome.73 Non-human primate NT constructs demand two critical structures missing during cloning – the

HPE_Chapter26.qxp

7/13/2007

5:33 PM

Page 349


Figure 26.8 Abnormal microtubule patterns after nuclear transfer (NT) in monkey (Rh), but not bovine, constructs. (A)–(C) Disarrayed microtubules (Mt) (green) assembled near the transferred somatic cell nucleus (blue) indicate a dysfunctional somatic centrosome following intracytoplasmic nuclear injection (ICNI) into enucleated monkey oocytes and artificial activation. Some constructs demonstrate cortical microtubule patterns (C) similar to parthenogenetically activated oocytes suggesting no functional microtubule organizing activity by the transferred somatic cell centrosome. (D) and (E) Tightly focused microtubule arrays (green) emanating from the transferred nucleus (blue) in activated bovine enucleated cytoplasts following either embryonic (D) or somatic cell (E) nuclear transfer. (F) Normal, anastral, bipolar spindles (green) with aligned chromosomes assemble at metaphase (blue) in activated bovine nuclear transfer constructs. (G) A focused microtubule array (green) from a rhesus fibroblast cell (blue) transferred into a bovine enucleated oocyte suggests that normal somatic centrosomal function at interphase is dependent on the maternal cytoplasmic source more than the donor nuclear source. All images double-labeled for microtubules (green) and DNA (blue). Bars ⫽ 10 ␮m. Reprinted with permission from Simerly et al.67

zygotic centrosome formed from the sperm’s proximal centriole and egg’s cytoplasmic constituents, and the oocyte’s meiotic spindle–chromosome complex. Microtubule minus-end residing proteins like the M-phase kinesin HSET70 and the structural matrix proteins NuMA are present at the spindle poles in unfertilized meiotic and fertilized mitotic oocytes (Figure 26.9B–C). However, these vital spindle components are not detected in NT mitotic spindles after nuclear transfer (Figure 26.9D and E).

Conversely, the oppositely oriented kinesin motor Eg5 binds centromere pairs in meiotic and mitotic spindles, including those on NT-cloned mitotic spindles (Figure 26.9F). Why NuMA and HSET are disrupted from participating in NT mitotic bipolar spindle assembly is not entirely clear, although removal of the meiotic spindle during the initial enucleation step may perhaps remove a large majority of these proteins from the cytoplasmic pool.73 Supporting evidence for this theory is found in experiments 

HPE_Chapter26.qxp

7/13/2007

5:33 PM

Page 350


B

10 µm A

E

F

10 µm

10 µm

10 µm C

10 µm

10 µm

G

D

10 µm

Figure 26.9 Faulty mitotic spindles produce aneuploid embryos after primate nuclear transfer. (A) Defective nuclear transfer (NT) mitotic spindle with misaligned chromosomes. Centrosomal NuMA at meiosis (B) and mitosis (C), but not in mitotic spindles after NT (D). The centrosomal kinesin HSET is also missing after NT (E), but not centromeric Eg5. (F) Bipolar mitotic spindles with aligned chromosomes and centrosomal NuMA after NT into fertilized eggs (G). Blue, DNA; red, ␤-tubulin; green, NuMA in (B), (C), (D), and (G); green, HSET in (E); green Eg5 in (F). Reprinted with permission from Simerly et al.67

combining fertilization with nuclear transfer, where the resulting tetraploid mitotic spindles assemble with properly aligned chromosomes on bipolar spindles and NuMA is detected at each spindle pole (Figure 26.9G). Primate NT is challenged by molecular requirements in assembling the first mitotic spindle that appear to be stricter than in other mammals in which NT succeeds. In cattle, the somatic centrosome transferred during NT organizes a large, wellformed microtubule aster within the recipient’s cytoplasm,67,74 while mice rely on the maternal centrosome within the ooplasm.7 Successful reproductive cloning is achieved routinely only in those 

well-studied species that provide vast numbers of both oocytes and surrogates, systems often capable of multiple deliveries and amenable to transfers of supernumerary embryos.

FUTURE RESEARCH CHALLENGES FOR THE CENTROSOME BIOLOGISTS

The specialized field of NT, although controversial, is aimed at determining the scientific feasibility of ‘therapeutic cloning’.75 As shown in relevant animal studies, NT cloning fails owing to a number of defects, including nuclear reprogramming, cell cycle

HPE_Chapter26.qxp

7/13/2007

5:33 PM

Page 351


asynchrony, gene mis-expression during development, genomic imprinting errors, placental dysfunction, and technical damage, to name but a few.76 Additionally, epigenetic obstacles like cytoplasmic incompatibilities in mitotic spindle assembly are noteworthy hurdles. Primate NT constructs appear to be normal when imaged live and non-invasively, but they demonstrate limited reproductive potential due to severe aneuploidy. Understanding these abnormalities will facilitate the production of genetically identical non-human primates for characterizing human diseases in a relevant preclinical model. Such investigations will be invaluable for assessing the promise of innovative embryonic stem cell therapies for clarity on the root causes of diseases, the production of novel disease cures through new drug discovery, and exploring early human developmental origins, including the onset of fetal disorders.75,77 Current approaches of NT cloning in non-human primates cannot meet these goals due to chromosome imbalances resulting from spindle defects. A key element to resolving such defects will be in fully characterizing somatic cell centrosome inheritance and reconstitution following NT. Reproduction, typically error prone, is clinically successful in fewer than a quarter of natural conceptions.78 Knowledge of the cellular and molecular events of fertilization is, therefore, critical for the reproductive and developmental biologists. For the centrosomal biologists, progress in understanding the molecular and cellular underpinnings of the cell’s MTOC remains an exciting research area. Already, fundamental discoveries have been made in centrosomal inheritance, composition, assembly, duplication, and separation in a variety of systems. Future discoveries in understanding fertilization and early development hinge on a more complete knowledge underlying this crucial cellular structure, and will greatly facilitate modern molecular medicine in addressing a multitude of clinical problems in infertility, contraception, and birth defects. ACKNOWLEDGMENTS

We thank Dr Gerald Schatten for his deep inspiration, unbridled enthusiasm, and incredible support

of cell and developmental biologists. We also thank all of our wonderful scientific colleagues, both past and present, for stimulating scientific collaborations in centrosomal biology, including our current members: Ahmi Ben-Yehudah, Carlos Castro, Kevin Grund, Laura Hewitson, Ethan Jacoby, Chih Cheng Lin, Dave McFarland, Jodi Mich-Basso, Cindy Oberley, Kyle Orwig, Hina Qidwai, Carrie Redinger, Mario Rodriguez, Meena Sukhwani, and Jamie Tomko. We thank Dr Diane Carlisle for editorial assistance with the manuscript, and the continued support of the National Institute of Health (nonhuman primates, rodents, and domestic species). All protocols performed in this review were approved by the appropriate University’s Research Animal Review and Human Subjects Institutional Review committees. REFERENCES 1. Compton DA. Spindle assembly in animal cells. Annu Rev Biochem 2000; 69: 95–114. 2. Hyman AA. Centrosomes: Sic transit gloria centri. Curr Biol 2000; 10(7): R276–8. 3. Doxsey S, McCollum D, Theurkauf W. Centrosomes in cellular regulation. Annu Rev Cell Dev Biol 2005; 21: 411–34. 4. Tsou MF, Stearns T. Controlling centrosome number: licenses and blocks. Curr Opin Cell Biol 2006; 18(1): 74–8. 5. Andersen SS. Molecular characteristics of the centrosome. Int Rev Cytol 1999; 187: 51–109. 6. Boveri T. Zellen-studien: Ueber die natur der centrosomen. Jena, Germany: Fischer; 1901. 7. Schatten G. The centrosome and its mode of inheritance: the reduction of the centrosome during gametogenesis and its restoration during fertilization. Dev Biol 1994; 165(2): 299–335. 8. Steptoe PC, Edwards RG. Birth after the reimplantation of a human embryo. Lancet 1978; 2(8085): 366. 9. Menezo YJ, Veiga A, Pouly JL. Assisted reproductive technology (ART) in humans: facts and uncertainties. Theriogenology 2000; 53(2): 599–610. 10. Schatten G, Hewitson L, Simerly C, Sutovsky P, Huszar G. Cell and molecular biological challenges of ICSI: ART before science? J Law Med Ethics 1998; 26(1): 29–37, 3. 11. Palermo G, Joris H, Devroey P, Van Steirteghem AC. Pregnancies after intracytoplasmic injection of single spermatozoon into an oocyte. Lancet 1992; 340(8810): 17–8. 12. Manandhar G, Simerly C, Schatten G. Centrosome reduction during mammalian spermiogenesis. Curr Top Dev Biol 2000; 49: 343–63. 13. Szollosi D, Calarco P, Donahue RP. Absence of centrioles in the first and second meiotic spindles of mouse oocytes. J Cell Sci 1972; 11(2): 521–41. 14. Simerly C, Wu GJ, Zoran S et al. The paternal inheritance of the centrosome, the cell’s microtubule-organizing center, in humans, and the implications for infertility. Nat Med 1995; 1(1): 47–52.



HPE_Chapter26.qxp

7/13/2007

5:33 PM

Page 352


15. Sathananthan AH, Kola I, Osborne J et al. Centrioles in the beginning of human development. Proc Natl Acad Sci U S A 1991; 88(11): 4806–10. 16. Palermo GD, Colombero LT, Rosenwaks Z. The human sperm centrosome is responsible for normal syngamy and early embryonic development. Rev Reprod 1997; 2(1): 19–27. 17. Balczon R, Bao L, Zimmer WE et al. Dissociation of centrosome replication events from cycles of DNA synthesis and mitotic division in hydroxyurea-arrested Chinese hamster ovary cells. J Cell Biol 1995; 130(1): 105–15. 18. Hinchcliffe EH, Sluder G. Two for two: Cdk2 and its role in centrosome doubling. Oncogene 2002; 21(40): 6154–60. 19. Simerly C, Zoran SS, Payne C et al. Biparental inheritance of gammatubulin during human fertilization: molecular reconstitution of functional zygotic centrosomes in inseminated human oocytes and in cell-free extracts nucleated by human sperm. Mol Biol Cell 1999; 10(9): 2955–69. 20. Sathananthan AH. Mitosis in the human embryo: the vital role of the sperm centrosome (centriole). Histol Histopathol 1997; 12(3): 827–56. 21. Rawe VY, Terada Y, Nakamura S et al. A pathology of the sperm centriole responsible for defective sperm aster formation, syngamy and cleavage. Hum Reprod 2002; 17(9): 2344–9. 22. Terada Y. Human sperm centrosomal function during fertilization, a novel assessment for male sterility. Hum Cell 2004; 17(4): 181–6. 23. Hewitson L, Haavisto A, Simerly C, Jones J, Schatten G. Microtubule organization and chromatin configurations in hamster oocytes during fertilization and parthenogenetic activation, and after insemination with human sperm. Biol Reprod 1997; 57(5): 967–75. 24. Sutovsky P, Terada Y, Schatten G. Ubiquitin-based sperm assay for the diagnosis of male factor infertility. Hum Reprod 2001; 16(2): 250–8. 25. Van Blerkom J. Sperm centrosome dysfunction: a possible new class of male factor infertility in the human. Mol Hum Reprod 1996; 2(5): 349–54. 26. Silber S, Escudero T, Lenahan K et al. Chromosomal abnormalities in embryos derived from testicular sperm extraction. Fertil Steril 2003; 79(1): 30–8. 27. Navara CS, First NL, Schatten G. Phenotypic variations among paternal centrosomes expressed within the zygote as disparate microtubule lengths and sperm aster organization: correlations between centrosome activity and developmental success. Proc Natl Acad Sci USA 1996; 93(11): 5384–8. 28. Comizzoli P, Wildt DE, Pukazhenthi BS. Poor centrosomal function of cat testicular spermatozoa impairs embryo development in vitro after intracytoplasmic sperm injection. Biol Reprod 2006; 75(2): 252–60. 29. Sadowy S, Tomkin G, Munne S, Ferrara-Congedo T, Cohen J. Impaired development of zygotes with uneven pronuclear size. Zygote 1998; 6(2): 137–41. 30. Doxsey SJ, Stein P, Evans L, Calarco PD, Kirschner M. Pericentrin, a highly conserved centrosome protein involved in microtubule organization. Cell 1994; 76(4): 639–50. 31. Felix MA, Antony C, Wright M, Maro B. Centrosome assembly in vitro: role of gamma-tubulin recruitment in Xenopus sperm aster formation. J Cell Biol 1994; 124(1–2): 19–31. 32. Stearns T, Kirschner M. In vitro reconstitution of centrosome assembly and function: the central role of gamma-tubulin. Cell 1994; 76(4): 623–37. 33. Salisbury JL, Suino KM, Busby R, Springett M. Centrin-2 is required for centriole duplication in mammalian cells. Curr Biol 2002; 12(15): 1287–92. 34. Fouquet JP, Kann ML, Combeau C, Melki R. Gamma-tubulin during the differentiation of spermatozoa in various mammals and man. Mol Hum Reprod 1998; 4(12): 1122–9.



35. Beisson J, Wright M. Basal body/centriole assembly and continuity. Curr Opin Cell Biol 2003; 15(1): 96–104. 36. Nakamura S, Terada Y, Horiuchi T et al. Analysis of the human sperm centrosomal function and the oocyte activation ability in a case of globozoospermia, by ICSI into bovine oocytes. Hum Reprod 2002; 17(11): 2930–4. 37. Moomjy M, Colombero LT, Veeck LL, Rosenwaks Z, Palermo GD. Sperm integrity is critical for normal mitotic division and early embryonic development. Mol Hum Reprod 1999; 5(9): 836–44. 38. Schatten G, Simerly C, Schatten H. Maternal inheritance of centrosomes in mammals? Studies on parthenogenesis and polyspermy in mice. Proc Natl Acad Sci USA 1991; 88(15): 6785–9. 39. Crozet N, Dahirel M, Chesne P. Centrosome inheritance in sheep zygotes: centrioles are contributed by the sperm. Microsc Res Tech 2000; 49(5): 445–50. 40. Kim NH, Simerly C, Funahashi H, Schatten G, Day BN. Microtubule organization in porcine oocytes during fertilization and parthenogenesis. Biol Reprod 1996; 54(6): 1397–404. 41. Wu GJ, Simerly C, Zoran SS, Funte LR, Schatten G. Microtubule and chromatin dynamics during fertilization and early development in rhesus monkeys, and regulation by intracellular calcium ions. Biol Reprod 1996; 55(2): 260–70. 42. Kola I, Trounson A. Dispermic Human Fertilization: Violation of Expected Cell Behaviour. San Diego: Academic Press Inc., 1989. 43. Plachot M, Crozet N. Fertilization abnormalities in human in-vitro fertilization. Hum Reprod 1992; 7(Suppl 1): 89–94. 44. Phelan MC, Curtis Rogers R, Michaelis RC, Moore CL, Blackburn W. Prenatal diagnosis of mosaicism for triploidy and trisomy 13. Prenat Diagn 2001; 21(6): 457–60. 45. Golubovsky MD. Postzygotic diploidization of triploids as a source of unusual cases of mosaicism, chimerism and twinning. Hum Reprod 2003; 18(2): 236–42. 46. Egozcue S, Blanco J, Vidal F, Egozcue J. Diploid sperm and the origin of triploidy. Hum Reprod 2002; 17(1): 5–7. 47. Bestor TH. Imprinting errors and developmental asymmetry. Philos Trans R Soc Lond B Biol Sci 2003; 358(1436): 1411–5. 48. Verona RI, Mann MR, Bartolomei MS. Genomic imprinting: intricacies of epigenetic regulation in clusters. Annu Rev Cell Dev Biol 2003; 19: 237–59. 49. Battaglia DE, Goodwin P, Klein NA, Soules MR. Influence of maternal age on meiotic spindle assembly in oocytes from naturally cycling women. Hum Reprod 1996; 11(10): 2217–22. 50. Munne S, Cohen J. Chromosome abnormalities in human embryos. Hum Reprod Update 1998; 4(6): 842–55. 51. Hassold T, Hunt P. To err (meiotically) is human: the genesis of human aneuploidy. Nat Rev Genet 2001; 2(4): 280–91. 52. Hunt PA, Hassold TJ. Sex matters in meiosis. Science 2002; 296(5576): 2181–3. 53. Verlinsky Y, Cieslak J, Kuliev A. Preimplantation FISH diagnosis of aneuploidies. Meth Mol Biol 2002; 204: 259–73. 54. Handyside AH, Scriven PN, Ogilvie CM. The future of preimplantation genetic diagnosis. Hum Reprod 1998; 13(Suppl 4): 249–55. 55. Werlin L, Rodi I, DeCherney A et al. Preimplantation genetic diagnosis as both a therapeutic and diagnostic tool in assisted reproductive technology. Fertil Steril 2003; 80(2): 467–8. 56. Shirato Y, Tamura M, Yoneda M, Nemoto S. Centrosome destined to decay in starfish oocytes. Development 2006; 133(2): 343–50. 57. Briggs D, Miller D, Gosden R. Molecular Biology of Female Gametogenesis. New York: Parthenon Publishing, 1999. 58. Paulson RJ, Boostanfar R, Saadat P et al. Pregnancy in the sixth decade of life: obstetric outcomes in women of advanced reproductive age. JAMA 2002; 288(18): 2320–3.

HPE_Chapter26.qxp

7/13/2007

5:33 PM

Page 353


59. Kovacic B, Vlaisavljevic V. Configuration of maternal and paternal chromatin and pertaining microtubules in human oocytes failing to fertilize after intracytoplasmic sperm injection. Mol Reprod Dev 2000; 55(2): 197–204. 60. Brinster RL, Nagano M. Spermatogonial stem cell transplantation, cryopreservation and culture. Semin Cell Dev Biol 1998; 9(4): 401–9. 61. Szollosi D, Ozil JP. De novo formation of centrioles in parthenogenetically activated, diploidized rabbit embryos. Biol Cell 1991; 72(1–2): 61–6. 62. Cibelli JB, Grant KA, Chapman KB et al. Parthenogenetic stem cells in nonhuman primates. Science 2002; 295(5556): 819. 63. Vrana KE, Hipp JD, Goss AM et al. Nonhuman primate parthenogenetic stem cells. Proc Natl Acad Sci USA 2003; 100(Suppl 1): 11911–6. 64. Prather RS. Basic mechanisms of fertilization and parthenogenesis in pigs. Reprod Suppl 2001; 58: 105–12. 65. Lee J, Miyano T, Moor RM. Spindle formation and dynamics of gamma-tubulin and nuclear mitotic apparatus protein distribution during meiosis in pig and mouse oocytes. Biol Reprod 2000; 62(5): 1184–92. 66. Navara CS, First NL, Schatten G. Microtubule organization in the cow during fertilization, polyspermy, parthenogenesis, and nuclear transfer: the role of the sperm aster. Dev Biol 1994; 162(1): 29–40. 67. Simerly C, Navara C, Hyun SH et al. Embryogenesis and blastocyst development after somatic cell nuclear transfer in nonhuman primates: overcoming defects caused by meiotic spindle extraction. Dev Biol 2004; 276(2): 237–52.

68. Megraw TL, Kao LR, Kaufman TC. Zygotic development without functional mitotic centrosomes. Curr Biol 2001; 11(2): 116–20. 69. Nedelec F, Surrey T, Karsenti E. Self-organisation and forces in the microtubule cytoskeleton. Curr Opin Cell Biol 2003; 15(1): 118–24. 70. Mountain V, Simerly C, Howard L et al. The kinesin-related protein, HSET, opposes the activity of Eg5 and cross-links microtubules in the mammalian mitotic spindle. J Cell Biol 1999; 147(2): 351–66. 71. Endow SA. Microtubule motors in spindle and chromosome motility. Eur J Biochem 1999; 262(1): 12–8. 72. Alberio R, Campbell KH. Epigenetics and nuclear transfer. Lancet 2003; 361(9365): 1239–40. 73. Simerly C, Dominko T, Navara C et al. Molecular correlates of primate nuclear transfer failures. Science 2003; 300(5617): 297. 74. Shin MR, Park SW, Shim H, Kim NH. Nuclear and microtubule reorganization in nuclear-transferred bovine embryos. Mol Reprod Dev 2002; 62(1): 74–82. 75. McLaren A. Ethical and social considerations of stem cell research. Nature 2001; 414(6859): 129–31. 76. Humpherys D, Eggan K, Akutsu H et al. Abnormal gene expression in cloned mice derived from embryonic stem cell and cumulus cell nuclei. Proc Natl Acad Sci USA 2002; 99(20): 12889–94. 77. Munne S, Velilla E, Colls P et al. Self-correction of chromosomally abnormal embryos in culture and implications for stem cell production. Fertil Steril 2005; 84(5): 1328–34. 78. Edwards RG. IVF and the history of stem cells. Nature 2001; 413(6854): 349–51.



HPE_Chapter26.qxp

7/13/2007

5:33 PM

Page 354

HPE_Index_new.qxp

7/16/2007

8:51 PM

Page 355

Index N.B. Page numbers in italic denote figure or table legends acidified Tyrode’s solution 137, 139–40 microneedle application 137, 138–9 adaptive immune system 155, 156 age see maternal age alanine 186 Alzheimer’s disease 260 amino acids 185, 193 endogenous pool factors affecting 195 in older females 197 essential amino acids 195–6 sulfur amino acids 196–7 amino acid metabolism 185–6, 193–8 ammonia toxicity 197 analysis 181 glutamine degradation 197–8 ammonia elimination 197 toxicity 197 anaphase promoting complex (APC) 267, 268 androgenetic embryos 236 aneuploidy 309 chromosome-specific 205 and embryonic competence 215 FISH, discovered by 205 and fragmentation 57–8 frequency in cleavage stage 209 after nuclear transfer 350 polyspermy 343, 345 reduction due to PGD 288 specific chromosome rates 214 testing in IVF 206 Angelman syndrome 240, 326, 345 angiogenic factors in follicular fluid 311–13 anti-Mullerian hormone (AMH) 314–15 antioxidants 191–2 antrum formation 302 apoptosis causing fragmentation 66–8 SNP-induced 172 in spermatogenesis 327–8 marker proteins 328

ART see assisted reproductive technologies (ART) ascorbic acid 198 assisted hatching (AH) technique 16 in cryopreservation 130 embryo selection 138–9 general considerations 135–8 meta–analyses 136–7 monozygotic twinning 141 patient selection 138 usefulness 136 zona opening acidified Tyrode’s solution 139–40 laser 140 assisted reproductive technologies (ART) current trends 1 HLA-G in see assisted reproductive technologies (ART) in large animals 238–9 success rate 228–9 asters see cytasters; sperm-derived centrosomes (sperm asters) asymmetric blastomeres blastomere size during cleavage 96 and chromosomal abnormalities 212 Barker hypothesis 164 basic fibroblast growth factor (bFGF) 265 Bax proteins 67–8, 73 Bcl-2 proteins 67–8, 73 Beckwith-Wiedemann syndrome 240 binucleated MNEs 47–8 implantation/pregnancy rates 49 bioinformatics 247–8 biomarker discovery with metabolomics 250–1 biopsy see embryo biopsy; polar body biopsy biospectroscopy-based metabolomics 249, 253 bipolarization 6 blastocysts 2, 79–87, 156 abnormal 12, 13 conversion 112, 114 culture 9, 79–81 development 108–9, 109



HPE_Index_new.qxp

7/16/2007

8:51 PM

Page 356

INDEX

blastocysts (Contd.) with IVF vs ICSI 329 and sperm quality 329 formation and chromosome abnormalities 215–17 embryo survival to blastocyst stage 216 fragments 13 hatching 2, 10, 12, 13 problems 16 implantation potential 84 inner cell mass (ICM) assessment 109 morphology 108–9 analysis 81 effect of score on pregnancy outcome 84 scoring system 81, 82–4 normal 10 E-cadherin distribution 65 after ICSI 13 in vivo maturation 102 physiology analysis 81, 84, 85 respiration 281 blastocyst transfer advantages 79, 79, 84, 117 disadvantages 117 single 118–19 blastomeres asymmetry blastomere size during cleavage 96 and chromosomal abnormalities 212 lysis in cryopreservation 129–30 morphometric analysis 93–7 sampling, preimplantation 205 size asymmetric blastomeres 96, 212 cell size cut-off limits 96 and fragmentation 93–6, 95, 97 total cytoplasmic volume 95, 95–6 multinucleation 96–7 variations in size/cell generations 94 blastulation 52 impact of fragmentation 59, 59 patterns 60 bone morphogenic protein (Bmp) genes 264, 265 BORIS 238 BRCA1 gene 268, 270, 293 calcium-calmodulin in NO production 169–70 calcium oscillations 21 capillary electrophoresis (CE) 249 carbohydrate metabolism 181–5 glucose 184 

lactate 185 pyruvate 183–4 Carnoy’s fixation method 221 CCCTC-binding factor protein (CTCF) 237–8 cell cycle regulation 170–1, 172 checkpoints 171 in oocytes 268 cell type/number analyzed in PGD 218–21 centrin 343 centrioles absence in oocytes 347 see also sperm centriole structure centrosomes definition 337 embryo development, effect on 337–53 female absence in meiotic spindles 347 anomalies in 345–7 future research 350–1 inheritance 337, 338, 339 male 6, 24 centrosome function assays 342–3 dysfunction and infertility 340–2 during fertilization 339–40 during nuclear transfer 348–50 phenotypic expression in bulls 341–2 reconstitution during fertilization 342 zygotic centrosome formation 342 centrosome function assays 342–3 centrosome microinjection therapy 343 CGH see comparative genome hybridization (CGH) checkpoints 171, 173–4 during meiosis 266–8 in oocytes 267 cholesterol synthesis 193, 194 chromatin, nucleolar–associated 6 alignment 6, 6, 24 rotation and pronuclear scoring 32 chromosomal abnormalities in ART embryos causes 210–11 prevalence of 288 and asymmetric blastomeres 212 and blastocyst formation 215–17 and cytoplasmic irregularities 213 and dysmorphism 212–13 and elongated embryo shape 213 of embryos 14 survival to blastocyst stage 216 and fragmentation 212

HPE_Index_new.qxp

7/16/2007

8:51 PM

Page 357

INDEX

frequency in cleavage stage 209 and giant oocytes 212–13 and implantation failure 208 and maternal age 245 morphological selection against 212–13 and multinucleation 212 in oocytes 302 of polar bodies 201–7 detection rate 203 embryos, relationship with 204–6 frequency/types 202–4 postmeiotic 209, 215 pronuclear morphology 213 self-correction (balancing) 204 and structural defects 58 study requirements 209 in unselected blastocyst studies 215–16, 216 in XO female mice 203 see also specific conditions chromosomal status of embryos 209–33 of MNEs 44–8 and multinucleation severity 47–8 of non-MNEs 46–7 and pronuclear morphology 38–9 cleavage rate and Qa-2/HLA-G levels 161 and sHLA-G levels 152 vs fragmentation 214–15 stage 156 and chromosomal abnormalities 213–15 in morphometric analysis 93 patterns haploidy 217 monosomies 217 mosaicism 216 polyploidy 217 trisomies 216–17 timing early cleavage 103–4 assessment 104, 110, 113 and embryo competence 98 cloning experiments in mice 239 nuclear transfer 350–1 therapeutic 350 COH see also controlled ovarian hyperstimulation (COH) cohesins 267–8

compaction 2, 179 morphology analysis 107 regional (partial) and fragmentation 60 comparative genome hybridization (CGH) 209–10, 289–92, 290 difficulties of application 291–2 microarray CGH 292 competence related determinants 313–15 cortisol:cortisone ratio 313–14 protein growth factors 314–15 computational biology 247–8 computer-assisted embryo assessment 90 controlled ovarian hyperstimulation (COH) 302 Doppler blood flow assessment 305–9 ‘good beginners’ vs ‘bad beginners’ 307–8 repeated cycles and fertility potential 319 corona radiata 2 corrected embryo score 109, 110–12, 115 cortical granules 2 and cryopreservation–induced damage 126 postmature (aging) oocytes 5 cortisol:cortisone ratio 313–14 cryopreservation 123–34 8-cell embryos 129 cleavage-stage embryos 129 cooling 124, 131 embryo morphology, effects on 123 and embryo transfer timing 117–18 fragments/lysed cells, removal of 140–1 plunging/vitrification 124 pre-equilibration 124, 131 seeding 124, 131 sensitivity of in vitro produced cells 125 slow cooling 124, 130, 131 thawing/rehydration 124, 131–2 theory 124 cryopreservation–induced damage causes 130–2 cellular metabolism 130 consequences 123–4 cortical granules in 126 and embryo origin 124–5 embryo quality 129–30 embryo ultrastructure 127–9 intra/interspecies differences 125–6 and intracellular lipids 125 meiotic spindle in 126–7 types of damage 124–7 cryoprotectants 124 impermeable 131 

HPE_Index_new.qxp

7/16/2007

8:51 PM

Page 358

INDEX

cryoprotectants (Contd.) permeable 131 toxicity 131 culture conditions and multinucleation 42–3, 43 culture, extended see blastocyst transfer culture media amino acids in 197–8 ammonium ions in 197 development 181, 191 embryo ‘plasticity’ 199 sequential media 199 culture system and embryo metabolism 191–2 culture techniques 199 cumulus cell assessment 297 cyclic guanosine 3´,5´–monophosphate (cGMP) 173 cyclin-CDK complexes 171 cyclin-dependent kinases (CDKs) 171 cyclins 171, 173–4, 268 cytasters 339 cytokinesis and fragmentation 71–3 and multinucleation 95–6, 97 cytoplasm cytoplasmic halo 32–3 cytoplasmic rotation 24 distribution of types 279–80 irregularities 302–3 and chromosomal abnormalities 213 in spermatozoa 326 oocyte 302–3 cytoplasmic microtubule asters 339 cytosine methylation 237 cytoskeleton anomalies in spermatozoa 326 and fragmentation 69–73 cytostatic factor (CSF) 265 cytotoxic T lymphocytes 158 Dazla gene 264 development see embryo development; oocyte development; preimplantation development diandric triploidy 345 dietary status of patient 79 differentiation and effect of fragmentation 59 digital image analysis multilevel system 90–1, 91, 92, 95 single image system 90 dimethylsulfoxide (DMSO) 124, 125 and cortical granules 126 dispermy 345, 346 

dispermic embryos 6–7, 11, 12 ‘dispermy hypothesis’ 343, 345 DNA damage 268, 268 mitochondrial 276 organization in inseminated oocytes 339 in polyspermy/parthenogenesis 340 repair 326–7 synthesis, post-fertilization 23 see also sperm DNA DNA fragmentation index (DFI) 330–1 DNA precursors 198 Doppler blood flow assessment for abnormalities in natural menstrual cycle 309 in controlled ovarian hyperstimulation 305–9 Doppler ultrasonography 305–9 cost vs benefits 310–11 follicle images 306 higher resolution 308 in ‘poor responders’ 309 and treatment options 306 double stranded DNA breaks (DSBs) 268 dysmorphism asymmetric blastomeres 212 and chromosomal abnormalities 212–13 cytoplasmic irregularities 213 elongated embryo shape 213 fragmentation 212 and giant oocytes 212–13 multinucleation 212 E-cadherin distribution 64, 65 early embryonic wastage 245 electrophoretic sperm isolation 332 elongated embryo shape 213 embryos 1 abnormalities, early 9, 9 arrested 8 degenerating 9, 12 assessment 8–10, 12, 14 grading for embryo transfer in laboratory 8, 8 preimplantation development 8 without blastomere biopsy 295–7 chimeric aggregates 63, 64 chromosome abnormalities in 14 chromosome status 209–33 computer-assisted assessment 90 dispermic 6–7, 11, 12 early stages 2

HPE_Index_new.qxp

7/16/2007

8:51 PM

Page 359

INDEX

fragmented 10, 11, 12 morphology preimplantation development 58–60 and structural defects 58 and viability 58 morphology analysis 103–8 2-cell embryos 103–4 4-cell embryos 104, 106 8-cell embryos 106–7 cell number 106–7 compaction 107 fragmentation 107 symmetry 107 cumulative scoring day 3 grading 107–8 multiday assessments 109, 115–16 day 3 grading 107–8 viability assessment 287–8 zygote stage 103 morphometric analysis 89–99 multinucleated 11 normal 9 1-cell to morula 10 2-cell 10 cleavage stage 9 day 2–8 8 proteomic analysis 295 quality 107–8, 116 and cryopreservation–induced damage 129–30 and single embryo transfer 118 ‘top quality’ 107–8, 116 and single embryo transfer 118 ultrastructure and cryopreservation-induced damage 127–9 viability see viability embryo biopsy 141–2 avoidance 297 comparison with cryopreservation 220 technique in PGD 221 vs polar body biopsy 218–20 zona opening techniques 137 see also preimplantation genetic diagnosis embryo development atypical 51, 52 after cell loss 220 developmental rate, normal timeline of 101–3 early, kinetics of 97–8 fragmentation, impact of 66 immunological aspects 155–68 manipulation of 135–44

history 135 see also assisted hatching (AH) technique; embryo biopsy normal 51, 52 sequential testing 205–6 slow rate, advantages of 162–3 and sperm quality 329 timeline of events 102, 103 embryo development rate (EDR) 104 embryo selection and chromosomal abnormalities of oocytes 302 criteria 275, 282 developmental selection, impact of 217 ideal methods 101 importance of 245 by metabolomic profiling 245–61 metabolomics-based 252 morphology analysis 288 non-invasive methods 245 sHLA-G levels for 151, 152 embryo transfer day 2 vs day 3 117 day 3 101 number of embryos 118–19 numbers, proscription of 305 optimum timing 116–18 sHLA-G levels for selection 151, 152 single 118 technique 224–5 see also multiple embryo transfer embryonic cleavage evaluation 26 and pronuclear scoring 33 embryonic wastage 268 empty follicle syndrome 315–16 endothelial NOS (eNOS) 169–70 expression in embryos 174–5 immunocytochemistry 174 epigenetics 235–44 and ART 235 problems, evidence for 238–41 animal experiments 238–40 human population studies 240–1 and ART offspring 240–1 basic concepts 235–7 consequences/future prospects 241–3 definitions 235–6 male germ cell in 326 mechanistic concepts 237–8 origin 236 

HPE_Index_new.qxp

7/16/2007

8:51 PM

Page 360

INDEX

estrogen and embryo NO production 172 extravillous cytotrophoblasts (EVTs) 145, 147, 148 factor in germline alpha (FIGLA) 264 fertility potential 319 rate, and maternal age 316 screens, potential 342 see also infertility fertilization 2 assessment 5–7 and centrosome inheritance 337, 338 early events 5–6 failed sperm incorporation 7, 8 failure see infertility rates, and oocyte size 92–3 silent 7 sperm-derived centrosomes 339–40 start 5 fertilized ovum abnormally fertilized digyny in ICSI 7 dispermy in IVF 6–7 normally fertilized after 3 hours 6 after 12–14 hours 6 FertiMorph 90, 94 ‘first in, first out’ theory 317 FISH 201 fixation loss of micronuclei 223 method in PGD 221–2 fluorescence in situ hybridization (FISH) 201 with CGH 291, 292 in chromosome abnormality detection 209 errors minimization 222–3 technical sources 223 rescue 223–4 in multinucleation studies 41 probes in PGD 222 hybridization, unsuitable 223 sequential testing 209 signal overlaps 223 stretched signals/double chromosomes 223 fluorescent mitochondrial measurement 282–3 folic acid in vitro 198 follicles aging, biological and chronological 303 antral (tertiary) (Graafian) 302 

and maternal age 317 chemistry/physiology studies 304–5 and maternal age 317–18 non-invasive assessment 301 clinical parameters 304 defect detectability 303–4 pre-antral 301 follicle stimulating hormone (FSH) 210, 301, 302 follicular biochemistry 311–13 follicular biology and maternal age/lifestyle 316–19 follicular development, clinical parameters of 301–5 follicular fluid angiogenic factors in 311–13 anti-Mullerian hormone in 314–15 biochemical composition 304 competence determinants in 313–15 components 313 cortisol:cortisone ratio in 313–14 inhibins in 314–15 follicular markers, search for 301–23 folliculogenesis 301 gene expression during 263–5 Fourier Transform infrared (FTIR) spectroscopy 249, 252 fragmentation 51–77, 52, 72–3 ‘aberrant cytokinesis’ model 71–3 and aneuploidy 57–8 animal models, relevance of 73 blastomeres, impact on fragmenting blastomere 60 sister blastomeres 60–2 blastulation, impact on 59, 59, 60 causes apoptosis in human embryos 67–8 in non-human eggs/embryos 66–7 metabolic/genetic defects 68–9 cell cycle phases 69–71 and cell division 53, 56, 56, 57 and chromosomal abnormalities 212 classification 52–4, 56 and cytoskeleton 69–73 ‘damaged chromosome’ model 72 and differentiation 59 events 51–2 initial 52 extensive 52 fragment disappearance 52 fragment removal 61, 63 future studies 73–4

HPE_Index_new.qxp

7/16/2007

8:51 PM

Page 361

INDEX

impact on preimplantation development 58 and implantation/pregnancy rates 55, 55–7, 57 in IVF 51 mechanisms 69–73 in meiosis II 71, 72 microtubule status 69–70, 70, 72–3 in mitosis I 71 morphology analysis 106, 107 morphometric analysis 93–6 total cytoplasmic volume 95–6 and mosaicism 68 and neonatal outcome 57–8 patterns 53, 53–5, 56 and cell division 56 frequency/fragmentation degree 54, 55 impact on blastula/blastocyte formation 60 prevalence 51 and regional (partial) compaction 60 spindles in 73 and telomere length 69 ‘toxicity’ of fragments 52, 61 and viability 51, 55, 57, 61 freezing, blastocysts after 84 frozen-thawed embryos 61 fructose metabolism 192 functional genomics 251 gamete selection by metabolomic profiling 245–61 gas chromatography (GC) 248, 252 Gdf-9 gene 264–5 gene expression and embryo viability 292–5 embryonic 22, 268 control, shift from maternal to embryonic 170, 170 downstream effects 247 during folliculogenesis 263–5 oocytes vs embryos 266 during preimplantation development 268–70 gene expression analysis 263–73, 293–5, 294 as diagnostic tool 270 genome activation 179 genomic imprinting 235, 236 maternal 236–7 paternal 236 in spermatozoa 326–7 genomics, functional 251 germinal vesicle breakdown (GVBD) 1 signaling after 266 giant oocytes 93 and chromosomal abnormalities 212–13

glucose, excess 192–3, 193 glucose metabolism 180, 181–3, 184, 192, 192–3 catabolic pathways 181–3, 182 optimal glucose level 192 species variation 192 glutamine degradation 197 glutathione 192 glycine 193–4 glycolysis 181–2, 182 gonadotropin releasing hormone agonist (GnRH-a) 210, 211 graduated embryo score (GES) 110–14, 115, 115 granulocyte macrophage colony stimulating factor GMCSF 198 granulosa cell layer development 301–2 zona proteins 15 growth differentiation factor (Gdf-9) gene 264–5 growth factors 198–9 and preimplantation development 199 GVBD see germinal vesicle breakdown (GVBD) gynogenetic embryos 236, 239 H19 genes 237, 239–40 half-embryo development 62 haploidy 217 HELLP syndrome 150 hierarchical cluster analysis 294, 296 high performance liquid chromatography (HPLC) 248–9, 252 highly DNA stainable cells (HDSs) 330, 331 histone protein methylation 237 HLA-G see human leukocyte antigen G (HLA-G) Hoffman modulation contrast optics 25 homocysteine 196 host-pathogen interactions 251 human chorionic gonadotropin (hCG) 186 in cortisol:cortisone ratio 313 in empty follicle syndrome 316 human leukocyte antigens 145, 146 human leukocyte antigen G (HLA-G) 146 and ART 151–2 assessment in human embryos 145–54 characterization 146–7 cytotrophoblast protection 150 effect of presence/absence 162–3 function 145 and implantation 147–51 inhibition of NK cells 146, 147, 148, 150–1 localization of expression 145 

HPE_Index_new.qxp

7/16/2007

8:51 PM

Page 362

INDEX

human leukocyte antigen G (HLA-G) (Contd.) membrane bound, in pregnancy 147 and PED gene 162–3 polymorphism 147 on preimplantation embryos 160–1 as signaling molecule 162 tissue distribution 147 see also soluble human leukocyte antigen G (sHLA-G) human serum albumin (HSA) and multinucleation 42 11?-hydroxysteroid dehydrogenase (11?-HSD) 313 hypoxia inducible factor (HIF-1) pathway 311–12 ICSI see intracytoplasmic sperm injection (ICSI) Igf2 expression in imprinting 236–7, 239 immune system overview 155–6 and preimplantation development 156–61 immunocytochemistry of NO synthase 174 implantation 145 and high ovarian response 211 and immune system 145 and impact of PGD on multiple pregnancies 288–9 implantation rates with assisted hatching 137 and immunosuppression 158 and biopsy type 220–1 and embryo transfer timing 117 and fragmentation 55, 55–7, 57 improvement with PGD 226–7 and multiday embryo scoring 112, 114 and multinucleation 48–9 and numbers of embryos for biopsy 227 and sHLA-G levels 151–2 imprinting 235, 236 maternal 236–7 paternal 236 in spermatozoa 326–7 imprinting control regions (ICRs) 237 in vitro environment see culture system in vitro fertilization (IVF) abnormal fertilization dispermy 6–7, 11 failed sperm incorporation 7, 8 chromosome abnormalities 208 culture system media composition 80–1 oxygen content 81 prescreening 80 dietary status of patient and 79 embryonic wastage 268 

holistic analysis 79, 80 ‘IVF dilemma’ 245 repeated cycles 319 and SCSA values 330–1 and sperm damage 331 timeline of events 101 in vitro produced cells and cryopreservation 125 in vivo embryo maturation 101, 102 inducible NOS (iNOS) 170 expression in embryos 174–5 immunocytochemistry 174 infertility and centrosome dysfunction 340–2 and centrosome function assays 342–3 see also fertility inhibins 315 innate immune system 155, 156 inner cell mass (ICM) assessment 109 in cryopreservation 128 formation 270 ICM-specific markers 270 Institute for Reproductive Medicine and Science (IRMS) 41 insulin 198 intracellular pH (pHi) 310 intracytoplasmic sperm injection (ICSI) digyny 7 and imprinting 240 and SCSA values 330–1 and sperm damage 331 and sperm DNA breaks 325 intrafollicular oxygen regulation 305–9 IVF see in vitro fertilization karyokinesis 95–6, 97 kinesin motors 347, 349 N-omega-nitro-L-arginine methyl ester (L-NAME) 172 lactate production 180, 181, 185, 193 laser, assisted hatching with 137, 140 leucine 186, 195 lipid metabolism 193 lipids, intracellular 125, 126 luteinizing hormone (LH) 302 in cortisol:cortisone ratio 313 macromolecules 198–9 magnetic sperm sorting 332 major histocompatability complex (MHC) 159–61

HPE_Index_new.qxp

7/16/2007

8:51 PM

Page 363

INDEX

class I proteins on preimplantation embryos 160 membrane-bound vs soluble 160–1 human vs mouse MHC 159, 159 and immune rejection 159–60 in implantation 145 inheritance 160 Nobel prize research 159 markers of ART outcome 16–17 follicular, search for 301–23 oxidative stress biomarkers 245–61 search for 270 viability 179, 186 mass spectrometry (MS) 249, 295 maternal age and centrosome anomalies 345 and chromosomal abnormalities 245, 288 and follicular biology/oocyte competence 316–19 influence on genes 266 and meiosis rescue 204 and mitochondrial activity 281 and multinucleation 42, 43 and oocyte competence 303 PB1 testing 201 see also older females maternal effect genes 268 maternal-zygotic transfer (MZT) 195, 292 and embryo metabolism 191 maturation promoting factor (MPF) 265 maturational status of oocytes 1–2 meiosis chromosome abnormalities in 201 male/female differences 347 testing for errors 205 see also blastomeres, sampling; polar bodies, FISH analysis meiosis II fragmentation in 71, 72 mitochondrial morphology 277 meiosis rescue (balancing) 204 meiotic spindle aging of 346 and cryopreservation–induced damage 126–7 messenger RNA (mRNA) during oogenesis 265–6 transcript quantification 293 metabolism cleavage stage vs postcompaction 180 CO2 production 180–1 precompaction 180

preimplantation embryos 179–80 species variation 180 and in vitro environment 191–200 metabolism measurement techniques 180–1 metabolite uptake/release 179–89 amino acid metabolism 185–6 carbohydrate metabolism 181–5 glucose 184 lactate 185 pyruvate 183–4 human chorionic gonadotropin (hCG) 186 oxygen consumption 186 as viability marker 179 metabolome 246 metabolomic profile/fingerprint 253 metabolomics 245–7 analytical techniques 248–9 detection methods 249 separation methods 248–9 applications in ART 252–60 current 252 key 250–1 case for 246–7 definition 246 definitions 246 history 247 indications in other fields 259–60 in reproductive health 259–60 studies Metabolomics Study Group for ART 253–8 other 259 metabolomics-based embryo selection 252 Metabolomics Study Group for ART 253–8 Bayesian probability model 258 instrumentation 254, 254 metabolomic profiling of OS biomarkers 255 results near infrared spectroscopy 257 Raman spectral signature 255, 256 study aims 253 study conclusion 257–8 study design 254–5, 255 study results 255–7 technology platform 253, 253 metabonomics 246 metaphase I mitochondrial morphology 277 methionine 195–7, 196 microarray CGH 292 

HPE_Index_new.qxp

7/16/2007

8:51 PM

Page 364

INDEX

microdrops 199 microfluorescence assays 181 micromanipulation 135 importance of skills 143 micronucleated blastomeres 47–8 microtubule asters 339, 340 microtubule motors 347–8 microtubule organization abnormal, after nuclear transfer 349 in fertilization failures 341 in inseminated oocytes 339 in polyspermy/parthenogenesis 340 without MTOCs 347 microtubule-organizing centers (MTOCs) 24, 337 absence 347 in fragmentation 69–70, 72–3 importance of 351 microtubule status in fragmentation 69–70, 70 mineral oil in culture medium 199 misdiagnosis in PGD 224, 225, 225 mitochondria 275–6 activity and embryo development 280–2 oocytes 280–1 in embryo selection 282–3 and localization 279 and maternal age 281 measurement in embryos 282–3 and fragmentation 68–9 localization and embryo development 278, 279–80 matrix 276 morphology and developmental characteristics 278, 279 in embryo selection 282 in embryos 278 in oocytes 281 germinal vesicle stage 277–9 immature 277 mature 277 in zygotes 278, 278 in oocytes/embryos 276–9 cleavage stage 281 after fertilization 278, 281 in reproduction 275–85 role of 275–6 mitochondrial ‘bottleneck’ theory 276 mitochondrial DNA 276 mitosis modulation by NO 173–4 spindle formation 

in dispermy 346 faulty, in aneuploidy 350 in parthenogenesis 348 MNEs see multinucleated embryos (MNEs) molecular biometrics 253, 260 molecular diagnostics 250 monosomies 217 monozygotic twinning 141 morphological abnormalities 96, 212–13 pronuclear 213 relation to chromosomal abnormalities 213 morphological selection against chromosome abnormalities 212–13 impact of 217 morphology analysis blastocysts 81 effect of score on pregnancy outcome 84 scoring system 81, 82–4 embryos 103–8 2-cell embryos 103–4 4-cell embryos 104, 106 8-cell embryos 106–7 cell number 106, 106–7 compaction 107 fragmentation 106, 107 symmetry 106, 107 cumulative scoring day 3 characteristics 107–8 multiday assessments 109, 110–14, 115–16 day 2 characteristics 104, 105, 106, 111 day 3 characteristics 107–8, 111, 113 zygote stage 103 and scoring systems 103–8 morphometric analysis advantages 89–90 blastomere size during cleavage 93–7 asymmetric blastomeres 96 cell size cut-off limits 96 and fragmentation 93–6, 95, 97 total cytoplasmic volume 95, 95–6 multinucleation 96–7 variations in size/cell generations 94 digital image analysis multilevel system 90–1, 91, 92, 95 single image system 90 embryos 89–99 kinetics of early development 97–8 nuclei, size of 96–7 oocyte size 92–3 technique 90

HPE_Index_new.qxp

7/16/2007

8:51 PM

Page 365

INDEX

time-lapse monitoring 91 zygote size 92–3 morula 2, 156 E-cadherin localization 65 mosaicism cleavage stage patterns 215, 216 and fragmentation 68 and multinucleation 45–6, 46, 47 and PGD misdiagnosis 224, 225, 225 and polar body chromosome abnormalities 204–5 pronuclei in 24, 25, 38 in unselected blastocyst studies 215–16 multiday embryo scoring 109, 115–16 novel method 116 studies prospective 113–14 retrospective 110–12 multinucleated blastomeres (MNBs) 41, 44, 52 and chromosomal abnormalities 210, 211 morphometric analysis 95–6, 97 multinucleated embryos (MNEs) 41 multinucleation 47–8 in 4-cell embryo assessment 106 chromosomal status of embryos 44–8 MNEs 44–8 and severity of multinucleation 46, 47–8 non-MNEs 46, 46–7 and chromosome abnormalities 212 clinical outcomes 41, 44, 48, 48–9 day of multinucleation 48–9 and culture conditions 42–3 definition 41 frequency of 41–3 impact on embryo morphology 41, 44, 45 and maternal age 42 mechanisms 43–4, 45 morphometric analysis 96–7 and mosaicism 45–6, 46, 47 and repetition of IVF cycles 42–3, 43 types of 41, 42 multiple embryo transfer 245, 269 legal controls 305 outcomes 305 multiple pregnancies implantation, impact of PGD on 288–9 ovarian stimulation 287 reducing 287 natural killer (NK) cells in implantation 148

inhibition by HLA-G 146, 147, 148, 150–1 near infrared (NIR) spectroscopy 249 follicular fluid 259 Metabolomics Study Group for ART 257 from seminal plasma 258 neonatal outcome with fragmentation 57–8 neuronal NOS (nNOS) 169–70 expression in embryos 174, 175 immunocytochemistry 174 newborn ovary homeobox (Nobox) gene 265 nitric oxide inhibitors 172 nitric oxide (NO) 169–70 induction of excessive levels 172 regulation of preimplantation embryos 169–78 signal transduction modulation 173–4 multiple mechanisms 175–6 nitric oxide synthase (NOS) 169 expression in embryos 174–6 see also endothelial NOS (eNOS); inducible NOS (iNOS); neuronal NOS (nNOS) nitric oxide synthase knockout mice 175–6 NG-nitro-L-arginine (NLA) 172 NO see nitric oxide (NO) Nobox gene 265 nuclear envelope formation 22, 23 nuclear magnetic resonance (NMR) spectroscopy 249 nuclear transfer (NT) abnormal microtubule patterns 348, 349 centrosomes during 348–50 in mammals 348 nuclear transfer cloning 350–1 nucleolar precursor bodies (NPBs) 23, 23 alignment and pronuclear scoring 32 and RNA synthesis 25 size/distribution 26 visualization 25 nucleus size in morphometric analysis 96–7 nutrigenomics 251 older females endogenous amino acid pool 197 see also maternal age oocytes 1 activated 2 activation 21–2 aneuploidy/dysmorphism development 303 assessment 1–5 maturational status 1–2 mature egg, structure of 2–5, 3 without blastomere biopsy 295–7 

HPE_Index_new.qxp

7/16/2007

8:51 PM

Page 366

INDEX

oocytes (Contd.) defect detectability 303–4 esthetics, unreliablilty of 303 growth to terminal size 301 maturation/maturational status 4 mouse 156 organelles 2, 3, 4 GVBD, oocytes in 4 immature oocyte 3, 3–4 mature oocyte 3, 3 maturing oocyte 3 postmature (aging) oocytes 4, 4–5, 5 unfertilized 7 polar body 3 polyspermic penetration 21 quiescence 301 size 92–3 spindle, metaphase II 2, 4, 5 oocyte competence and aging 303 and Doppler ultrasonography 308–9 effect of maternal age/lifestyle 316–19 follicular fluid, determinants in 313–15 follicular markers, search for 301 oocyte development atypical 51, 52 E-cadherin distribution 64, 65 normal 51, 52 oocyte selection 1 in contemporary IVF 301–23 oogenesis 301 mRNA during 265–6 osmolarity 191 outcome predictors of ART 16–17 ovarian hyperstimulation and euploidy 211 see also controlled ovarian hyperstimulation (COH) ovarian hyperstimulation syndrome (OHSS) 312 ovarian stimulation current trends 1 and higher-order multiple pregnancies 287 and pronuclear developmental perturbations 27 repeated, adverse effects of 319 oxidative stress markers 245–61 oxygen consumption 186 analysis 181 measurement in embryos 282 intrafollicular 309–11 

and VEGF levels 312 reduced, in vitro 191 Parkinson’s disease 260 parthenogenesis 7, 21, 337, 340 and centrosome anomalies 347–8, 348 and mosaicism 24 Ped gene 161–4 and embryo survival 162–3 mechanism of action 161–2 mouse model 163–4 clinical relevance 164, 164 and Qa-2/HLA-G 162–3 pentose phosphate shunt 182, 182 pericentrosomal matrix (PCM) 337 perifollicular blood flow and intrafollicular oxygenation 311 in oocyte selection 305–8, 311 rates, Doppler-assessed 305–8 and embryo transfer outcome 307 and 11?-HSD 314 as oocyte/embryo competence marker 308–9 and VEGF levels 312 perifollicular blood flow regulation 309–11 PGD see preimplantation genetic diagnosis (PGD) pH intracellular (pHi) 309 in vitro 191 pharmacodiagnostics 250 physiology analysis of blastocysts 81, 84, 85 platelet activating factor (PAF) 199 and Qa-2 levels 163–4 polar bodies 3 alignment and pronuclear scoring 32 chromosome abnormalities 201–7 detection rate 203 embryos, relationship with 204–6 frequency/types 202–4 first (PB1) 201 FISH analysis 201 detection of aneuploidies 205 missing chromatids after 219 preimplantation testing 205 technique 202 second (PB2) 201 as fertilization marker 21 polar body biopsy vs cleavage stage biopsy 218–20 Polscope readings 17 polyploidy 215, 217 polyspermic oocyte penetration 21

HPE_Index_new.qxp

7/16/2007

8:51 PM

Page 367

INDEX

nucleolar precursor bodies in 23 polyspermy 337, 340, 343, 345 postmeiotic abnormalities 209 and embryonic competence 215 Prader-Willi syndrome 196, 326, 345 pre-eclampsia, HLA-G in 147, 150 pregnancy rates ‘early cleavage’ embryos 103–4 and embryo transfer timing 117 embryo transfer timing, optimum 116–18 and fragmentation 55, 55–7, 57 improvement with PGD 226–7 after IVF with damaged sperm 330 and multiday embryo scoring 112, 114 and multinucleation 49 and numbers of embryos for biopsy 227 and sHLA-G levels 151–2 preimplantation development early 179–80 effect of Qa–2 163 effect of sex 163 and immune system 156–61 nitric oxide in 171–3 timeline of events 101 preimplantation embryos MHC class I proteins on 160 membrane-bound vs soluble 160–1 mouse 155, 156 preimplantation genetic diagnosis (PGD) 141–2, 269–70 blastocysts after 84 cell loss in 220 in chromosomal status determination 44–5 chromosomes selected 291 clinical impacts 228–9 future directions 289–92 impact on multiple pregnancies 288–9 indications 217–18, 227–8 in infertility, rationale for 217 methodology 217–27 biopsy technique 221 cell type/number analyzed 218–21 embryo transfer technique 224–5 FISH error minimization 222–3 fixation method 221–2 implantation rates and embryos for biopsy 227 improvement 226–7 misdiagnosis and mosaicism 224, 225, 225 pregnancy rates and embryos for biopsy 227

improvement 226–7 probes 222 spontaneous abortion, decrease in 226 take-home baby rate improvement 226–7 technical errors 223 rescue 223–4 trisomy reduction 225–6 number of procedures performed 218 polar body chromosome abnormalities 204, 206 vs cleavage stage biopsy 218–19 prognosis in subsequent ART 228–9 and reproductive history of couples 228 as selection method 217–29 studies, summary of 218 viability assessment 288–9 see also embryo biopsy preimplantation genetic screening see preimplantation genetic diagnosis (PGD) primordial germ cells (PGCs) 263–4 probes in PGD 222 procentrosomes 342 pronuclear morphology 213 pronuclear scoring systems 25, 26–7, 33, 34, 35–7 0 patterns 34, 35, 35, 37–8 criteria 31–2 cumulative point scoring 35–6, 36 methodology 31–2 parameters 32–3 pronuclear morphology scoring (PNMS) systems 36–7, 37, 38 studies prospective 113 retrospective 110 studies using zygote scoring systems 37–9 Z score 33, 33, 34, 35, 35 pronuclei 5–6, 21–9 alignment and pronuclear scoring 32, 38 analysis in living zygotes 25 possibilities/limits 25–6 asymmetric 25 developmental changes 27 function 23–4 structure 23 establishment 22 experimental approaches 22 as fertilization marker 21 interpronuclear synchrony 24–5 intrapronulcear structure polarization 24 male 23, 23, 340 

HPE_Index_new.qxp

7/16/2007

8:51 PM

Page 368

INDEX

pronuclei (Contd.) in polar axis formation 24 in polyspermically penetrated oocyte 25 movement 24 recovery from abnormalities 26 in viability prediction 31–40 visualization 25 propanediol 124 and cortical granules 126 protein growth factors 314–15 proteomic embryo analysis 295 purine metabolism 193 pyruvate metabolism 180, 181, 182–4, 193 in fertilization/first cleavage 183, 183 with/without glutamine 183 and transfer time 184, 184 Qa-2 effect of presence/absence 162–3 on preimplantation embryo surface 161, 162 as signaling molecule 162, 163 Raman and NIR spectroscopy 249 Metabolomics Study Group for ART 254, 255, 256, 257 reactive oxygen species (ROS) effect of glucose excess 193 metabolomics 253 protection against 191, 199 recurrent pregnancy loss (RPL) 227 recurrent spontaneous abortions 150 redox potential in vitro 191–2 repeated implantation failure (RIF) 227 respiration blastocyst 281 and maternal age 281 respiratory chain 276 respirometry 252 RNA precursors 198 synthesis, intrapronulcear 25 see also messenger RNA (mRNA) S-adenosyl methionine (SAM) 196 scoring systems morphology analysis 103–8 2-day embryos 104, 105, 106, 111 3-day embryos 107–8, 111 blastocysts 81, 82–4 see also pronuclear scoring systems 

SCSA see sperm chromatin structure assay (SCSA) securin 267 sex of embryo and glucose levels 192–3 and preimplantation development 163 signal transduction cascade after GVBD 266 smoking 318–19, 327 smooth endoplasmic reticulum (SER) 2, 5 sodium nitroprusside (SNP) 172 SNP-induced apoptosis 172 soluble human leukocyte antigen G (sHLA-G) 145–6, 146 in apoptosis of activated CD8+ T cells 148, 149 assessment in human embryos 145–54 distribution in pregnancy 147 expression in pregnancy 146 inhibition of NK cells 149 in normal vs pathological pregnancies 149–51 on preimplantation embryos 160–1 secretion from cytotrophoblasts 148 see also human leukocyte antigen G (HLA–G) sperm asters see sperm-derived centrosomes (sperm asters) sperm centriole structure 343, 344 sperm chromatin structure assay (SCSA) 328, 329 and clinical outcome 330–1 parameters measured 330 sperm-derived centrosomes (sperm asters) 6, 24, 339–40 centrosome function assays 342–3 clinical assays 343, 344 dysfunction and infertility 340–2 during fertilization 339–40 phenotypic expression in bulls 341 sperm DNA damaged, removal of spermatozoa with 332 and embryo development 325–35 remodeling 328 repair 327 and reproductive outcome 326–7 strand breaks 327 and reproductive outcome 328–31, 329 sperm monoaster formation 6 sperm-oocyte fusion 21 sperm reconstitution model 343 spermatozoa apoptosis 327–8 components 326 damaged inheritance of 325 novel removal techniques 332

HPE_Index_new.qxp

7/16/2007

8:51 PM

Page 369

INDEX

reproductive outcome 332 spindle structural proteins 347–8 spindles checkpoints 267 and cryopreservation-induced damage 126–7 formation in dispermy 346 after nuclear transfer 350 in parthenogenesis 348 in fragmentation 73 meiotic spindle 346 metaphase II 2, 4, 5 spontaneous abortion 226 surface-enhanced laser desorption and ionization time of-flight mass spectrometry (SELDI-TOF MS) 295 survival rates and embryo quality 129 symmetry 106, 107 systems biology 248

viability assessment morphology 287–8 preimplantation genetic screening 288–9 and gene expression 292–5 markers 179, 186 oxidative stress markers 245–61 non-invasive analysis see blastocysts, physiology analysis; morphology analysis; morphometric analysis prediction using pronuclei 31–40 and sHLA-G levels 152 vitamins 198 vitrification 124, 128 volatile compounds and ART 211

temperature and impact on ART 211 testicular sperm extraction and intracytoplasmic sperm injection (TESE-ICSI) 326–7 theranostics 250 time-lapse monitoring of morphometric analysis 91 tobacco smoking 318–19, 327 total cytoplasmic volume 93, 95 morphometric analysis 95–6 multinucleated blastomeres 97 toxicity assessment with metabolomics 251 Tp53 gene 270 transamination 182, 182, 194, 197 trisomies age-related increase 201 cleavage stage patterns 216–17 and fragmentation 58 reduction with PGD 225–6 trisomy/monosomy ratio discordance 205 ‘trisomy rescue’ 204 trophectoderm (TE) E-cadherin localization 65 formation 270 TE-specific markers 270 trophoblast cells, fetal 148 tumor protein p53 (Tp53) gene 270 TUNEL assay 129, 328

Z score system 33, 33, 34, 35, 35 prospective studies 113–14 zona drilling 135–6, 137, 140 patient selection 135 zona opening techniques 137, 139–40 zona pellucida 15–19, 156–7 assessment of morphology 16–17 in cryopreservation 127 embryonic cell surface 158 during fertilization/implantation 15 filaments 15 function 15 hardening 126 hatching problems 16 and immune system 157–8 immunological role 15 intact/removed from blastocyst 157 as mechanical barrier 157–8 modification see assisted hatching (AH) technique modulation 15–16 morphological changes in ART 16–17 external influences 16 proteins 157 detection 158 mRNA expression 158 schematic of protective elements 158 surface morphology and ART outcome 157 thickness and ART outcome 157

uniparental disomy disorders (UDD) 206, 345 vascular endothelial growth factors (VEGF) 311, 312

weight and fertility 318 whole genome amplification (WGA) 290



HPE_Index_new.qxp

7/16/2007

8:51 PM

Page 370

INDEX

zona pellucida (Contd.) in morphometric analysis 92, 93 thinning 16 zona proteins 15 zona pellucida thickness variation (ZPTV) 16–17 zona pellucida thickness (ZPT) 16, 17 calculation 17 zygotes 156 development



features of 101–2 morphology analysis 103 developmental characteristics 280 embryonic function 22 evaluation, predictive value of 26 morphology 280 physiology 21–2 size 92–3 zygote centrosomes 339

Human Preimplantation Embryo Selection

Preimplantation Genetic Diagnosis

Practical Preimplantation Genetic Diagnosis

Preimplantation Genetic Diagnosis

Embrion ludzki – pytanie o status Human Embryo Status

Single Embryo Transfer

Embryo Culture: Methods and Protocols

Unnatural Selection

Unnatural Selection

Natural Selection

Unnatural Selection

The Selection

Selection Event

Natural Selection

Unnatural Selection

Selection Event

Natural Selection

Unnatural Selection

Cultural Selection

Unnatural Selection

Unnatural Selection

Unnatural Selection

Natural Selection

Unnatural Selection

Unnatural Selection

Natural Selection

Natural Selection

Unnatural Selection

Unnatural Selection

Unnatural Selection

The Selection

Human Preimplantation Embryo Selection

Preimplantation Genetic Diagnosis

Practical Preimplantation Genetic Diagnosis

Preimplantation Genetic Diagnosis

Embrion ludzki – pytanie o status Human Embryo Status

Single Embryo Transfer

Embryo Culture: Methods and Protocols

Unnatural Selection

Unnatural Selection

Natural Selection

Unnatural Selection

The Selection

Selection Event

Natural Selection

Unnatural Selection

Selection Event

Natural Selection

Unnatural Selection

Cultural Selection

Unnatural Selection

Unnatural Selection

Unnatural Selection

Natural Selection

Unnatural Selection

Unnatural Selection

Natural Selection

Natural Selection

Unnatural Selection

Unnatural Selection

Unnatural Selection

The Selection

Recommend Documents