Biotechnology Annual Review, Vol 13

Biotechnology Annual Review Volume 13 Biotechnology Annual Review Volume 1 Volume 2 Volume 3 Volume 4 Volume 5 Volume...

Author: M. Raafat El-Gewely

91 downloads 2700 Views 5MB 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

Biotechnology Annual Review Volume 13

Biotechnology Annual Review Volume 1 Volume 2 Volume 3 Volume 4 Volume 5 Volume 6 Volume 7 Volume 8 Volume 9 Volume 10 Volume 11 Volume 12 Volume 13

editor M.R. El-Gewely, Published by Elsevier Science B.V., Amsterdam, editor M.R. El-Gewely, Published by Elsevier Science B.V., Amsterdam, editor M.R. El-Gewely, Published by Elsevier Science B.V., Amsterdam, editor M.R. El-Gewely, Published by Elsevier Science B.V., Amsterdam, editor M.R. El-Gewely, Published by Elsevier Science B.V., Amsterdam, editor M.R. El-Gewely, Published by Elsevier Science B.V., Amsterdam, editor M.R. El-Gewely, Published by Elsevier Science B.V., Amsterdam, editor M.R. El-Gewely, Published by Elsevier Science B.V., Amsterdam, Chief editor M.R. El-Gewely, Published by Elsevier Science B.V., Amsterdam, Chief editor M.R. El-Gewely, Published by Elsevier B.V., Amsterdam, 2004 Chief editor M.R. El-Gewely, Published by Elsevier B.V., Amsterdam, 2005 Chief editor M.R. El-Gewely, Published by Elsevier B.V., Amsterdam, 2006 Chief editor M.R. El-Gewely, Published by Elsevier B.V., Amsterdam, 2007

1995 1996 1997 1998 2000 2000 2001 2002 2003

More information on the above mentioned volumes can be accessed at: http://www.elsevier.com/locate/issn/13872656

Biotechnology Annual Review Volume 13

Chief Editor M. Raafat El-Gewely Department of Biotechnology University of Tromsø Tromsø, Norway

Amsterdam – Boston – Heidelberg – London – New York – Oxford Paris – San Diego – San Francisco – Singapore – Sydney – Tokyo

Elsevier Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands Linacre House, Jordan Hill, Oxford OX2 8DP, UK First edition 2007 Copyright r 2007 Elsevier B.V. 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 written permission of the publisher Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333; email: [email protected]. Alternatively you can submit your request online by visiting the Elsevier web site at http://www.elsevier.com/locate/permissions, and selecting Obtaining permission to use Elsevier material Notice No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN: 978-0-444-53032-5 ISSN: 1387-2656 For information on all Elsevier publications visit our website at books.elsevier.com Printed and bound in The Netherlands 07 08 09 10 11 10 9 8 7 6 5 4 3 2 1

v

Foreword

Biotechnology, Health, and Peace

Progress in biotechnology is a prerequisite for global health, and global health is a prerequisite for peace in the world. Therefore, biotechnology, health, and peace are strongly linked together. Unfortunately, the gap between developing and industrialized countries continues to increase [1]. It is, therefore, immensely important that the international community becomes active in reducing this gap by promoting science and technology. There is a tremendous potential in the genomics revolution to promote health in developing countries [1], and thus promoting peace with justice in the world. The 2001 United Nations Millennium Development Goals (MDGs) by 2015 are:

Eradicate extreme poverty and hunger. Achieve universal primary education. Promote gender equality and empower women. Reduce child mortality. Improve maternal health. Combat HIV/AIDS, malaria, and other diseases. Ensure environmental sustainability. Develop a global partnership for development.

The 2001 United Nations Development Program report, Making New Technologies Work for Human Development, identified technical progress as the largest factor in reducing mortality rates and improving life expectancy from 1960 to 1990 [2]. Addressing health challenges of the developing world will require new forms of international partnerships that take into account emerging opportunities in the globalization of scientific knowledge [3]. A few years ago an international panel of 28 scientists, experts in biotechnology and global health, were asked to answer the question: ‘‘What do you think are the major biotechnologies that can help improve health in developing countries in the next five to ten years?’’ [4]. The list of ‘‘top ten biotechnologies’’ was:

molecular diagnostics, recombinant vaccines, vaccine delivery systems, bioremediation,

vi

sequencing pathogen genomes, female-controlled protection against sexually transmitted diseases, bioinformatics, nutritionally enhanced genetically modified crops, recombinant therapeutic proteins, and combinatorial chemistry.

The ‘‘top ten biotechnologies’’ fit in well with the United Nations MDGs. Therefore, the important challenge is to achieve worldwide implementation of these technologies to reach the MDGs by the year 2015 [4]. It is obvious that vaccine programs are essential for global health. Development of new and efficient vaccines is dependent on modern biotechnology. As early as 1988, the World Health Assembly passed a resolution to eradicate poliomyelitis by the year 2000. By the turn of the 21st century, we have nearly reached this goal, although war seriously disrupted the polio campaign in Afghanistan, Angola, the Democratic Republic of Congo, Liberia, Sierra Leone, Somalia, Sudan, and Tajikistan, and less severe difficulties in several other regions also disrupted the campaign [5,6]. Nevertheless in several countries or regions, the health authorities made agreements with the combatants to cease fire while the vaccinations proceeded. For example, in the Democratic Republic of Congo and in Afghanistan, the UN or its agencies helped ‘‘negotiate’’ ‘‘days of tranquility’’ to allow the vaccination program to carry on [6]. MacQueen et al. [6] state: ‘‘Notice the health-peace symbiosis here: institutions devoted to negotiations and peacemakings assist an enormously important health initiative, while the health campaign impedes the progress of armed conflict. This symbiosis has important implications, not just for the eradication of polio, but for the achievement of broader health objectives and ultimately the attainment of some semblance of sustainable peace’’ [7]. MacQueen et al. [6] further states: ‘‘The important question is not whether peace must be pursued in order to health to flourish – the answer to this is obvious – but health workers can contribute in a unique way to peace-making. We believe they can. But there is a need for a new discipline of ‘peace through health’ that studies both the downward spiral of war and disease and the positive symbiosis of peace and health’’. Health as a Bridge for Peace (HBP) was formally accepted by the 51st World Health Assembly in May 1998 as a feature of the ‘‘Health for All in the 21st Century’’ strategy. Since that time, WHO has worked continuously with the issue of peace-through health. HBP is a multidimensional policy and planning framework which supports health workers in delivering health programs in conflict and postconflict situations, and at the same time contributes to peace-building. The HBP concept is based on values derived from human rights and humanitarian principles as well as medical ethics. It is supported by the conviction that it is imperative to adopt peace-building strategies to ensure lasting health gains in the context of social instability and complex emergencies. In 2004, the Nobel Peace Prize was awarded to Wangari Maathai from Kenya to underscore the importance of sustained development for peace. In 2006, the Nobel Peace Prize was awarded to Mohammad Yunus from Bangladesh and Grameen Bank for their fight against poverty through microcredit. Yunus said: ‘‘We have created a

vii slavery-free world, a polio-free world, an apartheid-free world. Creating a povertyfree world would be greater than all these accomplishments while at the same time reinforcing them. This would be a world that we could all be proud to live in’’. In fighting diseases and poverty in the world, modern biotechnology is an immensely important tool. Therefore, progress in biotechnology is of major importance both for global health and world peace. Professor Ole D. Mjøs, MD, Ph.D. Chairman of the Norwegian Nobel Committee, Professor of Medical Physiology, Institute of Medical Biology, University of Tromsø, Norway

Acknowledgements I am thankful that the chief editor and my good colleague for many years, Professor M. Raafat El-Gewely, for inviting me to write the foreword of this volume.

Biographical

Ole D. Mjøs, was born on 8th March 1939. He is Professor of Medical Physiology, University of Tromsø, Norway (from 1975). He was Dean of Medical Faculty, University of Tromsø (1983–86) and Rector of University of Tromsø (1989–96). He is Chairman of the Norwegian Nobel Committee (from 2003).

viii References [1] [2]

[3] [4] [5] [6] [7]

Acharya T, Daar AS, Thorsteinsdottir H, Dowdeswell E and Singer PA. PLoS Med 2004;1(3):e40. United Nations Development Programme. Human Development Report 2001: Making New Technologies Work for Human Development, Oxford, Oxford University Press, 2001. p. 278. Juma C and Yee-Cheong. Reinventing global health: the role of science, technology and innovation. Lancet 2005;365:1105–1107. Acharya T, Daar AS and Singer PA. Biotechnology and the UN’s millennium development goals. Nature Biotechnol 2003;21(12):1434–1436. Tangermann R, Hull H, Jafari H, Nkowane B, Everts H and Aylward R. Eradication of poliomyelitis in countries affected by conflict. Bull World Health Org 2000;78:330–338. MacQueen G, Santa.Barbara J, Neufeld V, Yusuf S and Horton R. Health and peace: time for a new discipline. Lancet 2001;357:1460–1461. Bush K. Polio, war and peace. Bull World Health Org 2000;78:281–282.

ix

Editorial board Chief Editor M. Raafat El-Gewely, Department of Molecular Biotechnology, Gene Systems Group, Institute of Medical Biology, University of Tromsø, 9037 Tromsø, Norway, E-mail: [email protected]

Editors MaryAnn Foote, MaryAnn Foote Associates, 4284 Par Five Court, Westlake Village, California 91362, USA, E-mail: [email protected] Guido Krupp, AmpTec GmbH, Koenigstr. 4A, D-22767 Hamburg, Germany, E-mail: [email protected] Alfons Lawen, Monash University, Clayton Campus, Department of Biochemistry and Molecular Biology, Room 312, Building 13D, Clayton, Victoria 3800, Australia, E-mail: [email protected]

Associate Editors Carlos F. Barbas, The Skaggs Institute for Chemical Biology and the Departments of Molecular Biology and Chemistry, The Scripps Research Institute, 10550 North Torrey Pines Rd., La Jolla, CA 92037, USA, E-mail: [email protected] Marin Berovicˇ, Department of Chemical and Biochemical Engineering, University of Ljubljana, Askercˇeva 5, 1001 Ljubljana, Slovenia, E-mail: marin.berovic@fkkt. uni-lj.si Thomas M.S. Chang, Artificial Cells & Organs Research Centre, McGill University, 3655 Drummond St., Room 1005, Montreal, Quebec, Canada H3G 1Y6, E-mail: [email protected] Thomas T. Chen, University of Connecticut, Molecular & Cell Biology, TLS 415, 75 North Eagleville Road, Unit 3125, Storrs, CT 06269-3125, USA, E-Mail: Thomas. [email protected] Frank Desiere, F. Hoffmann – La Roche Ltd, Diagnostics Division, Bldg. 52/1607, CH-4070 Basel, Switzerland, E-mail: [email protected]

x Franco Felici, Department of Microbiological, Genetic and Molecular Science, University of Messina, Salita Sperone 31, 98166 Messina, Italy, E-mail: franco.felici @unime.it Leodevico (Vic) L. Ilag, Patrys Ltd, Lv2, 517 Flinders Lane, Melbourne, VIC 3000, Australia, E-mail: [email protected] Kuniyo Inouye, Laboratory of Enzyme Chemistry, Division of Food Science and Biotechnology, Graduate School of Agriculture, Kyoto University, Sakyo-ku, Kyoto 606-8502, Japan, E-mail: [email protected] K. John Morrow, 625 Washington Avenue, Newport, KY 41071, USA, E-mail: [email protected] Jocelyn H. Ng, CSIRO Molecular and Health Technologies, 343 Royal Parade, Parkville, VIC 3052, Australia, E-mail: [email protected] Eric Olson, Vertex Pharmaceuticals, 130 Waverly Street, Cambridge, MA 02139, USA, E-mail: [email protected] Vincenzo Romano-Spica, University Institute of Motor Science, IUSM, P.zza Lauro e Bosis 15, 00194 Rome, Italy, E-mail: [email protected]

xi

List of contributors Arjumand Ather, The Norwegian Structural Biology Center (NorStruct), Department of Chemistry, University of Tromsø, Tromsø 9037, Norway Marin Berovic, Department of Chemical and Biochemical Engineering, Faculty of Chemistry and Chemical Engineering, University of Ljubljana, Askerceva 5, 1001, Ljubljana, Slovenia Bojana Boh, Faculty of Natural Sciences and Engineering, University of Ljubljana, Vegova 4, 1000 Ljubljana, Slovenia Marcus Droege, Roche Diagnostics, Penzberg, Germany Loı¨ c Faye, Universite´ de Rouen, CNRS UMR 6037, IFRMP 23, GDR 2590, Faculte´ des Sciences, Baˆt. Ext. Biologie, 76821 Mont-Saint-Aignan cedex, France MaryAnn Foote, MA Foote Associates, 4284 Par Five Court, Westlake Village, CA 91362, USA Ve´ronique Gomord, Universite´ de Rouen, CNRS UMR 6037, IFRMP 23, GDR 2590, Faculte´ des Sciences, Baˆt. Ext. Biologie, 76821 Mont-Saint-Aignan cedex, France Yasuhiko Hashida, Division of Food Science and Biotechnology, Graduate School of Agriculture, Kyoto University, Sakyo-ku, Kyoto 606-8502, Japan G. Wesley Hatfield, The Institute for Genomics and Bioinformatics, Donald Bren School of Information and Computer Sciences, University of California, Irvine, CA 92497, USA Louis-Marie Houdebine, Biologie du De´veloppement et Reproduction; Institut National de la Recherche, Agronomique, 78350 Jouy en Josas, France Kuniyo Inouye, Division of Food Science and Biotechnology, Graduate School of Agriculture, Kyoto University, Sakyo-ku, Kyoto 606-8502, Japan Mahmud Tareq Hassan Khan, School of Molecular and Structural Biology and Department of Pharmacology, Institute of Medical Biology, University of Tromsø, Tromsø 9037, Norway Masayuki Kusano, Division of Food Science and Biotechnology, Graduate School of Agriculture, Kyoto University, Sakyo-ku, Kyoto 606-8502, Japan

xii Thomas Laurell, Department of Electrical Measurements, Lund Institute of Technology, Lund University, P.O. Box 118, SE-221 00 Lund, Sweden Matic Legisa, National Institute of Chemistry, Hajdrihova 19, 61000 Ljubljana, Slovenia David Lieńard, Universite´ de Rouen, CNRS UMR 6037, IFRMP 23, GDR 2590, Faculte´ des Sciences, Baˆt. Ext. Biologie, 76821 Mont-Saint-Aignan cedex, France Gyo¨rgy Marko-Varga, Department of Analytical Chemistry, Lund University, P.O. Box 124, S-221 00 Lund, Sweden Masashi Minoda, Daiwa Kasei K.K., 4-19 Hie-cho, Konan-shi, Shiga 520-3203, Japan K. John Morrow Jr., Newport Biotechnology Consultants, 625 Washington Avenue, Newport, KY 41071, USA Anton Ressine, Department of Electrical Measurements, Lund Institute of Technology, Lund University, P.O. Box 118, SE-221 00, Lund, Sweden David A. Roth, CODA Genomics Inc., 26061 Merit Circle # 101 Laguna Hills, CA 92653-7015, USA Eric Soler, Cell Biology Department, Erasmus MC, dr. Molewaterplein 50, 3015 GE, Rotterdam, The Netherlands Christophe Sourrouille, Universite´ de Rouen, CNRS UMR 6037, IFRMP 23, GDR 2590, Faculte´ des Sciences, Baˆt. Ext. Biologie, 76821 Mont-Saint-Aignan cedex, France Carsten VoX, Fermentation Engineering, Faculty of Technology, Bielefeld University, Germany Kiyoshi Yasukawa, Division of Food Science and Biotechnology, Graduate School of Agriculture, Kyoto University, Sakyo-ku, Kyoto 606-8502, Japan Jingsong Zhang, Institute of Edible Fungi, Shanghai Academy of Agriculture Sciences, Shanghai, P.R. China Lin Zhi-Bin, Department of Pharmacology, Peking University Health Science Center, Beijing 10083, P.R. China Burkhard Ziebolz, Roche Diagnostics, Penzberg, Germany

xiii

Contents Foreword Editorial board List of contributors

v ix xi

Toward a new era in sequencing Burkhard Ziebolz and Marcus Droege

1

Optimizing scaleup yield for protein production: Computationally Optimized DNA Assembly (CODA) and Translation EngineeringTM G. Wesley Hatfield and David A. Roth

27

Engineering, expression, purification, and production of recombinant thermolysin Kuniyo Inouye, Masayuki Kusano, Yasuhiko Hashida, Masashi Minoda and Kiyoshi Yasukawa

43

Preparation of recombinant vaccines Eric Soler and Louis-Marie Houdebine

65

Advances in antibody manufacturing using mammalian cells K. John Morrow Jr.

95

Pharming and transgenic plants David Lieńard, Christophe Sourrouille, Ve´ronique Gomord and Loı¨c Faye

115

Porous silicon protein microarray technology and ultra-/superhydrophobic states for improved bioanalytical readout Anton Ressine, Gyo¨rgy Marko-Varga and Thomas Laurell

149

Production of plasmid DNA for pharmaceutical use Carsten VoX

201

Potentials of phenolic molecules of natural origin and their derivatives as anti-HIV agents Mahmud Tareq Hassan Khan and Arjumand Ather

223

Ganoderma lucidum and its pharmaceutically active compounds Bojana Boh, Marin Berovic, Jingsong Zhang and Lin Zhi-Bin

265

Citric acid production Marin Berovic and Matic Legisa

303

xiv Using nanotechnology to improve the characteristics of antineoplastic drugs: Improved characteristics of nab-paclitaxel compared with solvent-based paclitatel MaryAnn Foote

345

Index of authors Keyword index

359 361

1

Toward a new era in sequencing Burkhard Ziebolz and Marcus Droege Roche Diagnostics, Penzberg, Germany Abstract. Sequencing is a powerful tool that helps scientists in gaining new insights in many areas of medicine and biology. The electrophoresis-based Sanger method is currently the most popular sequencing technology and was the foundation stone of the human genome project. With the Sanger technique it became possible to sequence not only complete genomes, but also fragments of genomes. Nowadays, this standard method is very close to reach its limits. Keywords: gene therapy, genetic vaccination, plasmid, manufacturing, cultivation, cell lysis, recombinant RNase, chromatography, affinity separation, capillary gel electrophoresis, stability

Sequencing technologies of the next generation Sequencing is a powerful and versatile tool that helps scientists in gaining new insights in many areas of medicine and biology. Thanks to sequencing we now know much more about the molecular principles of diseases. New findings about drug resistance have been obtained through comparative sequencing of drug resistant genomes of pathogenic microorganisms and their drug sensitive counterparts. Sequencing of complete or partial viral genomes from clinical samples has given us some idea of the course of an infection over time, the response to treatments and indications of possible strategies for the future development of drugs. The electrophoresis-based Sanger method is currently the most popular sequencing technology and was the foundation stone of the human genome project. With the Sanger technique it became possible to sequence not only complete genomes, but also fragments of genomes, for example for clone verification or for the detection of mutations in the genome that may be related to the development of illnesses. The method has been steadily improved over the past 10 years. During this period sequencing costs have fallen by around 90% while, at the same time, the throughput of a modern automated sequencer has increased 10-fold. Nowadays, this standard method is very close to reach its limits. If sequencing is ever to become a constituent part of individualized diagnosis and preventive medicine – and the potential is certainly there – the costs must be reduced still further. The international planning target is 1,000 dollars for a complete human genome. Corresponding author: Tel: +8856-60-4830. Fax: +8856-60-7881.

E-mail: [email protected] (B. Ziebolz). BIOTECHNOLOGY ANNUAL REVIEW VOLUME 13 ISSN 1387-2656 DOI: 10.1016/S1387-2656(07)13001-5

r 2007 ELSEVIER B.V. ALL RIGHTS RESERVED

2 New sequencing techniques are also of interest to the pharmaceutical sector, for example in research on drug resistance and the pathogenicity of bacteria, for identifying human DNA variations [1,2], establishing the onset of drug resistance in HIV [3,4] or HCV, producing tumor profiles as a guide to cancer treatments [5] or for differentiating the modes of action of antibiotics [6]. A whole range of research institutes and companies are now working on technologies designed to lower costs and increase throughput. While most of these are still in their development phase, one – the 454 sequencing technology – is already commercially available and perfectly suited to address a variety of different sequencing applications in the fields of whole genome sequencing, transcriptome and gene regulation studies as well as amplicon analysis. Microelectrophoretic methods Microelectrophoretic methods are based on existing technologies of capillary electrophoresis used in Sanger sequencing. One advantage is a smaller electrophoresis platform – with consequent savings on reagent costs. In addition, a greater number of lanes can be used in the electrophoresis [9–11], and sample preparation and sequencing processes can be integrated in a single device [12,13]. Research on these techniques is mainly taking place in universities. Sequencing by hybridization The basis of sequencing by hybridization is the microarray (DNA chip) technology, which also provides the basis for a large part of the gene expression analyses in drug development. In this technique, single-stranded sample DNA is hybridized onto a microtechnically produced array of DNA oligonucleotide probes. Each base in a sequence is scanned by moving the middle base in the oligonucleotide probe to all four possible positions (A, C, G and T), while leaving the rest of the sequence unchanged. The sequence of the DNA is determined according to which of the four oligo probes produces the strongest hybridization signal in each case. Large sequences can be analyzed by hybridization-based sequencing. The disadvantage, however, is that the read width is restricted to the length of the oligonucleotide probe (o25 bases). While the technique is suitable for investigating changes in individual nucleotides, more complex changes, for example the insertion or deletion of one (or more) codons, multiple point mutations in close proximity to each other or the insertion/deletion of fairly large sections of genetic material (such as complete ORFs), pose a challenge. Although the technique has been used for resequencing and de novo

3 sequencing operations [14–19], its strengths tend to lie in the mass resequencing of a limited number of genomic positions. Single molecules detected in real time Two approaches currently exist in this area, although these are still far from the stage of implementation in a practical and marketable form. 1. Direct observation of nucleotide incorporation: Here the researcher more or less watches a genetically modified polymerase during synthesis of the second DNA strand. Various fluorescence markers identify the nucleotide. Difficulties have been encountered with this technique in obtaining sufficient signals from individual incorporation events faced with a background of labeled nucleotides and in detecting all nucleotide incorporation events. 2. Nanopore-based sequencing method: Here the DNA is observed during its transport through pores on the nanometer scale. While the nucleotides pass through the nanopores, the chemical and physical properties of the individual bases are converted into electrical signals [20,21] and then into the sequence – at least that is the theory. We can only speculate about the likelihood of this technology actually coming to fruition. Although DNA fragments have already been observed during transport through the pores, it has not yet been proved possible to sequence individual bases. However, intense research efforts are focusing on the further development of the nanopores and the detection systems [22,23]. Sequencing by synthesis Sequencing by synthesis methods can roughly be subdivided into two groups: methods that sequence clonally amplified templates and methods that sequence single molecule templates. In both cases, the DNA fragments to be sequenced are physically separated in an array and subjected to repeated cycles of reagent addition/enzymatic manipulation in order to produce the sequences. Sequencing by synthesis of clonally amplified templates Most cyclic array sequencing technologies are based on sequencing by synthesis, i.e., the gradual assembling of the sequence by a polymerase, associated with a detection mechanism. The method developed by the US company 454 Life Sciences was the first of these techniques to be launched on the market [25]. The Genome Sequencer 20 (GS20) system was developed on the basis of this method and is distributed by Roche Diagnostics. Roche and 454 are working together on the further development of this system.

4 The technique is based on the clonal amplification of single molecules on beads (microparticles), which are isolated in an emulsion, and the subsequent massively parallel sequencing of the amplified DNA on the beads placed in the wells of a PicoTiterPlateTM. The GS20 system achieves a sequencing output of at least 20 Mbp (200,000 fragments) per 4.5 h cycle, with an average read width of 100 bp. The detection system of the device is based on the conversion of pyrophosphate – released during the polymerase-catalyzed attachment of a nucleotide to the complementary strand – into light via an enzyme cascade. Figure 1 provides an overview of the process. In all the cyclic array methods of this group, the fragments to be sequenced are amplified [26], either after isolation of the molecules (by emulsification or by an acrylamide matrix [27–29]) or by labeling and subsequent separation of the molecules. Although these are not, strictly speaking, single molecule detection systems, they are suitable for the sequencing of single molecules from an originally complex mixture. Sequencing by synthesis from single molecules This sequencing approach currently under development is similar to that of the first group, but with the difference that a more sensitive detection system is needed to detect the incorporation of nucleotides into the growing strand. While all single molecule methods employ the principle of the gradual incorporation of fluorescent nucleotides, they differ in their signal detection systems and the detailed biochemical aspects. Outlook The role of sequencing in Life Science research will change dramatically over the next few years (Table 1). The GS20 system from Roche Diagnostics is the first commercially available new-generation device (Fig. 2). The attempts to obtain new high-throughput sequencing technologies are ultimately aimed at lowering costs sufficiently to allow personalized genome sequencing. The 1000 dollar genome is the eventual goal of these attempts. To this end, the hardware will need to be improved in the field of bioinformatics in order to keep pace with the huge data volumes and new applications in the future. Thanks to new-generation sequencing technologies, new applications are gradually opening up. This refers in particular to the medical research field, including drug target identification. The determination of disease-relevant exons within a population, the analysis of the development of viral quasispecies in patients in relation to the course over time and drug use, as well as the search for genome changes of pathogens in relation to drug resistance are just some of the possibilities. In most cases, however, doors will open up to applications which were due to technical and economical reasons previously unattainable.

5

Fig. 1. Schematic overview of the shotgun sequencing and assembling of a bacterial

genome with the Genome Sequencer 20 System: (a) DNA library preparation, (b) emPCR, and (c) Sequencing.

6

Table 1. Novel sequencing technologies at a glance. Technology

Microelectrophoretic sequencing

Sequencing by hybridization

Real-time detection of single molecules

Sequencing by synthesis

Advantages

Sequencing by electrophoresis is well established Long read widths Low error rate

Potentially lower throughput than other methods Potentially not as cost effective as other methods [4,7–13]

Suitable for the detection of single molecules in complex mixtures Suitable for re– and de novo sequencing Not yet commercially implemented

Suitable for the detection of single molecules in complex mixtures Suitable for re– and de novo sequencing Narrower read widths compared to electrophoretic methods

[20–23]

[6,24,26–32]

Disadvantages

References

Suitable for the resequencing of known point mutations

Narrower read widths compared to electrophoretic methods De novo sequencing slow [14–19]

7

Fig. 2. The Genome Sequencer 20 System.

In resequencing, the genome sequence of a closely related organism or strain (reference sequence) is used as template for the assembly of shotgun sequencing reads generated from the genome of interest. Often the goal is to identify sequence deviations amongst the two genomes. In de novo sequencing, a genome sequence is assembled without using a reference sequence as template. In this case either no closely related genome is available or structural changes in the genome of interest shall be elucidated. Comparison: New fast sequencing technique vs. established Sanger method In shotgun sequencing, the genome to be sequenced is first sheared into randomly overlapping shotgun fragments. These fragments are then sequenced. In a process similar to solving a jigsaw puzzle, sequence overlaps between the individual fragments are now identified by special computer programs which assemble the individual sections into a whole genome. Already available in the market: The Genome Sequencer 20 system For the first time since 1980, when Gilbert and Sanger received the Nobel Prize for chemistry for DNA-sequencing, a radically new approach is

8 available on the market. The cost-effective GS20 system from Roche Applied Science can perform sequencing runs up to 60 times faster than conventional commercially available platforms. A single preparation step, without extensive robots for colony picking and handling of the microtiter plates, is all that is needed to prepare a whole genome sequencing project. The automated DNA sequencing system is capable of carrying out and monitoring sequencing reactions in a massively parallel fashion. Within days, an individual researcher can for the first time prepare samples, sequencing reactions, generate sequence reads and assemble genome sequence data. The whole genome sequencing workflow from sample input to data output consists of DNA library preparation, emulsion-based clonal PCR amplification (emPCR), PicoTiterPlateTM device preparation and sequencing instrument run, and data analysis. The output of a single run is typically 20 106 nucleotides or more (for the 70 75 mm PicoTiterPlateTM device) at an average read length of 100 high-quality bases, and multiple runs can be pooled for off-line assembly/mapping. The final consensus sequence is output as a FASTA file, with an associated basecall quality score file.

Shotgun sequencing with the Genome Sequencer 20 system The approach developed by 454 Life Sciences for the shotgun sequencing of whole genomes is much less complex than the Sanger method (see Fig. 3) as it does not require either physical cloning or microtiter plate handling with robots. Each step is simplified and the throughput increased by several orders of magnitude. The DNA sample is first fragmented on a scale appropriate to the sequencing system. Each fragment is ligated to DNA adaptors carrying the primer sequences required for amplification and sequencing. The ligation products are then merged to produce a library of random fragments of single-stranded template DNA (sstDNA). After the fragments have been purified and quantified, the sstDNA library is immobilized on microparticles known as beads – with just one amplifiable sstDNA molecule per bead. The whole bead-bound library is emulsified with the amplification reagents in a water-in-oil mixture. Each bead is enclosed within its own micro-droplet. These droplets function like small reactors to amplify an individual (clonal) sstDNA fragment. The amplification causes each bead to be covered with millions of copies of a single clonal DNA fragment, and each bead has a different fragment. After amplification, the genome-carrying beads – each of which now bears millions of copies of ‘‘their’’ DNA – are each deposited by centrifugation onto a well in a special PicoTiterPlateTM device. The loaded plate is then placed in the GS20 instrument and sequentially coated with nucleotide solutions. If nucleotide binding occurs a light signal is produced by a chemical reaction.

9

Genomic DNA

DNA Library Preparation 1. DNA fragmentation 2. Adaptor ligation 3. Single-stranded template DNA (sstDNA) library isolation 4. sstDNA library quality assessment and quantitation

Input Genomic DNA and GS 20 DNA Library Prep Kit Output sstDNA library in 1.5-ml tube Time 4 hours

Emulsion-Based Clonal Amplification 1. Preparation of the amplification mix 2. sstDNA library capture 3. Emulsification 4. Amplification 5. Bead recovery 6. sstDNA library bead enrichment 7. Sequencing primer annealing

Input Output from above and GS 20 emPCR Kit Output Sequence ready beads in 1.5-ml tube Time 1 day

Sequencing 1. The pre-wash run 2. PicoTiterPlate preparation 3. The sequencing run

Input Output from above and GS 20 PicoTiterPlate Kit and GS 20 Sequencing Kit Output Sequence ready for assembly Time 6 hours

Fig. 3. Flow diagram for the shotgun sequencing and assembling of bacterial gen-

omes according to 454 Life Sciences. Inputs, outputs and time are stated for each step. (gDNA ¼ genomic DNA, GS ¼ Genome Sequencer).

These signals are captured by a CCD camera that simultaneously records the information from all wells in the plate. The onboard computer determines the sequence of the individual fragments from the series of images and assembles them to form the whole sequence of long-chain baseline DNA. The raw data consist of a set of digital images from which the sequence is determined for each fragment and then assembled into a consensus sequence. The reads can be mapped against a scaffold sequence, assembled de novo or output as individual reads for subsequent processing as FASTA files. The traditional assembler programs were developed for Sanger sequencing runs. In view of the read width of 100 bases and the resulting large number of

10 individual reads, these are not directly suitable for the GS20 system. Roche offers a de novo assembler software, called Newbler assembler, specially adapted to the system’s reads. Table 2 shows the output of the 454 sequencing system in respect of coverage and accuracy through de novo sequencing of known genomes. The bacterial results presented in the table were obtained from clones whose sequence is available in GenBank. The results show that the 454 sequencing system achieves a high degree of coverage for all genomes and substantial concordance with the published genomes. Breaks in the scaffold are the result of incomplete coverage of the genome due to random chance and repeat regions that are longer than the raw sequence reads and thus cannot be consistently recorded. The total accuracy of the consensus sequence is at least equal, if not better compared to those obtained during Sanger sequencing efforts. The same refers to the number of missassemblies. Sequencing and assembling microbial genomes by the Sanger method The shotgun sequencing method according to Sanger is currently the most popular sequencing technology [31]. Some 280 microbial genomes have been completely sequenced to date with this electrophoresis-based technique [32]. The time taken to sequence and assemble a microbial genome with the Sanger method depends greatly on the resources of the implementing organization. Thanks to their sophisticated technical equipment, large genome centers are capable of sequencing a genome and assembling an initial draft in around two weeks. An individual laboratory would take several months, or even years, to complete the same task. In view of its limitations in terms of cost and capacity (throughput), the Sanger sequencing method is only used to a very limited extent [33]. Figure 4 shows a flow diagram for the shotgun sequencing of complete genomes according to Sanger. First, libraries are generated from randomly fragmented genomic DNA. After fragmentation and size selection, the libraries are produced by ligating the DNA fragments into plasmid (for small inserts) or cosmid/fosmid (for large inserts) vectors. These plasmids, cosmids or fosmids are transformed into bacteria. The resulting bacterial colonies are transferred (by an automatic colony picker) to 96-well or 384-well plates and incubated in a liquid growth medium in order to culture bacterial clones containing the plasmids. During this phase the libraries hopefully contain a comprehensive, random distribution of genomic DNA. In the next step, sufficient templates must be produced to enable the sequencing chemistry to be implemented. For the most part, plasmids are isolated and purified from each individual bacterial clone. For sequencing, a standard primer is hybridized to a single-stranded template and the DNA molecule is extended by a DNA polymerase. The DNA chain is terminated by a specific dideoxyribonucleotide triphosphate labeled

Table 2. The output of the Newbler assembler developed by 454 Life Sciences for various bacterial genomes (genome sequences publicly available in GenBank). The total accuracy of consensus sequences is striking. Strain Genome size (bp) Coverage depth (Xfold) Number of contigs Average contig size (kb) Size of largest contig (kb) Total genomic coverage (%) Total genomic coverage of nonrepeat regions (%) Total accuracy of consensus sequence (%) Number of runs Misassemblies

M.genitalium

B. licheniformis

E. coli

S. pneumoniae

580,069 20.68

4,222,645 21.98

4,639,675 23.5

2,076,278 22.37

8,641,205 20.5

12,070,820 25

20 28,008

136 30,657

139 32,603

229 8,802

1,013 8,383

717 15,817

154,741

200,162

163,595

59,578

70,534

98,656

96.57%

98.63%

97.45%

92.99%

96.96%

92.86%

99.42%

100.00%

100.00%

100.00%

99.9940% 0.3 0

99.9970% 2.5 2

99.9990% 3 3

99.9950% 1.5 4

S. coelicolor

97.9097%

S. cerevisae

100.00%

99.9920%

99.9780%

6 36

11 16

11

12

Genomic DNA

DNA Library Preparation 1. DNA fragmentation − generate large (20 kb), and small (3 kb) fragment libraries 2. Insert libraries into suitable vector and transform into bacteria 3. Plate bacteria onto solid growth media (Petri dishes) 4. After incubation, pick colonies into 96 (or 384)-well plates

Template Purification 1. Plasmid preparation a. Many different methods of template preparation are available. Plasmid preparation is a standard.

Input Genomic DNA, cloning vector, bacteria, Petri Dishes, 96 (or 384)-well plates, growth media Output Bacterial clones in hundreds of 96-well plates Time 1 week with full automation

Input Output from above and plasmid preparation kits, 96 (or 384)-well plates Output Purified plasmid in hundreds of 96 (or 384)-well plates Time 1 day with full automation

Sequencing 1. Set-up sequencing chemistry 2. Thermal-cycle sequencing chemistry reactions 3. Hundreds of sequencing runs 4. Generation of paired-end reads

Input Output from above and sequencing chemistry kits, 96 (or 384)-well plates Output Sequence ready for assembly Time Week to months depending on number of thermal cyclers and sequencers and level of automation

Fig. 4. Flow diagram for the shotgun sequencing and assembling of bacterial gen-

omes according to Sanger. Inputs, outputs and time are stated for each step.

with fluorescent dye. The mixture of DNA fragments of differing length is electrophoretically separated into capillaries according to their length. The fluorescent signal detected by a laser identifies the specific nucleotide that is labeled at the end of a DNA fragment of a precise length. Comparison of the shotgun methods developed by 454 and Sanger Compared to the Sanger method, the sequencing with the GS20 system is much faster and more cost-effective. Moreover, the genomes resulting from

13 the new method are much more comprehensive in many cases, because the Sanger method frequently involves the systematic error arising from nonclonable sections, and because the new technique offers good coverage of regions that are traditionally difficult to sequence (e.g., full stops). Due to these advantages, sequencing with the GS20 system not only allows rapid investigation of individual genomes, but for the first time also comparative studies with many genomes in a reasonable period of time. Applications of the Genome Sequencer 20 system Whole genome sequencing The GS20 system already has revolutionized whole genome sequencing of microorganisms. For instance, sequencing of a 3 megabase bacterial genome up to a high-quality draft is now possible within days, rather than months. Due to the fact that high-quality reads at an average read length of 100 bases are generated, both de novo assembly as well as resequencing (mapping) of genomes is possible. The Mapping application generates the consensus DNA sequence by mapping, or alignment, of the reads to a reference sequence; as well as a list of high-confidence mutations. The current version of the GS20 software has the capacity of analyzing genomes up to 50 Mbp in size at 15–25X depth of coverage. The Mapping application will typically result in X99.99% accuracy over 95% of the non-repeat parts of the genome (Q40+bases), when the average genome coverage is at least 15X. The Assembler application will yield N50 contig sizes X10 kb with X99.99% accuracy over 95% of the nonrepeat parts of the genome (Q40+bases), when the average genome coverage is at least 25X. Examples of several bacterial genome assemblies are shown in Table 1 (N50 contig size: The size of a contig above and including that half of the nucleotides are assembled). Since the GS20 system utilizes neither cloning in bacteria nor electrophoretic separation, sequence coverage biases normally associated with these techniques are eliminated. The lack of sequence coverage bias had been confirmed by sequencing several tens of bacterial genomes. The remaining gaps in assembled genome sequences are largely due to the presence of sequence repeats longer than 75 bp. This means that the GS20 system is particularly useful in sequencing of AT rich organisms resistant to subcloning in E. coli. One example is sequencing of the filamentous fungus Neurospora crassa. By using the 454 sequencing technique, 2.5% additional sequence information compared to a Sanger sequencing approach has been identified. Not surprisingly, the GC content of this additional information was quite low (27%; http://www.genome-sequencing.com). Recently 454 Life Sciences has developed a new protocol that makes whole genome sequencing on the GS20 system even more efficient. Paired-end

14

Intact gDNA

1) Fragment 2) EcoR1 Methylase

Bio

Met

Met

1) EcoR1 2 2) Ligation

Bio

3) Polish 4) Adaptor ligation Approx 2.5 Kb Fragments

Bio

1) MmeI Bio

Two 20 base tags that were ~2.5 kb apart

2) Isolate with SA-beads Bio

Bio Bio

Amplification A

Bio

B

A

B

Fig. 5. Paired-end library preparation scheme: genomic DNA is fragmented to yield average fragment size around 2.5 kb. The fragmented genomic DNA is methylated with EcoR I methylase to protect the EcoR I restriction sites. The ends of the fragments are blunt-ended, polished and an oligonucleotide adaptor is blunt-end ligated onto both ends of the digested DNA fragments. Subsequent digestion with EcoR I restriction enzyme cleaves a portion of the adaptor DNA, leaving sticky ends. The fragments are circularized and ligated resulting in 2.5 kb circular fragments. The adaptor DNA contains two Mme I restriction sites and after treatment with Mme I the circularized DNA is cleaved 20 nucleotides away from the restriction sites in the adaptor DNA. This digestion generates small DNA fragments that have the adaptor DNA in middle and 20 nucleotides of genomic DNA that were once approximately 2.5 Kb apart on each end. These small, biotinylated DNA fragments are purified from the rest of the genomic DNA by streptavidin beads.

libraries are generated and sequenced in order to determine the orientation and relative positions of contigs produced by the de novo shotgun sequencing and assembly (see Fig. 5 for an outline of the library generation process). Sequence data obtained from the paired-end libraries are combined with standard GS20 whole genome shotgun sequencing reads in a new version of the Assembler. The benefits of combining the GS20 shotgun sequence reads with the paired-end reads have been tested on several bacterial genomes and a Saccharomyces cerevisiae genome previously sequenced at 454 Life Sciences. For instance, the 4.6 Mbp genome of E. coli K12 strain was sequenced in three standard GS20 runs to a depth of 22-fold. The assembly performed with the Newbler assembly software resulted in 140 un-oriented contigs. One additional sequencing run of a paired-end library yielded approximately 112,000 reads. The paired-end data improved the genome assembly to 20 multi-contig scaffolds covering 98.6% of the genome. An illustration of this result is shown in Fig. 6. The 12.2 Mbp genome of S. cerevisiae S288C (16 haploid

15

Fig. 6. De novo assembly results for E. coli K12 aligned against a reference genome.

The reference genome is represented by the top black line. The standard whole genome shotgun sequence and assembly is represented by the light grey bars. Repeat regions of the genome are represented by the dark grey bars at the bottom. Spaces between the light grey bars are typically a result of repeat regions that cannot be uniquely assigned to a region in the genome. With the addition of one run of pairedend reads, represented by the grey bars, the genome sequence becomes much more complete.

chromosomes and one 86 kbp mitochondrion) was shotgun sequenced in nine sequencing runs yielding approximately 23X over sampling. The assembly performed with the Newbler assembler resulted in 821 un-oriented contigs. Two additional sequencing runs of a paired-end library yielded approximately 395,000 reads. The paired-end data reduced the assembly to 153 scaffolds covering 93.2% of the genome. Large-scale sequencing of mammoth DNA In a recent publication [34], Poinar et al. described sequencing of 28 million base pairs (bp) of DNA derived from a woolly mammoth sample from Siberia, of which 13 million bp (45.4%) were identified as mammoth DNA. The amount of endogenous DNA available from this single animal would allow for completion of its genome, revolutionizing the field of paleogenetics. Poinar et al. used the GS20 system for their work. Sequencing DNA of extinct species could provide valuable insights into the course of evolution. Traditional PCR-based sequencing approaches, however, are limited by the poor preservation of fossil DNA. Fossil DNA is often fragmented to less than 300 bp due to degradation [35–37], impeding Polymerase-based PCR amplification. Contaminations with bacterial and/or fungal DNA are frequent. Therefore, sequencing DNA of extinct species was

16 mammoth (alignable to elephant) 45.43% mammoth (predigted to align) 9.09% alignable to human 1.40% alignable to dog 1.25% bacteria 5.76% archaea 0.24% other Eukaryota 4.15% virus 0.09% environmental sequences 14.15% unidentified sequences 18.42%

Fig. 7. Characterization of the mammoth metagenomic library. Percentage of read

distributions to various taxa. Host organism prediction was based on BLAST comparison against GenBank and environmental sequences database [34].

restricted to the mitochondrial genome, which exists in higher copy numbers. In addition, sequencing of nuclear DNA requires known sequences from specific genes as targets. The sequencing approach utilizing the GS20 system provides an alternative, and for the first time enabled large-scale sequencing of nuclear mammoth DNA. The 454 sequencing approach yielded 302,692 sequence reads with an average length of 95 bp, resulting in 28 106 bp of data transferred to the NCBI Trace Archive (SID131303). Of these, 45.4% aligned to the genome of the African elephant with an identity of 490%. This indicates that sequences were derived from mammoth DNA rather than from external contamination. Sequence identity between mammoth and elephant in reads aligning to a single position of the elephant genome was 98.55% (result not corrected for sequence damage due to base degradation) (Fig. 7; Table 3). Sequencing by using the GS20 system enabled the most comprehensive sequencing of DNA derived from a fossil sample of an extinct species so far. The special design of the sequencing process, relying on small fragments of DNA and adaptor-sequences rather than on known target sequences, render it an ideal approach for the study of less-explored genomes in general. Amplification by means of standard adaptor sequences prior to sequencing avoids cloning or amplification biases. The original template distribution is maintained, which may improve genome coverage. In addition, the process results in shorter, overlapping reads compared with the traditional approach. This may serve as a correction filter for DNA-degradation-based reading mistakes. Applying the new methodology, researchers were able to create the first large-scale sequencing of the nuclear genome of the mammoth to an extent that enables the detailed analysis of functional genes and a fine-scale refinement of mutation rates. Valuable insights into the reaction of Pleistocene mammals to climatic changes may soon be expected.

17 Table 3. Total percent of aligned reads and their relative identities to African elephant, human and dog (modified from Poinar et al. 2006).

Total number of reads Aligned reads Uniquely aligned reads Multiply aligned reads Reads with at least 95% identity Reads with 100% identity Uniquely aligning base pairs Identity in unique alignments Mitochondrial reads Identity in mitochondrial reads Mitochondrial base pairs

Elephant

Human

Dog

302,692 (100%)

302,692 (100%)

302,692 (100%)

137,527 (45.4%) 44,442 (14.7%)

4,237 (1.4%) 3,901 (1.3%)

3,775 (1.2%) 3,548(1.2%)

93,085 (30.8%)

336 (0.1%)

227 (0.1%)

90,507 (30.0%)

1,184 (0.4%)

1,140 (0.4%)

21,952 (7.3%)

116 (0.04%)

142 (0.05%)

4,332,350

318,966

291,714

98.55%

92.68%

92.91%

209

–

–

95.93%

–

–

16,419

–

–

Ultrafast sequencing boosts target identification in diarylquinoline drug research After AIDS, tuberculosis (TB) is the leading cause of infectious disease mortality in the world, with 2–3 million deaths per year. The TB and HIV epidemics fuel one another in coinfected people, and at least 11 million adults are infected with both pathogens. Consequently, one of the factors contributing to the TB burden is the recent increase in the number of HIV-infected individuals. Although current first-line anti-TB drug regimens can achieve more than 99% efficacy, this is often reduced because of drug resistance. New drugs that could shorten or simplify effective treatment of TB would substantially improve TB control programs. In a recent publication [38], K. Andries et al. reported on the antimycobacterial properties of the diarylquinolines (DARQs). The lead compound, R207910, not only has several properties that may improve the treatment of TB, but also appears to act at a new target, providing an antimycobacterial spectrum different from those of current drugs.

18 Andries et al. used the GS20 system to select mutants of M. tuberculosis and M. smegmatis resistant to R207910 in vitro, for assessment of the crossresistance pattern, and to investigate the mechanism of action. By comparison with rifampin, they quantified the proportion of resistant mutants arising. The results indicate that the target and mechanism of action of R207910 is different from those of other anti-TB agents. Inhibition of ATP synthase function may lead to ATP depletion and imbalance in pH homeostasis, both contributing to decreased survival. In comparing the ATP synthases sequences of different bacteria and of eukaryotic ATP synthase, the researchers found a rationale for the specificity of the antibacterial spectrum, and – to a lesser extent – the safety profile of R207910. A further finding was that the distinct target of R207910 indicates the lack of cross-resistance with existing anti-TB drugs. The studies verified that R207910 is as effective against MDR strains as it is against fully antibiotic-susceptible strains.

Amplicon analysis As mentioned above, the GS20 system is based on clonal, single DNA fragment molecule amplification in combination with a high-throughput sequencing chemistry. The characteristics of resulting sequence reads, currently on average 100 bases long, but tens-of-thousand-fold deep, open a unique opportunity to employ the Genome sequencer in applications where detection of rare variants of a known sequence in complex mixtures of sequences is crucial. Direct sequencing of mixed, non-clonal PCR products (amplicons) using Sanger dideoxy terminator chemistry is not sensitive enough to identify and quantify many of the sequence variants present in biological specimens. Bacterial cloning of amplicons into a vector prior to traditional sequencing of individual clones will increase the sensitivity, but not without a large increase in time and cost, thus making this approach non-economical. The 454 technology provides ‘‘instant cloning’’ of hundreds-of-thousands of molecules via the emulsion PCR step and highly accurate sequencing, since now each fragment can be sequenced hundreds-of-thousand-fold deep. Although there are many potential uses for amplicon sequencing, the molecular biology and software developments at 454 Life Sciences have initially focused on the oncology research applications, more specifically on the detection of rare somatic mutations in complex cancer samples. The early detection of drug resistance causing mutations in tumor cells (medical research on the GS20 system) may be of paramount importance for design of future diagnostic assays. Moreover, the ability to sensitively detect somatic mutations in cancer cells will help to understand the development of cancer on the genetic level in much greater detail (time series). Additionally, none of the existing high-throughput technologies, based on hybridization on microarrays or bead arrays, offer the possibility of novel variant detection.

19 DE15

51% C / 49% A

DD14

51% T / 49% C

GA9

81% T / 19% G

Fig. 8. Genotyping results of three SNPs in the HLA-DMA gene region (Class II

MHC). Base changes along the fragment sequence (x-axis) and their positions are shown as bars. The primary y-axis denotes base change frequency, the secondary yaxis as well as the black line above the mutation spectrogram represents sequencing coverage. Both high-frequency alleles (top panel) and low-frequency alleles (bottom panel) are shown.

To demonstrate the power of the GS20 system, previously described single nucleotide polymorphisms from upstream of the HLA-DMA gene to the TAP2 gene in the Class II region of the MHC were chosen as a model system. The researchers were able to reproduce the published data using the GS20 system, allele frequencies down to 3% were easily detected, as shown in Fig. 8.

Ultra-deep sequencing of EGFR in lung cancer reveals low abundance drug response mutations A current challenge in oncology is the detection of cancer causing and chemotherapy responsive mutations in clinical samples that may contain only a small sub-population of tumor-derived cells. Thomas et al. [39] reported high-sensitivity mutation detection using high-throughput DNA sequencing with the GS20 system. The researchers demonstrated the utility of the technology by studying mutations in the epidermal growth factor receptor (EGFR), associated with tumorigenesis and response to tyrosine kinase inhibitors in non-small cell lung cancer (NSCLC). Furthermore, they examined the

20 mutational status of exons 18–22 of EGFR in over 20 individuals suffering from NSCLC. The data revealed low-abundance mutations, including base-pair substitutions, deletions and insertions associated with disease and drug-response, which were – due to a lack of sensitivity caused by mainly economical reasons – completely missed using Sanger sequencing. In a pre-therapy pleural effusion sample from individuals with strong initial drug response, a drug-sensitizing 18 bp deletion was detected at 0.28%. Sequencing of a sample taken after relapse detected the deletion at 3% and, consistent with the clinically observed loss of drug response, also revealed a drug-desensitizing mutation, T790M, at 2.0%. The data demonstrate fast and accurate identification of cancer-associated mutations from complex samples at a sensitivity and speed that is unprecedented. According to Thomas et al., this rapid mutation detection could aid in critical decision making in the course of patient treatment. In summary, Thomas et al. reached the following conclusions: Low-level nucleotide variants, including substitutions, deletions and

insertions, can be detected and quantified down to the 0.1% frequency level, Subspecies are identified and quantified from within complex samples allowing for haplotype determination, Multiplexing is possible on the amplicon level, making the simultaneous interrogation of many target regions, including entire coding regions of specific genes and exons of whole gene families, Applied to tumor samples the technology has the capability to detect known and novel mutations at levels not readily detected with traditional sequencing approaches.

Transcriptome and gene regulation studies The GS20 system enables the study of transcriptomes at a previously impossible depth of coverage and sensitivity. This is due to the system’s massively parallel sequencing technology, which generates a high number of sequence reads (minimum of 200,000 single reads per 5 h run), facilitating the identification of previously unknown transcripts (see ‘‘Studying the role of the Arabidopsis thaliana DICER-like genes in small RNA processing’’). First results gained within the framework of a short tag sequencing project also revealed that the GS20 system is also very well suited for reliable expression profiling studies (data not shown). In terms of gene regulation, the 454 technology so far has been shown to be perfectly suited for the genome wide identification of small non-coding RNAs (sncRNAs), for the identification of transcription factor binding sites

21 or the elucidation of DNA-methylation patterns. Compared to sequencing of sncRNAs using the Sanger-approach, during which miRNA fragments are concatemerized in order to make sequencing more economical, the GS20 approach is much straightforward because the often ‘‘tricky’’ concatemerization step simply can be skipped. Moreover, costs per clonal read are much lower on the GS20 system, thus providing a real basis for screening for scnRNA on a genome wide level. In the following chapters please find some application examples on transcriptome and gene regulation studies. Identification of binding sites of DNA-binding proteins The identification of binding sites of DNA-binding proteins such as those of the transcription factor p53 on the GS20 system has recently been shown by Ng et al. [40]. DNA fragments that include binding-site sequences can be isolated after immunoprecipitation with their protecting transcription factors and characterized using high-throughput sequencing. This study revealed that using the GS20 system transcription factor binding sites can be identified with much higher efficiency compared to Sanger sequencing at a much better accuracy. DNA-Methylation studies Loss of methylation as well as hypermethylation of CpG islands within promoter regions is known to be a very important gene regulation mechanism of many genes. Genome methylation occurs at cytosine residues located 50 to a guanosine in a CpG dinucleotide. Dense areas of CpG dinucleotides within promoter regions are organized into CpG islands. Applying the known bisulfite treatment procedure, 454 recently has established a sequencing-based technology to quantitatively characterize the methylation state of each CpG dinucleotide in a given target genomic sequence. This is based on 454’s amplicon kits and the corresponding amplicon variant software. To understand how the chemistry will perform on clinical research samples, eight colorectal cancer (CRC) tumor samples and their matched normal adjacent tissue (NAT) were analyzed (Fig. 9), and the frequency of methylation was determined. The results obtained [42] are supported by the published literature that a significant percentage of CRCs show methylation of the p16 CpG island [41]. This proves that the described GS20 application, which is based on an open system without bias for what CpG dinucleotides leads is a straightforward technique to obtain accurate data regarding the quantification of methylation at each CpG dinucleotide addressed. It will be this data that can be analyzed by powerful statistical tools to determine what residues are important in the development and maintenance of cancer.

CRC methylation

Freq CpG Methylation

1

0.8

22

1.2

1 2 3 4 5 6 7 8

0.6

0.4

0.2

0 41 46 48 52 56 59 65 87 91 94 112 115 118 139 157 161 163 168 174 189 193 205 211 221 226 236 240 247

CpG Position Fig. 9. CpG Methylation Frequency in the p16 CpG island in CRC samples. This figure represents the methylation frequencies for each of the CpG dinucleotides assayed in the p16 promoter region in eight independent CRC tumor samples. Frequency of methylation is frequency of C to T conversion subtracted from 1.0 which would represent 100% conversion of C to T when there is no protection of a C from the bisulfite treatment. Some positions (e.g., 139) show a nearly complete methylation of that CpG in all the samples while others shown almost no methylation (e.g., position 59). The variability in methylation between the eight samples and from one CpG dinucleotide to another is striking. See position 221 through 247.

23 Genome wide identification of of piRNAs (small non-coding RNAs) Girad et al. [43] used the system in order to characterize a new class of small RNAs, called Piwi-interacting RNAs (piRNAs), in mouse testis. More than 87,000 reads were generated, around 53,000 of which would be classified as candidate piRNAs. Girard et al. found that a germline-specific class of small RNAs binds mammalian Piwi proteins. Small RNAs associate with Argonaute proteins and serve as sequence-specific guides to regulate messenger RNA stability, protein synthesis, chromatin organization and genome structure. In animals, Argonaute proteins segregate into two subfamilies. The Argonaute subfamily acts in RNA interference and in microRNA-mediated gene regulation using 21–22-nucleotide RNAs as guides. The Piwi subfamily is involved in germline-specific events such as germline stem cell maintenance and meiosis. However, neither the biochemical function of Piwi proteins nor the nature of their small RNA guides is known. The authors showed that MIWI, a murine Piwi protein, binds a previously uncharacterized class of 29–30-nucleotide RNAs that are highly abundant in testes. They therefore named these piRNAs. piRNAs show distinctive localization patterns in the genome, being predominantly grouped into 20–90 kb clusters, wherein long stretches of small RNAs are derived from only one strand. Similar piRNAs are also found in human and rat, with major clusters occurring in syntenic locations. Although their function must still be resolved, the abundance of piRNAs in germline cells and the male sterility of Miwi mutants suggest a role in gametogenesis. Studying the role of the Arabidopsis thaliana DICER-like genes in small RNA processing Henderson et al. studied the role of the four Arabidopsis thaliana DICERlike genes (DCL1–DCL4) in small RNA processing, gene silencing and DNA-methylation patterning [44]. By sequencing of RNAs from a dcl2 dcl3 dcl4 triple mutant, the researchers found markedly reduced tasiRNA and siRNA production. Their conclusion was that DCL1, which is also the major enzyme for processing miRNAs, has a previously unknown role in the production of small RNAs from endogenous-inverted repeats. DCL2, DCL3 and DCL4 were functionaly redundant in production of siRNA and tasiRNA and in the establishment and maintenance of DNA methylation. The studies also suggest that asymmetric DNA methylation can be maintained by pathways that do not require siRNAs. Searching mammalian factors for transcriptional gene silencing During a search for candidate mammalian factors for transcriptional gene silencing (TGS), Nelson C. Lau et al. purified a complex that contains small RNAs and Riwi, the rat homolog to human Piwi, from rat testes [45]. The RNAs, frequently 29–30 nucleotides in length, are called piRNAs, 94% of

24 which map to 100 defined (o100 kb) genomic regions. Within these regions, the piRNAs generally distribute across only one genomic strand, or distribute on two strands but in a divergent, non-overlapping manner. Preparations of piRNA complex (piRC) contain rRecQ1, which is homologous to qde-3 from Neurospora, a gene implicated in silencing pathways. Piwi has been genetically linked to TGS in flies and slicer activity cofractionates with the purified complex. These results are consistent with a gene silencing role for piRC in mammals. References 1. 2. 3.

4.

5. 6. 7. 8. 9. 10. 11. 12. 13.

14. 15.

16.

Hinds DA, et al. Whole-genome patterns of common DNA variation in three human populations. Science 2005;307:1072–1079. Hardenbol P, et al. Highly multiplexed molecular inversion probe genotyping: over 10,000 targeted SNPs genotyped in a single tube array. Genome Res. 2005;15:269–275. Gerhardt M, et al. In-depth, longitudinal analysis of viral quasispecies from an individual triply infected with late-stage human immunodeficiency virus type 1, using a multiple PCR primer approach. J Virol 2005;79:8249–8261. Kapoor A, et al. Sequencing-based detection of low-frequency human immunodeficiency virus type 1 drug-resistant mutants by an RNA/DNA heteroduplex generator-tracking assay. J Virol 2004;78:7112–7123. Kwak EL, et al. Irreversible inhibitors of the EGF receptor may circumvent acquired resistance to gefitinib. Proc Natl Acad Sci USA 2005;102:7665–7670. Andreis K, et al. A diarylquinoline drug active on the ATP synthase of Mycobacterium tuberculosis. Science 2005;307:223–227. Emrich CA, et al. Microfabricated 384-lane capillary array electrophoresis bioanalyzer for ultrahigh-throughput genetic analysis. Anal Chem 2002;74:5076–5083. Koutny L, et al. Eight hundred-base sequencing in a microfabricated electrophoresis device. Anal Chem 2000;72:3388–3391. Aborn JH, et al. A 768-lane microfabricated system for highthroughput DNA sequencing. Lab Chip 2005;5:669–674. Lagally ET and Mathies RA. Integrated genetic analysis microsystems. J Phys D: Appl Phys 2004;37:R245–R261. Paegel BM, et al. Microfluidic devices for DNA sequencing: sample preparation and electrophoretic analysis. Curr Opin Biotechnol 2003;14:42–50. Zwick ME, et al. Microarray-based resequencing of multiple Bacillus anthracis isolates. Genome Biol 2004;6:R10. Sougakoff W, et al. Use of a high-density DNA probe array for detection mutations involved in rifampicin resistance in Mycobacterium tuberculosis. Clin Microbiol Infect 2004;10:289–294. Miatra A, et al. The human MitoChip: a high-throughput sequencing microarray for mitochondrial mutation detection. Genome Res 2004;14:812–819. Gonzalez R, et al. Detection of human immunodeficiency virus type 1 antiretroviral resistance mutations by high-density DNA probe arrays. Clin Microbiol 2004;42:2907–2912. Read TD, et al. Comparative genome sequencing for discovery of novel polymorphisms in Bacillus anthracis. Science 2002;296:2028–2033.

25 17. Poly F, et al. Genomic diversity in Campylobacter jejuni: identification of C. jejuni 81-176-specific genes. J Clin Microbiol 2005;43:2330–2338. 18. Deamer DW and Akeson M. Nanopores and nucleic acids: prospects for ultrarapid sequencing. Trends Biotechnol 2000;18:147–151. 19. Meller A, et al. Dynamics of polynucleotide transport through nanometer pores. J Phys Condens Matter 2003;15:R581–R607. 20. Chen P, et al. Atomic layer deposition to fine-tune the surface properties and diameters of fabricated nanopores. NanoLett 2004;4:1333–1337. 21. Karhanek M, et al. Single molecule detection using nanopipettes and nanoparticles. NanoLett 2004;5:403–407. 22. Brenner S, et al. Gene expression analysis by massively parallel signature sequencing (MPSS) on microbead arrays. Nat Biotechnol 2000;18:630–634. 23. Margulies M et al. Genome sequencing in open microfabricated high density picoliter reactors. Nature 2005;10.1038/nature03959. 24. Dressman D, et al. Transforming DNA molecules into fluorescent magnetic particles for detection and enumeration of genetic variations. Proc Natl Acad Sci USA 2003; 100:8817–8822. 25. Mitra RD and Church GM. In situ localized amplification and contact replication of many individual DNA molecules. Nucleic Acids Res 1999;27:e34. 26. Mitra RD, et al. Digital genotyping and haplotyping with polymerase colonies. Proc Natl Acad Sci 2003;100:5926–5931. 27. Mitra RD, et al. Fluorescent in situ sequencing on polymerase colonies. Anal Biochem 2003;320:55–65. 28. Levene MJ, et al. Zero-mode waveguides for single-molecule analysis at high concentrations. Science 2003;299:682–685. 29. Braslavsky I, et al. Sequence information can be obtained from single DNA molecules. Proc Natl Acad Sci 2003;100:3960–3964. 30. Simons J et al. Ultra-deep sequencing of HIV from drug resistant patients. XIV International HIV Drug Resistance Workshop, Quebec City, Canada, June 7–11, 2005. 31. http://www.jgi.doe.gov/education/how/index.html. 32. http://www.genomesonline.org/. 33. Read TD, et al. Comparative genome sequencing for discovery of novel polymorphisms in Bacillus anthracis. Science 2002;296:2028–2033. 34. Poinar HN et al. Metagenomics to paleogenomics: large-cale sequencing of mammoth DNA. Science 2006;311:392–439. 35. Paäbo S, et al. Genetic analyses from ancient DNA. Annu Rev Genet 2004;38:645–679. 36. Poinar HN, et al. Molecular coproscopy: Dung and diet of the extinct ground sloth nothrotheriops shastensis. Science 1998;281:02–06. 37. Hoess et al. DNA damage and DNA sequence retrieval from ancient tissues. Nucleic Acids Res 1996;24(7):1304–1307. 38. Andries K, et al. A diarylquinoline drug active. On the ATP synthase of Mycobacterium tuberculosis. Science 2005;307:223–227. 39. Thomas, et al. Sensitive mutation detection in heterogeneous cancer specimens by massively parallel picoliter reactor sequencing. Nat Med 2006;12(7):852–855. 40. Ng, et al. Multiplex sequencing of paired-end ditags (MS-PET): A strategy for the ultrahigh-throughput analysis of transcriptomes and genomes. Nucleic Acids Res 2006;34(12). 41. Kim, et al. Methylation and expression of p16INK4 tumor suppressor gene in primary colorectal cancer tissues. Int J Oncol 2001;26:1217–1226.

26 42. Herman, et al. Methylation-specific PCR: A novel PCR assay for methylation status of CpG islands. Proc Natl Acad Sci 1996;93:9821–9826. 43. Girard A, Ravi Sachidanandam R, Hannon GJ and Carmell MA. A germline-specific class of small RNAs binds mammalian Piwi proteins. Nature 2006;4 (June). Advance online publication. 44. Henderson IR, Zhang X, Lu C, Johnson L, Meyers BC, Green PJ and Jacobsen SE. Dissecting Arabidopsis thaliana DICER function in small RNA processing, gene silencing and DNA methylation patterning. Nat Genet 2006;38:721–725. 45. Lau NC, Seto AG, Kim J, Kuramochi-Miyagawa S, Nakano T, Bartel DP and Kingston RE. Characterization of the piRNA complex from rat testes. Science 2006;15 (June). Advance online publication.

27

Optimizing scaleup yield for protein production: Computationally Optimized DNA Assembly (CODA) and Translation EngineeringTM G. Wesley Hatfield1,2 and David A. Roth2, 1

The Institute for Genomics and Bioinformatics, Donald Bren School of Information and Computer Sciences, University of California, Irvine, CA 92497, USA 2 CODA Genomics Inc., 26061 Merit Circle #101, Laguna Hills, CA 92653-7015, USA Abstract. Translation EngineeringTM combined with synthetic biology (gene synthesis) techniques makes it possible to deliberately alter the presumed translation kinetics of genes without altering the amino acid sequence. Here, we describe proprietary technologies that design and assemble synthetic genes for high expression and enhanced protein production, and offers new insights and methodologies for affecting protein structure and function. We have patented Translation EngineeringTM technologies to manage the complexity of gene design to account for codon pair usage, translational pausing signals, RNA secondary structure and user-defined sequences such as restriction sites. Failure to optimize for codon pair-encoded translation pauses often results in the relatively common occurrence of a slowly translated codon pair that slows the rate of protein elongation and decreases total protein production. Translation Engineering TM technology improves heterologous expression by tuning the gene sequence for translation in any well-characterized host, including cell-free expression techniques characterized by ‘‘broken’’ Escherichia coli systems used in kits for today’s molecular tools market. In addition, we have patented a novel gene assembly method (Computationally Optimized DNA Assembly; CODA) that uses the degeneracy of the genetic code to design oligonucleotides with thermodynamic properties for self-assembly into a single, linear DNA product. Fast translational kinetics and robust protein expression are optimized in synthetic ‘‘Hot RodTM’’ genes that are guaranteed to express in E. coli at high levels. These genes are optimized for codon usage and other properties known to aid protein expression, and importantly, they are engineered to be devoid of mRNA secondary structures that might impede transcription, and over-represented codon pairs that might impede translation. Hot Rod genes allow translating ribosomes and E. coli RNA polymerases to maintain coupled translation and transcription at maximal rates. As a result, the nascent mRNA is produced at a high level and is sequestered in polysomes where it is protected from degradation, even further enhancing protein production. In this review we demonstrate that codon context can profoundly influence translation kinetics, and that over-represented codon pairs are often present at protein domain boundaries and appear to control independent protein folding in several popular expression systems. Finally, we consider that over-represented codon pairs (pause sites) may be essential to solving problems of protein expression, solubility, folding and activity encountered when genes are introduced into heterologous expression systems, where the specific set of codon pairs controlling ribosome pausing are different. Thus, Translation Engineering TM combined with synthetic biology (gene synthesis) techniques may allow us to manipulate the translation kinetics of genes to restore or enhance function in a variety of traditional and novel expression systems.

Corresponding author. Tel: direct (949) 305-5296; cell (949) 228-0986.

E-mail: [email protected] (D.A. Roth). URL: http://www.codagenomics.com (D.A. Roth). BIOTECHNOLOGY ANNUAL REVIEW VOLUME 13 ISSN 1387-2656 DOI: 10.1016/S1387-2656(07)13002-7


28 Keywords: codon context, synthetic biology, translational step-times, codon pair utilization bias, protein expression, translational kinetics, protein domains, Translation Engineering, SpeedPlots, Hot Rod genes, Planned Pause genes, fusion proteins, monoclonal antibodies, protein therapeutics, protein subunit vaccines.

Introduction Many studies performed during the early years of molecular biology demonstrated that the translational efficiency of a codon is affected by neighboring mRNA sequences [1–24]. These early experiments mainly measured the ability of a nonsense suppressor tRNA molecule to rescue the in vivo translation of an mRNA containing a chain-terminating nonsense codon located at different positions in the coding sequence. At some positions, a specific suppressor tRNA was observed to restore translation to high levels, and at other positions to low or intermediate levels. Since the only apparent difference was the position of the nonsense code, this phenomenon became known as ‘‘codon context.’’ While the molecular basis for codon context remained controversial for sometime, subsequent experiments led to the conclusion that codon context effects primarily were the consequence of physical interactions between adjacent tRNA molecules at the A(minoacyl)and P(eptidyl)-sites of a translating ribosome [25–28]. Later experiments demonstrated that these favorable or unfavorable tRNA–tRNA isoacceptor interactions dictate translational step-times [29]. This realization that tRNA–tRNA interactions on the surface of a translating ribosome influence translational kinetics and therefore nascent protein elongation rates led us to suggest a relationship between codon context and the translational ‘‘step-times’’ of individual codon pairs. We proposed that the use of one codon next to another co-evolved with the abundance and structure of tRNA isoacceptors in order to allow control of the kinetics of translation of a growing polypeptide chain without imposing constraints on amino acid sequence or protein structure. We further suggested that if tRNA compatibilities play a major role in determining translational step-times, and if the regulation of translational step-times is an important aspect of protein synthesis, then the use of adjacent codons (codon pairs) that determine the positioning of tRNAs next to one another on the ribosome should be subject to a high degree of selection, and there should be a substantially nonrandom pattern of utilization of the 3,721(612) possible pairs of non-terminating codons in protein coding sequences. In 1989, we reported that codon pair utilization patterns in the bacterium Escherichia coli were indeed highly biased, and that this bias was independent of previously observed biases in codon usage and in di-nucleotide and amino acid pair frequencies [30]. By 1992, we had extended these studies to baker’s yeast, human, rat and mouse [24]. We now know that codon pair utilization patterns are highly biased and species-specific in all organisms for which there exists a genome sequence that we have examined.

29 In this chapter, we briefly review the salient features of codon pair biases and discuss the application of this information for the design of genes that yield high protein production levels in heterologous biological hosts, and some of the new applications that this technology, now called Translation EngineeringTM holds in the diverse but related realms of synthetic biology for novel protein-based therapeutics, protein manufacturing, X-ray crystallography and SAR and restored or enhanced protein function for basic science research and biologics/drug development. Codon pair utilization patterns are highly biased Owing to the limited DNA sequences were available in 1989, even for E. coli, our initial analyses of codon pair bias were based on a non-redundant collection of protein-coding regions of only 75,403 codon pairs from 237 sequences [24,30]. More recently, this analysis has been expanded to include all of the E. coli open reading frames (ORF; see Fig. 1). Only minor changes are observed with the complete data set. To minimize the influence of differing codon usage between genes [31], codon usage frequencies for the 61 codons were determined for each sequence

Fig. 1. Total CHISQ3 values (corrected for both amino acid pair and di-nucleotide

biases) for various modes of analysis of all non-redundant ORFs of the E. coli database. The numbers above each bar represent the number of standard deviations (SD) by which the value exceeds its expectation. The solid horizontal line represents the expected value, and the broken lines represent +/3 SD. Adjacent codon pairs were evaluated for all non-redundant ORFs of the E. coli database (ECO) as well as for a randomized (RAND) version of the same sequences. Non-adjacent codon pairs were evaluated at separations of 1, 2 and 3 intervening codons, as indicated.

30 independently. These sequences were used to determine expected values for each of the 3,721 codon pairs. Comparison of these expected values with the observed values yielded a set of chi-square values, which we referred to as CHISQ1. Next, we calculated a new set of expected values for each codon pair in such a manner as to remove that component of the codon pair bias associated with bias in amino acid ‘‘nearest neighbors’’; this yielded a new set of chi-square values (CHISQ2). Finally, we applied an additional correction for bias in di-nucleotide frequencies, yielding a new set of chi-square values (CHISQ3). Thus, the bias represented by CHISQ3 cannot be the consequence of bias in codon usage (since the actual codon frequencies were used to calculate the expected values), or of biases in either amino acid or di-nucleotide nearest neighbors. Several conclusions were drawn from our analyses of the E. coli database. First, as shown in Fig. 1, there is a huge high degree of bias in codon pair utilization. The sum of CHISQ3 is more than 120 standard deviations (SDs) removed from its expected value. Second, also shown in Fig. 1, the vast majority of this bias represents a short-range effect; analyzing codon pairs separated by two or three intervening codons removes more than 95% of the excess chi square. Third, the bias is independent of all of these features, as well as our observation that genes expressed at high levels tend to avoid highly over-represented pairs (in addition to the well-known avoidance of infrequently used codons), thus supporting the conclusion that codon context (codon pair bias) was a consequence of tRNA–tRNA interactions at adjacent codons (codon pairs) during the translation process. It also suggested that over-represented codon pairs might be codons for incompatible isoacceptor tRNAs and, therefore, translated slower than less abundant codon pairs that recognize more compatible tRNA isoacceptors. Perhaps, the few hundred over-represented (and more slowly translated) codon pairs are used as translational punctuation marks to pause ribosomes (see below), as well as to specifiy amino acids during trans-peptidation and chain elongation. Over-represented codon pairs are translated slowly It is well known that ribosomes translate mRNA with a variable rate [32–39]. Translational pauses that result in the in vivo accumulation of nascent polypeptide chains have been observed during the translation of many proteins, such as rabbit reticulocyte globins [32], fibroins in the silk gland of Bombyx mori larvae [33], the ampullate gland of a spider [34], the colicins A, El, E2 and E3 in E. coli [35], the bacteriophage MS2 coat protein [36], the E. coli maltose-binding protein [37] and in vitro, during the translation of rabbit globin and tobacco mosaic virus mRNAs [38] and preprolactin [39], as well as others. We and others have suggested that a major cause of these pauses is the consequence of tRNA–tRNA interactions on the surface of a translating ribosome. All such tRNA–tRNA interactions, in turn, are dictated by

31 adjacent codons (codon pairs) within the gene-coding region (ORF). Thus, codon pairs would be expected to control ribosome translational step-times. For example, there are 36 (6 6) codon pairs for leu–leu that can be read by 25 (5 5) E. coli encoded leucyl-tRNA isoacceptor combinations. Since each combination of codon pairs and tRNA isoacceptors can exhibit different translational step-times from slow to fast, the cell is able to select translational kinetics for the placement of translational pauses for synthesis of, for example, properly folded soluble proteins without any constraints on amino acid sequences. To determine if the translational step-times of over- and under-represented codon pairs are, in fact, demonstrably different, we developed two mechanistically independent in vivo assays capable of measuring relative translational step-times across a codon pair in a growing polypeptide chain [29]. One assay, the ‘‘trp attenuator assay,’’ exploited the fact that the transit time of a ribosome through the leader polypeptide coding sequence of the leader RNA of the attenuator region of the trp operon of E. coli (transcriptionally fused to the lacZ b-galactosidase gene) sets the basal level of transcriptional read-through into this reporter gene [40]. The other assay, the ‘‘translation initiation assay,’’ exploited the fact that ribosome pausing near the beginning of a bacterial mRNA coding sequence can inhibit translation initiation by physically occluding the attachment of a new ribosome to the message [41]. In the case of the trp attenuator assay, we showed that over-represented (slowly translated) codon pairs caused de-attenuation and increased expression of the downstream b-galactosidase gene. In the case of the translation initiation assay, we showed that over-represented codon pairs early in the lacZ gene stalled ribosomes, blocked attachment of new ribosomes to the mRNA, and caused decreased expression of the b-galactosidase gene. These assays were also used to demonstrate that the highly over-represented ThrLeu codon pair, ACG CUG, was translated ten-fold slower than the highly under-represented Thr-Leu codon pair, ACC CUG, and that the more overrepresented a codon pair, the slower it was translated. The results of experiments employing these assays also were consistent with the results of our statistical analyses. They demonstrated: that codon pair biases in E. coli are directional and limited to nearest neighbors; that there is very little correlation between the chi square values of any given codon pair and its reverse counterpart; and that codon pairs that are separated by two or three intervening codons have no measureable effect on translational efficiency [30]. In addition to these correlations between codon pair bias and translational efficiency, we also were able to show that translational efficiency is more closely related to codon context (codon pair bias) than it is to the utilization frequency of individual codons. This was obvious from the observation that both ACC and CUG are frequently used codons in E. coli, but the ACC CUG and CUG ACC pairs are translated at ten-fold different rates in a context where the biases of the flanking codon pairs are not significantly

32 altered. Thus, if the differing translation rates were primarily influenced by the frequency of usage of one or the other of these codons, then the translation rates in both orientations would be expected to be the same. Finally, we used these assays to emphasize the close relationship that exists between the chi square value of a codon pair and its translational efficiency. For example, we showed that two codon pairs with nearly equal codon pair bias values, but encoding different amino acid pairs and differing at all six nucleotide positions, can exhibit the same translational efficiency. The results of these experiments prompted us to hypothesize that the translation rates of genes are ‘‘hard-wired’’ into the sequence of each gene’s ORF, and that the use of one codon next to another has co-evolved with the structure and abundance of tRNA isoacceptors in order to control the rates of translational step-times without imposing additional constraints on amino acid sequences or protein structures. This hypothesis offers a simple explanation for the large, seemingly excessive number of tRNA isoacceptor molecules found in all living cells. It implies that, for any given amino acid sequence, appropriately biased codon pairs can be employed to set the translational step-times for the addition of amino acids to the growing polypeptide chain. For example, an E. coli cell can use any combination of the 36 codon pairs and five tRNA isoacceptors to set the translational step-time for the tRNA binding and trans-peptidation reactions for the attachment of tandem Leu (Leu–Leu) amino acids to a growing polypeptide chain. This suggested that translational pauses important for the folding, or other properties of nascent polypeptide chains, can be engineered into the DNA coding sequence of any synthetic gene of any organism for which codon pair data are available. In addition, and by extension, we surmised that we could regulate the kinetics of translation by experimentally removing natural pauses to enhance protein expression and production, or conversely, by placing ribosomal pauses in precise and pre-determined locations to induce translational slowing and ensure proper folding, thus positively affecting protein structure and function. Over-represented codon pairs occur at protein structural boundaries While it has been acknowledged that the structure of proteins may be influenced by the interactions of amino acids during the synthesis of the nascent polypeptide chain, little attention has been paid to the temporal aspects (translation kinetics) of protein folding that occur during the translation process. This may not be surprising, however, since there has been little rational basis or data available for incorporating the kinetic parameters of polypeptide chain elongation into an analysis of protein structure and function. Furthermore, until recently, we have not been able to synthesize genes to test large numbers of codon pair patterns. Also, these studies have probably been discouraged because it is well known that during the synthesis of a

33 nascent polypeptide chain secondary structures, which depend on interactions between amino acids located near one another, such as a-helices, form rapidly, on the order of microseconds [24,40]. Therefore, since the average translational step-time, for example, in E. coli, is about 65 ms, the formations of such structures in the nascent polypeptide chain are not likely to be affected by varying translation rates. However, it is also well known that the establishment of higher-order interactions involving secondary structures, such as the positioning of one newly synthesized a-helical region next to another, can take much longer; up to several minutes in vitro [40]. This is because there are many ways that the nascent secondary structures might associate with one another, and it takes time for each of these alternative structures to form the most thermodynamically stable folding intermediate. Failure to give one folding intermediate (structural domain) enough time to fold properly before the next domain is translated and formed might allow secondary structures from these two domains to erroneously interact with one another. To prevent this, the positioning of slowly translated over-represented codon pairs between these structural domains (now known as ‘‘planned pausing,’’ and available as Planned Pause GenesTM) might provide the time for the correct structures to form without interference from the following structural intermediate. Because temporally controlled domain interactions are theoretically possible, the structure of a native in vivo synthesized protein would be expected to be very different from the structure of the same multi-domain polypeptide chain following denaturation and refolding of the completed polypeptide. It would also be expected to be very different from the same polypeptide chain expressed in a heterologous host with different codon pair statistics and different ribosome translational kinetics, leading to very different positioning of ribosome translational pause sites. If we assume a continuous correlation between codon pair biases (chi square values) and translational step-times, as suggested by our limited in vivo translational step-time measurements [29],1 and if we arbitrarily assign a minus sign to the chi square values of under-represented codon pairs, then we can represent the translational kinetics of any protein of any organism for which we have substantial ORF sequence data. Such data can be expressed as a map of ribosomal kinetics that we now call a ‘‘SpeedPlotTM’’ and was used in analyzing kinetic data for the following recent study of the capsid encoding a portion of the GAG gene of the Ty3 retrotransposon of Saccharomyces cerevisiae, as shown in Fig. 2A with a SpeedPlot for the structurally related capsid encoding a portion of the GAG gene of the human immunodeficiency retrovirus type 1 (HIV-1), as shown in Fig. 2B. The structures of the protein products of these genes are shown above each of the respective SpeedPlots. 1 High-throughput experiments are currently underway in the laboratories of CODA Genomics, Inc. to measure the relative step-times of all 3,721 non-terminating codon pairs.

34

Fig. 2. SpeedPlots of the capsid protein of (A) the human immunodeficiency virus, HIV-1, and (B) the Capsid Protein of the Saccharomyces cerevisiae Retrotransposon, Ty3. The ribbon structure of each protein is shown above its SpeedPlot. The SpeedPlot regions of the amino terminal and the carboxy terminal domains of each protein are indicated by brackets. The thick black horizontal lines identify the positions of ahelices.

35 The Ty3 capsid SpeedPlot in Fig. 2A shows the appearance of highly overrepresented codon pairs (translational pause sites) at two sites located at either end of a random coil region between two predicted helix bundle domains. These data are consistent with the idea discussed above that translational pauses at these sites might allow helix bundle 1 to obtain a stable conformation before competing sequences in helix bundle 2 are synthesized and the entire molecule is assembled into a functional protein. A similar comparison can be made between the structural domains of HIV-1 and the positions of over-represented codon pairs as seen in the corresponding SpeedPlots. Again, the sequences encoding the two helix bundle domains are separated by two highly over-represented codon pairs. The functional importance of this is emphasized by the fact that these codon pair patterns are conserved between these two genes while their proteins share only 16% amino acid sequence identity, and the genes themselves share even less nucleotide sequence identity. It is interesting that the HIV-1 gene also encodes over-represented codon pairs that isolate a-helices 1 and 2 including amino acid positions 17–43. Sundquist and his colleagues [43,44] have shown that specific interactions between helices 1 and 2 are required for the assembly and stabilization of the capsid hexamer of the mature viral capsid. Is it possible that the pauses following a-helices 1 and 2 allow proper folding and alignment in the absence of interference from downstream protein sequences? It is not known if these a-helices provide similar surface contacts important for assembly of the Ty3 capsid hexamer. Finally, it is noteworthy that most of the less over-represented codon pairs of both genes occur near or between ahelical secondary structures. These correlations among codon pair positions and protein structures are common, and may be essential to understanding the foundations for expressing functional proteins with complex domains in large quantities for research and manufacturing in many expression systems. Misplaced and over-represented codon pairs can inhibit protein expression Since codon pair bias, codon usage, and tRNA isoacceptor structure and abundance patterns are different in different organisms, the positioning of translational pauses must differ between organisms. For example, if a human gene sequence is translated in E. coli, a highly over-represented codon pair (translation pause) might be generated by chance near the beginning, or elsewhere in a gene where it would inhibit translation as demonstrated by our in vivo translation initiation and trp attenuator assays described above and detailed in Table 1 of reference [29]. For example, Trinh et al. [42] attempted to express a human protein in which a single Fc chain specific for the HER2/ neu gene containing VH and VL regions joined by a flexible (GGGGS)3 linker fused to a human anti-rat transferrin receptor IgG3 heavy chain with the same flexible (GGGGS)3 linker. In initial experiments they were unable to observe any expression of this protein.

36 However, using our human codon pair data [30], they found that a singlenucleotide change (GGA GGC, to GGT GGC) that resulted in the replacement of a highly over-represented codon pair with an under-represented codon pair in each (GGGGS)3 linker, but did not change the amino acid sequences, enabled them to achieve high-level expression of this fusion protein (Fig. 3). In another instance, Dr. Hung Fan of the UC-Irvine Cancer Research Institute attempted to express the cytoplasmic tail of a retroviral envelope protein as a GST fusion protein in E. coli without success. Again, removal of a highly over-represented codon pair generated by the fusion solved this problem without changing the amino acid sequence. In this case, it is likely that following coupled transcription and translation of the GST portion of the fusion gene, the over-represented codon pair causes an uncoupling of translating ribosomes from the rapidly transcribing T7 RNA polymerase of the E.coli BL21 strain, and that this uncoupling leads to degradation of the remaining unprotected portion of the mRNA encoding the cytoplasmic tail of the Jaagsiekte sheep retrovirus. Regardless of the explanation, this experimental example again illustrates that the chance occurrence of a misplaced over-represented codon pair can inhibit protein expression from a gene in a heterologous system (Fig. 4).

Fig. 3. Transient expression of fusion proteins. HEK-293T cells were transiently transfected with an anti-dansyl kappa light vector and either the fusion heavy chain containing the ‘‘slow’’ codon pairs (GGA GGC) or the heavy chain fusion containing two ‘‘fast’’ codon pairs (GGT GGC) in place of the ‘‘slow’’ ones. After 24 h [35], S-methionine was added and cells were labeled for 24 h. Supernatants were then collected, radioactive H and L chains precipitated using a mixture of rabbit anti-Fab and Fc and IgGSorb and analyzed by SDS-PAGE following reduction. (1) Cells transfected with an empty (E) vector; (2) cells transfected with the heavy chain with the ‘‘slow’’ codon pairs (GGAGGC); (3–7) serial dilutions of the immunoprecipitate from cells expressing heavy chain with the ‘‘fast’’ codon pairs [42].

37

Fig. 4. Expression of GST-JSRV cytoplasmic-tail (CT) fusion protein in E. coli strain BL21. The Coomassie blue-stained extracts of the glutathione-S-transferase protein alone (GST) and fused to the CT of the Jaagsiekte sheep retrovirus envelope protein before (GST+CT), and after the removal of the over-represented codon pair (GST+CT-CP), is shown in the Coommasie-stained protein gel in panel A. That no 32 kDa fusion protein at all was induced from the construct with the over-represented codon pair (GST+CT) is illustrated by the Western blot of this gel in Fig. 4B. (Experimental details are courtesy of Dr. Hung Fan, University of California, Irvine.)

Hot RodTM genes produce proteins at high levels CODA Genomics, Inc. now offers a custom-designed synthetic ‘‘Hot Rod’’ gene from any source organism that avoids the chance occurrences of overrepresented codon pairs and are guaranteed to express in E. coli at high levels. This guarantee is possible because these genes are not only optimized for codon usage and other properties know to help protein expression, but they are also engineered to be devoid of all mRNA secondary structures that might impede transcription, and all over-represented codon pairs that might impede translation. They are designated Hot Rod genes because translating ribosomes and E. coli RNA polymerases maintain coupled translation and transcription through these genes at maximum rates. As a consequence of this high-level coupled transcription and translation, the nascent mRNA is sequestered in polysomes and protected from degradation, even further enhancing protein production. To date, the company has produced hundreds of synthetic Hot Rod genes. The data in Fig. 5 compare the E. coli protein expression and SpeedPlots of synthetic genes for the capsid protein from the

38 A.

Hot Rod 0.02% 0.002%

Codon Optimized 0.02% 0.002%

[Arabinose]

B.

Native 0.02% 0.002%

15

1. Codon Optimized 10

5

0 0

50

100

150

200

250

300

350

(Noramilized CHISQ3)

Relative Codon Pair Usage

-5

4

Speed

2. Hot Rod 2 0 0

50

100

150

200

250

150

200

250

300

350

-2 14

3. Native

12 10

Plot of Native (Yeast) GAG

8 6 4 2 0 0

50

100

300

350

-2 -4

Codon Pair Position

Fig. 5. E. coli protein expression and SpeedPlots of synthetic genes for the capsid

protein from the capsid encoding region of the GAG gene of the Saccharomyces cerevisiae Ty3 retrotransposon. (B) (1) Gene optimized only for codon usage (codon optimized), (2) optimized for codon usage and devoid of all over-represented codon pairs (Hot Rod) and (3) containing the native yeast gene sequence. (A) Western blots of protein expressed from genes induced from the ara BAD promoter with noted concentrations of L-arabinose.

capsid, encoding region of the GAG gene of the Saccharomyces cerevisiae Ty3 retrotransposon optimized only for codon usage (1. Codon Optimized), optimized for codon usage and devoid of all over-represented codon pairs (2. Hot Rod), and containing the native yeast gene sequence (3. Native). In this

39 case, the Hot Rod gene is more highly expressed than the native yeast sequence, but only marginally better than the codon optimized gene. This is because, by chance, no highly over-represented codon pairs were generated early in the coding sequence of the codon optimized gene. However, the poorly expressed native yeast sequence contains three highly over-represented codon pairs in this same early coding sequence. By design, the Hot Rod gene expresses well because it is totally devoid of over-represented codon pairs so that the translation kinetics are optimal and the gene is highly expressed. While Hot Rod genes are always active and produce high levels of protein in E. coli, it is our experience that synthetic genes optimized only for codon usage produce higher protein levels above native sequences only about onethird of the time; that they produce nearly the same amount of protein as the native gene about one-third of the time and about one-third of the time they produce less protein than the native sequence, or even no protein at all. In most cases, SpeedPlots reveal that low protein production from these synthetic genes can be accounted for by the presence of over-represented codon pairs early in the coding sequence. In conclusion, the fact that translationally engineered Hot Rod genes always express at high levels provides, at least anecdotally, evidence in favor of the importance of Translation Engineering in synthetic gene construction for high-level protein production in heterologous host systems. In addition, the fact that Hot Rod genes usually produce protein at higher levels than native, or simply codon-optimized genes in heterologous host expression systems other than E. coli (e.g. fungi, insects, mammals and plants) suggests the universal applicability of Translation Engineering with codon pairs. As we obtain a more complete definition of the relative translational step-times across all 3,721 non-terminating codon pairs, we will become increasingly able to rationally design synthetic genes for function in addition to high-level protein production. Applications of CODA technologies in modern biological drug, diagnostics and vaccine development While the ability to identify proteins that represent exciting targets for therapeutic intervention has been scaled up, a remaining roadblock for biologicsbased drug discovery is the ability to obtain reliably functioning proteins from economical protein production systems. Simultaneously, large pharmaceutical and biotechnology companies, initially reluctant to develop protein-based drugs, therapeutics and vaccines, now have several examples of blockbuster biologic drugs such as GM-CSF, EPO, human growth hormone and insulin that have catalyzed further R&D investment, especially in the field of therapeutic antibodies, growth factors and protein hormones. They have also noted that while only 1 in 5,000 small molecule drugs that enter

40 clinical trials are approved by the FDA, the protein-based biologics have a success rates in clinical trials that are several hundreds-fold better, changing the past pharmaceutical development risk profile and ushering in a new era of protein-based biological therapeutics and vaccines. Here, we describe technology that will enable pharmaceutical and biotechnology researchers to routinely utilize fully assembled synthetic genes that have been optimized for codon pair usage that will produce large amounts of functional proteins in traditional and developing expression systems that can employ large scale bioreactor and fermentor technologies. This technology allows scalable, rational creation of functional proteins for developing protein and antibodybased drugs and vaccines that may safely treat or prevent formidable diseases with few side effects at reasonable costs for manufacturers, health organizations and patients.

References 1.

2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

Akaboshi E, Inouye M and Tsugita A. Effect of neighboring nucleotide sequences on suppression efficiency in amber mutants of T4 phage lysozyme. Mol Gen Genet 1976;149:1. Bossi L. Context effects: translation of UAG codon by suppressor tRNA is affected by the sequence following UAG in the message. J Mol Biol 1983;164:73. Fluck MM, Salser W and IL.Epstein R. The influence of the reading context upon the suppression of nonsense codons. Mol Gen Genet 1977;151:137. Corner MM, Guthrie C and McClain WH. An ochre suppressor of bacteriophage T4 that is associated with a transfer RNA. J Mol Biol 1974;90:665. Comer MM, Foss K and McClain WH. A mutation of the wobble nucleotide of a bacteriophage T4 transfer RNA. J Mol Biol 1975;99:283. Bossi L and Roth JR. The influence of codon context on genetic code translation. Nature (London) 1980;286:123. Colby DS, Schedl P and Guthrie C. A functional requirement for modification of the wobble nucleotide in the anticodon of a T4 suppressor tRNA. Cell 1976;9:449. Fluck MM and Epstein RH. Isolation and characterization of context mutations affecting the suppressibility of nonsense mutations. Mol Gen Genet 1980;177:615. Feinstein SI and Altman S. Coding properties of an ochre-suppressing derivative of Escherichia coli tRNATyr. J Mol Biol 1977;112:453. Engelberg-Kulka H. UGA suppression by normal tRNATrp in E. coli: codon context effects. Nucleic Acids Res 1981;9:983. Murgola EJ. Restricted wobble in UGA codon recognition by glycine tRNA suppressors of UGG. J Mol Biol 1981;149:1. Traboni C, Ciliberto G and Cortese R. A novel method for site-directed mutagenesis: its application to an eukaryotic tRNA’ gene promoter. EMBO J 1982;1:415. Miller JH and Albertini AM. Effects of surrounding sequence on the suppression of nonsense codons. J Mot Biol 1983;164:59. Beaudet AL and Caskey CT. Release factor translation of RNA phage terminator codons. Nature (London) 1970;227:38.

41 15. Ganoza MX and Tomkins JKM. Polypeptide chain termination in vitro: competition for nonsense codons between a purified release factor and suppressor tRNA. Biochem Biophys Res Commun 1970;40:1455. 16. Murgola EJ, Pagel FT and Hijazi KA. Codon context effects in missense suppression. J Mol Biol 1984;175:19. 17. Martin R, Weiner M and Gallant J. Effects of release factor context at UAA codons in E. coli. J Bact 1988;170:4714. 18. Murgola EJ, Hijazi HA, Groinger HU and Dahlberg AE. Mutant 16S ribosomal RNA: a codon specific translational suppressor. Proc Natl Acad Sci USA 1988;85:4162. 19. Murgola EJ. Suppression and the code: beyond codons and anticodons. Experientia 1990;46:1134. 20. Culbertson MR, Leeds P, Sandbaken MG and Wilson PG. Frameshift suppression. In: The Ribosome: Structure, Function, and Evolution, Hill WE, Dahlberg A, Garrett RA, Moore PB, Schlessinger D and Warner JR (eds), Washington, DC, American Society for Microbiology, 1990, p. 559. 21. Taniguichi T and Weissman C. Inhibition of QP RNA 70S ribosome initiation complex formation by an oligonucleotide complementary to the 30 terminal region of E. coli 16S ribosomal RNA. Nature (London) 1978;275:770. 22. Uhlenbeck OC, Bailer J and Doty P. Complementary oligonucleotide binding to the anticodon loop of fMet-transfer RNA. Nature (London) 1970;225:508. 23. Hatfield, GW and Gutman, GA. Non-random utilization of codon pairs in E. coli. U.S. Patent #5, 1992;082:767. 24. Hatfield. GW and Gutman GA. Codon pair utilization bias in bacteria, yeast and mammals. In: Transfer RNA in Protein Synthesis, Hatfield DL, Lee BJ and Pirtle RM (eds), Boca Ration, LA, CRC Press, 1993. 25. Grosjean H, Soil DG and Crothers DM. Studies of the complex between transfer RNAs with complementary anticodons. I. Origins of enhanced affinity between complementary triplets. J Mol Biol 1976;103:499. 26. Andersson SGE, Buckingham RH and Kurland CG. Does codon composition influence ribosome function? EMBO J 1984;3:91. 27. Nishimura S. Minor components in transfer RNA: their characterization, location, and function. Progr Nucl Acid Res Mol Biol 1972;12:49. 28. Smith D and Yarus M. tRNA–tRNA interactions within cellular ribosomes. Proc Natl Acad Sci USA 1989;86:4397. 29. Irwin B, Heck JD and Hatfield GW. Codon pair utilization biases influence translational elongation step times. J Biol Chem 1995;270(39):22801–22806. 30. Gutman GA and Hatfield GW. Non-random utilization of codon pairs in E. coli. Proc Natl Acad Sci USA 1989;86:3699–3703. 31. Gouy M and Gautier C. Codon usage in bacteria: correlation with gene expressivity. Nucleic Acids Res 1982;10:7055. 32. Protzel A and Morris AJ. Gel chromatographic analysis of nascent globin chains. Evidence of nonuniform size distribution. J Biol Chem 1974;249:4594–4600. 33. Lizardi PM, Mahdavi V, Shields D and Candelas G. Discontinuous translation of silk fibroin in a reticulocyte cell-free system and in intact silk gland cells. Proc Natl Acad Sci USA 1979;76:6211–6215. 34. Candelas G, Candelas T, Ortiz A and Rodriguez O. Translational pauses during a spider fibroin synthesis. Biochem Biophys Res Commun 1983;116:1033–1038. 35. Varenne S, Knibiehler M, Cavard D, Morlon J and Lazdunski C. Variable rate of polypeptide chain elongation for colicins A, E2 and E3. J Mol Biol 1982;159:57–70.

42 36. Chaney WG and Morris AJ. Nonuniform size distribution of nascent peptides. The effect of messenger RNA structure upon the rate of translation. Arch Biochem Biophys 1979;194:283–291. 37. Randall LL, Josefsson LG and Hardy SJ. Novel intermediates in the synthesis of maltose-binding protein in Escherichia coli. Eur J Biochem 1980;107:375–379. 38. Abraham AK and Pihl A. Variable rate of polypeptide chain elongation in vitro. Effect of spermidine. Eur J Biochem 1980;106:257–262. 39. Wolin SL and Walter P. Ribosome pausing and stacking during translation of a eukaryotic mRNA. EMBO J 1988;11:3559–3569. 40. Landick R and Yanofsky C. Transcription attenuation. In: E. coli and Salmonella typhimurium: Cellular and Molecular Biology, Vol. 2, Neidhardt FC, Ingraham JL, Low KB, Magasanik B, Schaechter M and Umbarger HE (eds), Washington, DC, American Society for Microbiology, 1987. 41. Liljenstrom H and von Heijne G. Translation rate modification by preferential codon usage: intragenic position effects. J Theor Biol 1987;124:43–55. 42. Trinh R, Gurbaxani B, Morrison SL and Seyfzadeh M. Optimization of codon pair use within the (GGGGS)3 linker sequence results in enhanced protein expression. Mol Immunol 2004;40:717–722. 43. von Schwedler UK, Stemmer L, Klishko VY, Li S, Albertine KH, Davis DR and Sundquist WI. Proteolytic refolding of the HIV-1 capsid protein amino terminus facilitates viral core assembly. EMBO J 1998;17:1555–1568. 44. von Schwedler UK, Stray KM, Garrus JE and Sundquist WI. Functional surfaces of the human immunodeficiency virus type-1 capsid protein. J Virol 2003;77:5349–5450.

43

Engineering, expression, purification, and production of recombinant thermolysin Kuniyo Inouye1,, Masayuki Kusano1, Yasuhiko Hashida1, Masashi Minoda2 and Kiyoshi Yasukawa1 1 Division of Food Science and Biotechnology, Graduate School of Agriculture, Kyoto University, Sakyo-ku, Kyoto 606-8502, Japan 2 Daiwa Kasei K. K., 4-19 Hie-cho, Konan-shi, Shiga 520-3203, Japan

Abstract. Thermolysin [EC 3.4.24.27] is a thermostable neutral zinc metalloproteinase originally identified in the culture broth of Bacillus thermoproteolyticus Rokko. Since the discovery in 1962, the enzyme has been extensively studied regarding its structure and catalytic mechanism. Today, thermolysin is a representative of zinc metalloproteinase and an attractive target in protein engineering to understand the catalytic mechanism, thermostability, and halophilicity. Thermolysin is used in industry, especially for the enzymatic synthesis of N-carbobenzoxy L-Asp-L-Phe methyl ester (ZDFM), a precursor of an artificial sweetener, aspartame. Generation of genetically engineered thermolysin with higher activity in the synthesis of ZDFM has been highly desired. In accordance with the expansion of studies on thermolysin, various strategies for its expression and purification have been devised and successfully used. In this review, we aim to outline recombinant thermolysins associated with their engineering, expression, purification, and production. Keywords: affinity chromatography, Bacillus stearothermophilus, Bacillus subtilis, Bacillus thermoproteolyticus, engineering, enzyme, Escherichia coli, expression, hydrophobicinteraction chromatography, inclusion body, glycyl-D-phenylalanine, metalloproteinase, NprM, recombinant protein, production, propeptide, protein engineering, production, purification, thermolysin, TLP-ste.

Abbreviations B. E. FAGLA FM IPTG npr ZD ZDFM

Bacillus Escherichia N-[3-(2-furyl)acryloyl]-glycyl-L-leucine amide L-phenylalanine methyl ester isopropyl-b-D-thiogalactopyranoside neutral protease gene N-carbobenzoxy-L-aspartic acid N-carbobenzoxy-L-aspartyl-L-phenylalanine methyl ester

Corresponding author.

E-mail: [email protected] (K. Inouye). BIOTECHNOLOGY ANNUAL REVIEW VOLUME 13 ISSN 1387-2656 DOI: 10.1016/S1387-2656(07)13003-9


44 Introduction Thermolysin [EC 3.4.24.27] is a thermostable neutral zinc metalloproteinase originally identified in the culture broth of Bacillus thermoproteolyticus Rokko [1,2]. It requires one zinc ion for enzyme activity and four calcium ions for structural stability [3–5], and specifically catalyzes the hydrolysis of peptide bonds containing hydrophobic amino acid residues [6,7]. Thermolysin consists of 316 amino acid residues, and the amino acid sequence was determined [8]. The npr gene encoding thermolysin was cloned from B. thermoproteolyticus Rokko [9]. Thermolysin is a representative zinc metalloproteinase and an attractive target in protein engineering for the following reasons. Firstly, thermolysin is the first zinc metalloproteinase to be crystallized, and the three-dimensional structure was determined [10]. Based on the structural data (Fig. 1), thermolysin consists of a b-rich N-terminal domain and an a-helical C-terminal domain. These two domains are connected by a central a-helix, which is located at the bottom of the active site cleft. Currently, more than 60 structural data of thermolysin and thermolysin-inhibitor complex are available from Protein Data Bank. The active site model of thermolysin has been applied to other zinc hydrolases for which the structural data are lacking to speculate the catalytic mechanism and design the inhibitors [11]. Secondly, thermolysin has been extensively studied regarding the effects of solvent compositions and metal substitutions. We have reported the remarkable

Fig. 1. The overall protein structure of thermolysin. Catalytically important residues

Glu143 and His231 and active site Zn2+ and four Ca2+ of thermolysin (Protein Data Bank number 8TLN) [68] are shown.

45 activation and stabilization by high concentrations (1–4 M) of neutral salts [7,12–14], the activation by cobalt substitution of the active zinc ion [15], and inhibitory effects of alcohols [16,17]. These results show that halophilicity is one of the striking features of thermolysin and that thermolysin activity depends on the dielectric constant of the reaction medium. The reaction in organic solvents has been investigated using free thermolysin [18] or that immobilized to the resin [19]. It was suggested that particular amino acid residues of thermolysin are responsible for the interaction with solvents [12–17], but have not been identified yet. Thirdly, thermolysin is widely used in industry for the peptide bond formation through reverse reaction of hydrolysis [18,20]. The most extensive use is the synthesis of N-carbobenzoxy L-Asp-L-Phe methyl ester (ZDFM), a precursor of an artificial sweetener aspartame that is 200 times sweeter than sucrose, from N-carbobenzoxy L-Asp (ZD) and L-Phe methyl ester (PM) [20]. In production of ZDFM, the reduction of enzyme cost is an important subject. Generally, there has been no great advance in the study of proteases without the advances in the methods of their production. According to the first report of thermolysin in 1962 [1], B. thermoproteolyticus Rokko was cultured in the 2000-l fermentor and thermolysin was purified from the culture supernatants through ammonium sulfate fractionation followed by acetone fractionation and crystallization. Today, native thermolysin is commercially available and is used for the synthesis of ZDFM. Studies regarding the structure and the catalytic mechanism have been performed using native and recombinant thermolysins. Because the characteristics of thermolysin mutants vary widely, various techniques have been developed for their production. In this review, we will overview the techniques used to express the recombinant thermolysin in B. subtilis and E. coli together with the characteristics of the mutant enzymes expressed. We will also discuss the techniques to purify thermolysin from the culture supernatants or cellular fractions including the recovery of active enzymes from E. coli inclusion bodies. This review describes not only thermolysin but also two other neutral proteases from B. stearothermophilus, NprM, and TLP-ste. The nprM gene encoding NprM was cloned from B. stearothermophilus MK232 and sequenced [21]. At first, NprM and thermolysin were identified to be different proteins. Subsequently, the amino acid sequence of thermolysin was corrected by two residues (Asp37 and Glu117) [22], and it was revealed that the amino acid sequence of NprM is the same as that of thermolysin. Thermolysin-like protease from B. stearothermophilus (TLP-ste) (EC. 3. 4. 24. 4) is encoded in the nprT gene of B. stearothermophilus CU21. It consists of 319 amino acid residues and differs at 44 of its 319 residues from the 316-residue thermolysin including a three-residue insertion [23,24]. TLP-ste has been subjected to extensive site-directed mutagenesis studies and is the best characterized thermolysin-like protease. The amino acid numbering of TLP-ste used in this review is that of thermolysin.

46 Engineering and expression of recombinant thermolysin All extracellular bacterial neutral proteases are synthesized as inactive preproenzyme. Figure 2 shows the nucleotide sequence of the npr gene and the amino acid sequence of thermolysin. Thermolysin is synthesized as a preproenzyme consisting of a signal peptide (28 residues), a prosequence (204 residues), and a mature sequence (316 residues). The signal peptide has a typical signal peptide sequence including basic charged residues at the N terminus followed by a long hydrophobic sequence with a putative signal peptidase cleavage sequence of Ala-X-Ala. Figure 3 shows the processing mechanism of thermolysin. The signal peptide acts as a signal for translocation of pre-prothermolysin to membrane. The pre-prothermolysin is processed into the prothermolysin by the signal peptidase. The prosequence then acts as a molecular chaperone leading to an autocleavage of the peptide bond linking the pro and mature sequences. B. subtilis and E. coli have been used as host organisms to express recombinant thermolysins. Various mutant thermolysins have been expressed to identify its catalytically important residues and improve its activity and stability. The amino acid residues of thermolysin that have been mutated are shown in Fig. 2. Because the expression system of recombinant thermolysin in B. subtilis was developed earlier than that in E. coli, all mutants except that at the Asn112 and Glu143 positions have been expressed in B. subtilis. Recombinant thermolysin expressed in B. subtilis Genus Bacillus is a sporulating and aerobic bacterium. Several Bacillus species produce extracellular proteases. The best-characterized one is a serine protease subtilisin, which is widely used as a detergent protease in industry [25]. The B. subtilis cloning and expression systems were developed in the late 1970s through the establishments of highly transformable B. subtilis strains and plasmids that code for resistance to tetracycline and chloramphenicol [26]. Since then, various bacterial extracellular proteases have been expressed in B. subtilis. Recombinant thermolysin, NprM, and TLP-ste that have been expressed in B. subtilis are summarized in Table 1. Like native thermolysin, recombinant thermolysins are all expressed as a pre-proenzyme. Shuttle vectors between E. coli and B. subtilis, and B. subtilis strains lacking neutral protease activity have been developed. First, genes coding for thermolysin [9], NprM [21] and TLP-ste [23] were expressed in B. subtilis to access the enzymatic activity in the culture supernatants when they were cloned from B. thermoproteolyticus Rokko, B. stearothermophilus MK232, and B. stearothermophilus CU21, respectively. As for thermolysin, it was reported that the purified recombinant enzyme was indistinguishable from the natural enzyme in enzymatic activity [9]. Then, site-directed mutagenesis experiments have

47

Fig. 2. The nucleotide sequence of npr gene and the amino acid sequence of therm-

olysin. Nucleotides are numbered according to the previous paper [9] and indicated on the left-hand side. Amino acids numberings indicated on the left-hand side start at Met of the initiation codon and those on the right-hand side start at Ile of the N terminus of mature thermolysin. The underlined nucleotides show the promoter sequences, and that labeled with SD shows the Shine–Dalgarno sequence. The signal peptide sequence is double underlined. The arrow indicates an autocatalytic cleavage site. The box shows the amino acid residues that have been mutated in thermolysin or TLP-ste. The amino acid numbering of TLP-ste is that of thermolysin. The consensus sequence HEXXH in the active site of zinc metalloproteinases is underlined.

48

Extracellular medium

Secretion

Outer membrane Periplasm

Folding

Autocleavage

Inner membrane Cytosol pre

pro

mature

Translation mRNA

Fig. 3. Mechanism of the extracellular production of thermolysin. Native thermoly-

sin synthesized in B. thermoproteolyticus and recombinant thermolysin synthesized in B. subtilis or E. coli are pre-proproteins consisting of a signal peptide (28 residues), a prosequence (204 residues), and a mature sequence (316 residues). The prosequence acts as an intramolecular chaperone leading to an autocatalytic cleavage of the peptide bond linking the pro and mature sequences.

Table 1. Expression of recombinant thermolysin, NprM, and TLP-ste in B. subtilis. Enzyme

Vector

B. subtilis strain

References

Thermolysin (pre-proenzyme) Thermolysin (pre-proenzyme) NprM (pre-proenzyme) TLP-ste (pre-proenzyme) TLP-ste (pre-proenzyme) TLP-ste (pre-proenzyme) TLP-ste (pre-proenzyme) TLP-ste (pre-proenzyme) TLP-ste (pre-proenzyme)

pGE501 pUBTZ2 pTNM53 pTB90 pTNT53 pNP28 pGE501 pUBTZ2 pGE530

DB117 MT2 MT2 MT1 MT2 DB104 DB117 DB117 DB117

[9,41] [37,39] [21] [23] [21] [27,57] [28,38] [40] [32–34]

been performed in the B. subtilis expression system, which are summarized in Table 2. Glu143 and His231 were demonstrated to play significant catalytic roles [27,28] as predicted by the structural data [10,29,30]. The role of Nterminal domain of TLP-ste on thermal stability has been examined [31–33], and an 8-fold mutant enzyme that functions even at 1001C was generated [34]. The autodegradation site of thermolysin was determined and the mutation at that site increased thermal stability [35,36]. Thermolysin mutants that exhibited higher activity in the hydrolysis of N-[3-(2-furyl)acryloyl]glycyl-L-leucine amide (FAGLA) [37,38] and the synthesis of ZDFM [39]

49 Table 2. Recombinant thermolysin mutants. Enzyme TLN TLN TLN TLN TLN TLN NprM TLP-ste TLP-ste TLP-ste TLP-ste TLP-ste TLP-ste TLP-ste

Mutationa

Characteristics

References

R203M, R203A Q119E L144S/D150W/N227H S103A L155S N112D W115V, W115L E143W, E143S, E143R H231A L202G N116D/Q119R/D150Q/Q225R A4T/T56A/G58A/T63F/S65P/A69P G8C/N60C A4T/G8C/T56A/G58A/N60C/T63F/S65P/A69P

2300-fold decrease in kcat/Kmb 5-fold increase in kcat/Kmc 10-fold increase in kcat/Kmd 3-fold in increase in kcat/Kmc e

[41] [37] [39] [39] [35] [50] [48] [27] [28] [40] [38] [32] [33] [34]

Increase in stability at 801C D acidic pKa ¼ 0.4c No activitye No activityf 500-fold decrease in kcat/Kmb 22-fold decrease in kcat/Kmc 4-fold increase in kcat/Kmc WT50g ¼ 23.51Ce WT50g ¼ 16.71Ce T50h 4 1001Ce

a

The amino acid numbering of TLP-ste is that of thermolysin. Hydrolysis of enkephalin. c Hydrolysis of N-[3-(2-furyl)acryloyl]-Gly-L-Leu amide (FAGLA). d Synthesis of N-carbobenzoxy-L-Asp-L-Phe methyl ester (ZDFM). e Hydrolysis of casein. f Hydrolysis of N-carbobenzoxy-L-Ala-L-Leu-L-Ala. g Shift of T50 (the temperature required to reduce initial activity by 50% in 30 min) compared with that of the wild type. h The temperature required to reduce initial activity by 50% in 30 min. b

than the wild type were generated. Site-directed mutagenesis at the S10 site of TLP-ste changed the substrate specificity [40]. The expression level of thermolysin varies from 10 mg [41] to more than 50 mg/l [9] of culture, which are sufficient for use in usual structural and functional analysis. The mechanism for the extracellular production of the recombinant thermolysin in B. subtilis is probably the same as that of native thermolysin (Fig. 3). Therefore, the mutant enzymes that lose the autocatalytic activity, such as the mutants at the Glu143 position, cannot be expressed. Mature thermolysin lacking the propeptide is not expressed, either. The E. coli expression system was initially devised to circumvent this problem in the study of thermolysin maturation, which is described below. Recombinant thermolysin expressed in E. coli E. coli is clearly the most widely used host for the production of recombinant proteins. However, considerable kinds of proteins are expressed in E. coli in the form of inclusion bodies, and biologically active proteins have to be recovered from them by denaturation and refolding processes. Table 3 summarizes recombinant thermolysin, NprM, and TLP-ste that have been expressed in E. coli. Various forms of thermolysin that cannot be expressed in B. subtilis were expressed in E. coli [42–45]. For the expression of

50 Table 3. Expression of recombinant thermolysin, NprM, and TLP-ste in E. coli. Enzyme Expression in the forms of inclusion bodies Thermolysin (His6-proenzyme) TLP-ste (His6-mature enzyme) Expression as cellular soluble protein Thermolysin (proenzyme) Thermolysin (mature and proenzymes) TLP-ste (His6-mature enzyme) Extracellular production Thermolysin (pre-proenzyme) Thermolysin (pre-proenzyme) NprM (pre-proenzyme)

Vector

E. coli strain

References

pRSET pGE501

JM109 (DE3) BL21 (DE3)

[43] [45]

pRSET B pKK222-3 pGE501

JM109 JM109 Turner (DE3)

[42] [44] [45]

pUC19 pMK4 pMK4

JM109 JM109 JM109

[49] [48] [47]

thermolysin in E. coli, a combination of the conventional E. coli promoter sequences and the induction with isopropyl-b-D-thiogalactopyranoside (IPTG) have been used. Firstly, the propeptide alone was expressed as the cellular soluble proteins [42]. The purified propeptide inhibited thermolysin activity in a non-competitive manner but facilitated the refolding of denatured thermolysin [42]. Secondly, the wild type and the inactive mutant E143A were expressed as N-terminal histidine-tagged prothermolysin in the form of inclusion bodies, and then solubilized and immobilized on a cobalt-containing resin [43]. The wild type was transformed to mature enzyme spontaneously by dialysis against a refolding buffer, while E143A was not by the same procedures [43]. Because E143A was not transformed to mature enzyme even in the presence of exogenous active wild-type thermolysin, the authors concluded that thermolysin maturation proceeded through intramolecular, but not intermolecular, cleavage [43]. Thirdly, when the mature wild-type thermolysin and the prosequence were co-expressed in E. coli as independent polypeptides, the hydrolyzing activity was detected in the soluble fraction of the cells [44]. In the case of co-expression of the mature inactive mutant E143A and the propeptide, the substrate-binding ability was detected instead of the hydrolyzing activity in the soluble fraction [44]. The authors stated that the prosequence acted as an intermolecular chaperone and the covalent linking between the propeptide and the mature enzyme was not necessary [44]. Aside from the maturation mechanism of thermolysin, it can be said that, in contrast to the expression in B. subtilis, inactive thermolysin mutants can be expressed in E. coli. The expression level of thermolysin in E. coli was estimated to be 6–8 mg/l of culture [43,44], which is almost comparable to that in B. subtilis (10–50 mg/l) [9,41]. Recently, N-terminal histidine-tagged mature TLP-ste was expressed in E. coli in the cellular soluble fractions and in the forms of inclusion bodies [45]. The enzyme activity was recovered from the inclusion bodies by denaturation and refolding procedures without an addition of the propeptide [45], which is in contrast to the results with thermolysin that no active enzyme can be obtained by expression of the mature sequence alone

51 [44]. Although the precise mechanism of thermolysin maturation is still unclear, the E. coli expression system has made a contribution to elucidating the effects of the propeptide. It should be noted that, however, in comparison with the B. subtilis expression system, the E. coli expression system is not suitable for the production of a variety of thermolysin mutants. When expressing thermolysin in cellular soluble fractions, the induction and culture conditions as well as host species should be optimized to avoid the formation of inclusion bodies, for example, by reducing the concentration of IPTG from 1 mM to 50 mM and the incubation temperature from 371C to 251C [44] and using the particular E. coli strain such as lacZY-deficient Tuner cells [45]. Indeed, the expression of mature TLP-ste in the cellular soluble fractions was successful in Tuner cells but not in the conventional E. coli strains BL21 and JM109 [45]. Such an optimization for each thermolysin mutant is troublesome and time-consuming. Moreover, if thermolysin is expressed in the form of inclusion bodies, the denaturation and refolding processes have to be done empirically. A general procedure for recovering biologically active proteins from inclusion bodies has not been established yet. These obstacles might be overcome by extracellular production of recombinant thermolysin in E. coli. Indeed, a variety of techniques has been developed for extracellular production of recombinant proteins in an E. coli expression system [46]. Extracellular production of thermolysin was attained simply by expressing the pre-prothermolysin gene under the original promoter sequences in the npr gene. Casein hydrolysis activity was detected in the culture medium of E. coli cells transformed with the plasmid containing NprM gene [47,48]. Recently we established the system for extracellular production of thermolysin in E. coli [49]. The npr gene contains the putative promoter and Shine–Dalgarno sequences in E. coli (Fig. 4). Thermolysin is constitutively expressed, presumably at levels low enough to prevent formation of inclusion bodies, and then cleaves autocatalytically the peptide bond linking the pro and mature sequences as shown in Fig. 3. Thermolysin is secreted from the periplasm into the culture medium possibly due to a combination of hydrolysis of membrane proteins by active thermolysin and an increased permeability of the cell membrane. The purified thermolysin was shown to be indistinguishable from the natural enzyme in enzymatic activity (Fig. 5, Tables 4 and 5). The expression level is approximately 50 mg/l culture [49], which is comparable to the B. subtilis expression system (10–50 mg/l) [9,41]. This expression system, like that in B. subtilis, might be applicable to expression of various thermolysin mutants. A thermolysin mutant, in which the active site Asn112 was replaced with negatively charged amino acid Asp, was expressed in this system [50]. The acidic pKa of the mutant increased by 0.4 units compared to the wild type [50]. Taken together, thermolysin can be produced intracellularly and extracellularly in E. coli by selecting the promoter sequences. In the intracellular

52

pTE1 4694 bp npr Ori

EcoRI

pre-sequence

Amp

prosequence mature sequence

XbaI

ATCAGACTCTATTTTTCCCAATACAAATACTGTAAATATTGTGTT AATATTCTAAATACAAAGAATAAAGGAGGATGAAAAATGAAAATG SD M K M

Fig. 4. Structure of the expression plasmid pTE1. The shaded region represents the

npr gene. The underline shows the promoter sequences, and that labeled with SD shows the Shine–Dalgarno sequence.

production, the propeptide alone and the inactive enzyme can be expressed by optimizing the culture condition and/or refolding processes. The extracellular production does not require such optimizations and might be applicable to various mutants if they retain an autocatalytic activity. In vitro translation N-terminal histidine-tagged prothermolysin was produced in a cell-free system in the presence of [3H]leucine [43]. Upon SDS-PAGE, radioactive protein bands corresponding to prothermolysin and mature thermolysin were detected, showing that the translated thermolysin has an autolytic activity [43]. However, it seems difficult to obtain sufficient quantities of thermolysin for structural and functional analysis in this system at this stage. Purification of recombinant thermolysin The purification of thermolysin involves several specific problems. Firstly, thermolysin, like any other protease, is sensitive toward autodegradation [51]. Acidic pH and higher temperature at which thermolysin is destabilized and susceptible to autodegradation must be avoided. Secondly, thermolysin is a sparingly soluble protein and its solubility is rather low (1.0–1.2 mg/ml) [13]. When sample volumes are relatively large, gel filtration is less convenient

53 A

kcat/Kmx 10-4(M-1s-1)

4

3

2

1 0 4.0

5.0

6.0

7.0

8.0

9.0

0.9

1.2

1.5

pH B 2

Vo(µM/s)

1.5

1

0.5

0 0.0

0.3

0.6

[ZDFM]o(mM)

Fig. 5. Enzymatic activities of the native thermolysin expressed in B. thermoproteolyticus and the recombinant thermolysin expressed in E. coli. (A) Effect of pH on the thermolysin-catalyzed hydrolysis of FAGLA. Hydrolysis of FAGLA was measured following the decrease in absorbance at 345 nm. The amount of FA-dipeptide amides hydrolyzed was evaluated by using the molar absorption difference due to hydrolysis, De345 ¼ 310 M1cm1, at 251C. The reaction was carried out with native (J) and recombinant () thermolysins at 0.1 mM in 40 mM acetateNaOH buffer at pH 4.0–5.5, 40 mM MES buffer at pH 5.5–7.0, 40 mM HEPES buffer at pH 7.0–8.5, and TAPS buffer at pH 8.0–9.0 for each of which containing 10 mM CaCl2. The enzyme activity was evaluated by the specificity constant, kcat/ Km. (B) Dependence on the substrate concentration of the reaction rate of the thermolysin-catalyzed hydrolysis of ZDFM. Hydrolysis of ZDFM was measured by following the decrease in absorbance at 224 nm. The amount of ZDFM hydrolyzed was evaluated by using the molar absorption difference due to hydrolysis, e224 ¼ 493 M1cm1, at 251C. The reaction was carried out with native (J) and recombinant () thermolysins at 0.5 mM in 40 mM Tris–HCl (pH 7.5) buffer containing 10 mM CaCl2 at 251C. Solid line represents the best fit of the Michaelis–Menten equation.

54 Table 4. pKa values of native and recombinant thermolysins in the hydrolysis of FAGLA at 251C. Thermolysin

Acidic pKa

Alkaline pKa

Native Recombinant

5.270.1 5.370.1

8.370.0 8.270.1

Table 5. Kinetic parameters of native and recombinant thermolysins in the hydrolysis of ZDFM at 251C. Thermolysin

Km (mM)

kcat (s1)

kcat/Km (mM1s1)

Native Recombinant

0.5270.14 0.4070.07

4.370.4 3.870.2

8.371.3 9.671.0

at the first step because it requires concentration of the sample and might result in precipitation. Table 6 is the list of the procedures that have been applied successfully in the purification of thermolysin and TLP-ste. Affinity column chromatography and hydrophobic-interaction column chromatography are the most preferentially used procedures. Other procedures involve ammonium sulfate fractionation, gel filtration column chromatography, and ion-exchange column chromatography. Affinity column chromatography Affinity chromatography is a common laboratory technique for purification of enzymes. Substrate or its analogous inhibitor is used as a ligand. In the case of thermolysin, glycyl-D-phenylalanine (Gly-D-Phe) and bacitracin are preferentially used. Gly-D-Phe was reported as an affinity ligand for thermolysin [52,53]. It is based on the finding that thermolysin catalyzes specifically the hydrolysis of peptide bonds with bulky hydrophobic amino acid residues such as Phe or Leu at P10 position [6]. We made the analysis of adsorption isotherms and demonstrated that the association constant to thermolysin at pH 5.5 of the resins containing Gly-D-Phe is (3.370.8) 105 M1, which was ten times more than those of glycyl-L-leucine (Gly-L-Leu) and D-phenylalanine (D-Phe), indicating that Gly-D-Phe was the most suitable ligand for purification of thermolysin [54] (Fig. 6, Table 7). Bacitracin is an antibiotic produced by B. licheniformis as a family of branched antibiotic cyclopeptides [55,56]. It exhibits high affinity for various aspartyl, serine, and metalloproteinases including thermolysin at their active sites [56]. Bacitracin-coupled silica has been successfully applied for the purification of TLP-ste [57]. Affinity chromatography using Gly-D-Phe or bacitracin is performed in the

55 Table 6. Procedures for purification of recombinant thermolysin, NprM, and TLPste. Procedures

References

Gly-D-Phe affinity column chromatography Bacitracin affinity column chromatography Hydrophobic-interaction column chromatography Ammonium sulfate fractionation Gel filtration column chromatography Ion-exchange column chromatography

[9,28,41,49,50] [31–33,38,57] [21,37,39,41,45,49,50] [39,41,57] [21,37,39,45] [45]

pH range of 5.0–5.5 [9,28,31–33,38,41,49,57], which was in agreement with our finding that the optimal pH for Gly-D-Phe affinity chromatography was 5.5 [54] and the common idea that thermolysin becomes susceptible to autodegradation in acidic pH around 4. It also coincided with the previous reports that the Micahaelis constant (Km) value in hydrolysis of the synthetic substrate N-(4-methoxyphenylazoformyl)-Leu-Leu-OH by thermolysin [58] and the inhibitor constant (Ki) values of Streptomyces metalloproteinase inhibitor talopeptin [59] and phosphoramidon [60] against thermolysin were minimum at pH 5 and increased rapidly with increasing pH. In the elution of thermolysin and TLP-ste from Gly-D-Phe or bacitracin-coupled resin, acetate buffer (pH 5.0–5.5) containing 5–10 mM CaCl2, 2.5 M NaCl, and 20%(v/v) 2-propanol has been used. Including 2-propanol is essential for the storage as well as the elution because alcohols inhibit thermolysin activity and minimize its autodegradation [16,17]. We previously proposed two inhibitory mechanisms of alcohols, one is the binding of alcohol to the active site of thermolysin and the other is decreasing the dielectric constant of the reaction medium [17]. The affinity chromatography is suitable as the final step of purification. We have performed the Gly-D-Phe affinity chromatography as the last step in the purification of recombinant thermolysin from the culture supernatants of E. coli. The eluted thermolysin is stable in the elution buffer either at 41C [49] or 201C [9]. The sample is easily desalted before use by pre-packed, disposable gel filtration columns [38].

Hydrophobic-interaction chromatography Because thermolysin is sparingly soluble, the surface of the protein is presumably hydrophobic. Therefore, in the hydrophobic-interaction column chromatography, thermolysin is thought to interact with the resin more strongly than other contaminating proteins. Hydrophobic-interaction column chromatography is suitable as the first step of purification because it does not require concentration of samples.

56 A

Bound thermolysin concentration (µM)

800 600 400 200 0 0

5

10

15

20

Free thermolysin concentration (µM)

1/Bound (ml gel/mg thermolysin)

B 0.6

0.4

0.2

0 0

0.2

0.4

0.6

0.8

1/Free thermolysin concentration (1/µM)

Fig. 6. Isotherms for thermolysin adsorption to ligand-immobilized resin at 251C. Toyopearl AF-Formyl-650 M (100 ml) containing Gly-D-Phe (J), Gly-L-Leu (&), DPhe (W), and without a ligand (B) or Sepharose 4B (100 ml) containing Gly-D-Phe () (100 ml) were equilibrated with 20 ml of 20 mM acetate buffer (pH 5.5) containing 5 mM CaCl2 (Buffer A) and placed in a 15 ml tube. Thermolysin (10 ml of 0–0.8 mg/ ml in Buffer A) was added, and the tube was rotated for 1 h at 251C. The concentration of free thermolysin (C) was determined by A277, and that of adsorbed thermolysin on the gel was calculated by a total material balance. The n and C are expressed as the Langmuir equation [69]:

n¼

nm KC 1 þ KC

ð3Þ

where nm is the maximum capacities (mg thermolysin/ml gel), and the K is the association constant (M1). (A) Isotherms for thermolysin adsorption. (B) 1/n vs. 1/C plot. n means thermolysin binding capacity of the gel (mg thermolysin/ml gel).

57 Table 7. Langmuir isotherm parameters for thermolysin adsorption at 251C. Ligand

Resin

Association constant K (M1)

Maximum capacity nm (mg thermolysin/ml gel)

Gly-D-Phe Gly-L-Leu D-Phe Gly-D-Phe

Toyopearl AF-Formyl-650M Toyopearl AF-Formyl-650M Toyopearl AF-Formyl-650M Sepharose 4B

(2.570.2) 105 (3.171.4) 104 (1.971.6) 104 (4.170.3) 105

2471 1975 2179 3172

Purification of thermolysin by hydrophobic-interaction chromatography followed by affinity column chromatography As an example of purification of thermolysin, our method is presented below. We have used the hydrophobic-interaction chromatography as the first step and the Gly-D-Phe affinity chromatography as the second. The supernatant was applied onto the hydrophobic-interaction column, and each fraction was assessed for casein hydrolysis activity. The activity was found in the fractions at the NaCl concentration of 0 M (Fig. 7A). The active fractions were pooled and then applied to the affinity chromatography. Each fraction was assessed for casein hydrolysis activity by the casein plates, and the active fractions were pooled (Fig. 7B). The thermolysin preparation thus obtained yielded a single band with a molecular mass of 34 kDa (Fig. 7C). Recovery of active enzymes from inclusion bodies When a recombinant protein is expressed in E. coli in the form of inclusion bodies, denaturation and refolding processes are required. There are two reports of the denaturation and refolding processes, one for thermolysin [43] and the other for TLP-ste [45]: N-terminal histidine-tagged prothermolysin was solubilized in guanidinium hydrochloride, and then immobilized on a cobalt-containing resin. The immobilized thermolysin was refolded by dialysis of the resin against a refolding buffer with the guanidinium hydrochloride being gradually removed [43]. In the case of TLP-ste, N-terminal histidine-tagged mature enzyme was solubilized in guanidinium hydrochloride and dialyzed against a refolding buffer without being immobilized to a resin [45]. Active enzymes were recovered in both the cases. Comparison with culture supernatants and cellular fractions as a source of purification Generally, purification of recombinant proteins from culture supernatants is easier than that from cellular fractions due to fewer contaminating proteins.

58 A

A280 (

0.3 1.0

400

0.2 200

0.5

0.1

0

0 0

200

400

600

0 800

210

230

250

ml B

)

1.5

1

0.5

0 150

170

190 ml

C 1

97.4 66.3

42.4

30.0

20.1

2

3

4

5

6

Casein hydrolysis activity (units/ml) (O)

600

0.4 )

1.5

0.5

A 280 (

NaCl(M) ( - ---- )

2.0

59 In thermolysin and TLP-ste, one or two steps are enough for the purification from culture supernatants [9,28,37–39,49,57] while more than three steps are used for the purification from cellular soluble fraction or inclusion bodies [41,45]. The recovery rate of thermolysin in the purification from the culture supernatants was 33%, [49] while that from inclusion bodies was o0.5% [43]. It might be said that the extracellular production of thermolysin is preferable with respect to purification. Future perspectives In the 1980s, the pioneering work in site-directed mutagenesis experiments has been performed using subtilisin as a model enzyme. Introducing charged amino acid residues at its substrate-binding site increased the activity (kcat/ Km) in the hydrolysis of complementary charged substrates [61]. Introducing positively charged amino acid residues at the surface and in the proximity of the active site shifted the pH-activity profile to the acidic side [62]. Rational design of enzyme based on the structural data or the computer modeling was anticipated. However, it is now generally known that the results of mutagenesis study are unpredictable. Indeed, site-directed mutagenesis study of subtilisin BPN0 (275 amino acid residues) has been done at almost every amino acid residue to generate the enzyme of industrial use [25]. The same approach has been applied to another model enzymes, lysozyme [63] and dihydrofolate reductase [64]. In the case of thermolysin and TLP-ste, several mutants exhibiting the improved activity or thermal stability were isolated rather unexpectedly [32,37–39]. To gain further insights into the structural and functional relationship and generate a high-performance thermolysin mutant of industrial use, we think that the following points are important: (1) mutations displaying considerable effects should be combined to obtain more active and/or Fig. 7. Purification of recombinant thermolysin. (A) Elution pattern of hydrophobic-

interaction chromatography of culture supernatant of the JM109 transformed with pTE1. Casein hydrolysis was performed according to the methods described previously [70]. One unit of activity is defined as the amount of enzyme activity needed to liberate a quantity of acid soluble peptide corresponding to an increase in A275 of 0.0074 (A275 of 1 mg of tyrosine)/min. (B) Elution pattern of Gly-D-Phe affinity chromatography pattern of the active fractions from a hydrophobic-interaction chromatography. The broken line indicates the change of buffer to 2.5 M NaCl, 20 mM acetate-NaOH buffer (pH 5.5), 10 mM CaCl2, 20%(v/v) 2-propanol. The arrow indicates active fractions. (C) Coomassie blue-stained 12.5% SDS-polyacrylamide gel showing the marker proteins (lane 1), native thermolysin (lane 2), supernatants of JM109 cells transformed with pTE1 (lanes 3 and 4), active fractions of hydrophobic-interaction chromatography (lane 5), active fractions of Gly-D-Phe affinity chromatography, which is purified thermolysin (lane 6). The arrow indicates the band corresponding to mature thermolysin.

60 stable enzymes and to study the interdependence of the residues mutated. Thermodynamic analysis is effective to evaluate the interdependence. (2) Mutations should be introduced extensively to surface residues. A series of our study [7,12–17,65,66] have provided the idea that the remarkable increase in activity, thermal stability, and solubility of thermolysin by neutral salts results from not only the dielectric constant of the reaction medium but also the interaction between ions and particular surface residues of thermolysin. To our knowledge, there is only one paper focusing on the mutation of the surface residues [38]. (3) A novel expression system that enables extracellular production of thermolysin mutants without the autocatalytic ability is desired. We presume that certain mutations change the recognition sequences of autocatalytic cleavage in Streptomyces griseus protease B [67]. Various strategies for expression and purification of thermolysin have been devised as described in this review, and technical hurdles that arise due to working with thermolysin presents have almost been solved. However, further improvement in rapidness, easiness, and productivity might lead to accelerating site-directed mutagenesis study and reducing production cost of a powerful recombinant thermolysin of industrial use, which will hopefully be generated in the future. If an appropriate screening method is established, library screening might also be effective. We expect that the study of thermolysin will provide many benefits to science as well to the as industry. Conclusion Various techniques for the production of recombinant thermolysin have been developed, and protein engineering of thermolysin has been successively advanced. The expression system enabling extracellular production of thermolysin was established first in B. subtilis and then in E. coli. The E. coli expression system has also been used for intracellular production of various forms of thermolysin that cannot be expressed in B. subtilis, for the study of the maturation mechanism. The mutant enzymes with higher activity or thermal stability have been generated. Affinity chromatography and hydrophobic-interaction chromatography are the most effective methods for purification of thermolysin. In the production of thermolysin, suitable techniques should be properly chosen and combined. Acknowledgments This study was supported in part (K. I.) by Grants-in-Aid for Scientific Research (Nos. 14658203 and 17380065) from the Japan Society of the Promotion of Sciences, and grants (Nos. 0150 and 0345) from the Salt Science Foundation (Tokyo, Japan).

61 References 1. 2.

3. 4. 5.

6. 7.

8. 9.

10. 11. 12.

13.

14. 15. 16. 17. 18.

19. 20.

Endo S. Studies on protease produced by thermophilic bacteria. J Ferment Technol 1962;40:346–353. Matsubara H and Feder J. Other bacterial, mold, and yeast protease. In: The Enzymes, 3rd edn, Vol. 3, P.D. Boyer (eds), New York, USA, Academic Press, 1971 pp. 721–795. Latt SA, Holmquist B and Vallee BL. Thermolysin: a zinc metalloenzyme. Biochem Biophys Res Commun 1969;37:333–339. Feder J, Garrett LR and Wildi BS. Studies on the role of calcium in thermolysin. Biochemistry 1971;10:4552–4556. Tajima M, Urabe I, Yutani K and Okada H. Role of calcium ions in the thermostability of thermolysin and Bacillus subtilis var. amylosacchariticus neutral protease. Eur J Biochem 1976;64:243–247. Morihara K and Tsuzuki H. Thermolysin: kinetic study with oligopeptides. Eur J Biochem 1970;15:374–380. Inouye K, Lee SB and Tonomura B. Effect of amino acid residues at the cleavage site of substrates on the remarkable activation of thermolysin by salts. Biochem J 1996;315:133–138. Titani K, Hermodson MA, Ericsson LH, Walsh KA and Neurath H. Amino-acid sequence of thermolysin. Nature 1972;238:35–37. O’Donohue MJ, Roques BP and Beaumont A. Cloning and expression in Bacillus subtilis of the npr gene from Bacillus thermoproteolyticus Rokko coding for the thermostable metalloprotease thermolysin. Biochem J 1994;300:599–603. Holmes MA and Matthews BW. Structure of thermolysin refined at 1.6 A˚ resolution. J Mol Biol 1982;160:623–639. Matthews BW. Structural basis of the action of thermolysin and related zinc peptidases. Acc Chem Res 1988;21:333–340. Inouye K. Effects of salts on thermolysin: activation of hydrolysis and synthesis of N-carbobenzoxy-L-asparatyl-L-phenylalanine methyl ester, and a unique change in the absorption spectrum of thermolysin. J Biochem 1992;112:335–340. Inouye K, Kuzuya K and Tonomura B. Effects of salts on the solubility of thermolysin: a remarkable increase in the solubility as well as activity by the addition of salts without aggregation or dispersion of thermolysin. J Biochem 1998;123:847–852. Oneda H, Muta Y and Inouye K. Substrate-dependent activation of thermolysin by salt. Biosci Biotechnol Biochem 2004;68:1811–1813. Kuzuya K and Inouye K. Effects of cobalt-substitution of the active zinc ion in thermolysin on its activity and active-site microenvironment. J Biochem 2001;130:783–788. Inouye K, Lee SB, Nambu K and Tonomura B. Effects of pH, temperature, and alcohols on the remarkable activation of thermolysin by salts. J Biochem 1997;122:358–364. Muta Y and Inouye K. Inhibitory effects on alcohols on thermolysin activity as examined using a fluorescent substrate. J Biochem 2002;132:945–951. Trusek-Holownia A. Synthesis of ZAlaPheOMe. The precursor of bitter depeptide in the two-phase ethyl acetate–water system catalyzed by thermolysin. J Biotechnol 2003;102:153–163. Naskanishi K and Matsuno R. Continuous peptide synthesis in a water-immisicible organic solvent with an immobilized enzyme. Ann N Y Acad Sci 1990;613:652–655. Oyama K, Kihara K and Nonaka Y. Synthesis of an aspartame precursor by immobilized thermolysin in an organic solvent. J Chem Soc Perkin 1981;II:356–360.

62 21. Kubo M and Imanaka TJ. Cloning and nucleotide sequence of the highly thermostable neutral protease gene from Bacillus stearothermophilus. J Gen Microbiol 1988;134:1883–1892. 22. Miki Y. Reassignment of 37th and 119th amino-acid residues in thermolysin. J Ferm Bioeng 1994;77:457–458. 23. Fujii M, Takagi M, Imanaka T and Aiba S. Molecular cloning of a thermostable neutral protease gene from Bacillus stearothermophilus in a vector plasmid and its expression in Bacillus stearothermophilus and Bacillus subtilis. J Bacteriol 1983;154:831–837. 24. Takagi M, Imanaka T and Aiba S. Nucleotide sequence and promoter region for the neutral protease gene from Bacillus stearothermophilus. J Bacteriol 1985;163:824–831. 25. Cherry JR and Fidantsef AL. Directed evolution of industrial enzymes: an update. Curr Opin Biotechnol 2003;14:438–443. 26. Ehrlich SD. DNA cloning in Bacillus subtilis. Proc Natl Acad Sci USA 1978;75:1433–1436. 27. Toma S, Campagnoli S, De Gregoriis E, Gianna R, Margarit I, Zamai M and Grandi G. Effect of Glu-143 and His-231 substitutions on the catalytic activity and secretion of Bacillus subtilis neutral protease. Protein Eng 1989;2:359–364. 28. Beaumont A, O’Donohue MJ, Paredes N, Rousselet N, Assicot M, Bohuon C, FournieZaluski MC and Roques BP. The role of histidine 231 in thermolysin-like enzymes. A site-directed mutagenesis study. J Biol Chem 1995;270:16803–16808. 29. Hangauer DG, Monzingo AF and Matthews BW. An interactive computer graphics study of thermolysin-catalyzed peptide cleavage and inhibition by N-carboxymethyl dipeptides. Biochemistry 1984;23:5730–5741. 30. Mock WL and Aksamawati M. Binding to thermolysin of phenolate-containing inhibitors necessitates a reversed mechanism of catalysis. Biochem J 1994;302:57–68. 31. van den Burg B, Dijkstra BW, Vriend G, van der Vinne B, Venema G and Eijsink VG. Protein stabilization by hydrophobic interactions at the surface. Eur J Biochem 1994;220:981–985. 32. Eijsink VG, Veltman OR, Aukema W, Vriend G and Venema G. Structural determinants of the stability of thermolysin-like proteinases. Nat Struct Biol 1995;2:374–379. 33. Mansfeld J, Vriend G, Dijkstra BW, Veltman OR, van den Burg B, Venema G, UlbrichHofmann R and Eijsink VG. Extreme stabilization of a thermolysin-like protease by an engineered disulfide bond. J Biol Chem 1997;272:11152–11156. 34. van den Burg B, Vriend G, Veltman OR, Venema G and Eijsink VG. Engineering an enzyme to resist boiling. Proc Natl Acad Sci USA 1998;95:2056–2060. 35. Matsumiya Y, Nishikawa K, Aoshima H, Inouye K and Kubo M. Analysis of autodegradation sites of thermolysin and enhancement of its thermostability by modifying Leu155 at an autodegradation site. J Biochem 2004;135:547–553. 36. Matsumiya Y, Nishikawa K, Inouye K and Kubo M. Mutational effect for stability in a conserved region of thermolysin. Lett Appl Microbiol 2005;40:329–334. 37. Kidokoro S, Miki Y, Endo K, Wada A, Nagao H, Miyake T, Aoyama A, Yoneya T, Kai K and Ooe S. Remarkable activity enhancement of thermolysin mutants. FEBS Lett 1995;367:73–76. 38. de Kreij A, van den Burg B, Venema G, Vriend G, Eijsink VG and Nielsen JE. The effect of modifying the surface charge on the catalytic activity of a thermolysin-like protease. J Biol Chem 2002;277:15432–15438. 39. Hanzawa S and Kidokoro S. Thermolysin. In: Encyclopedia of Bioprocess Technology: Fermentation, Biocatalysis, and Bioseparation, Flickinger MC and Drew SW (eds), WileyNew York, USA, Wiley, 1999, pp. 2527–2535.

63 40. de Kreij A, van den Burg B, Neltman OR, Vriend G, Venema G and Eijsink VG. The effect of changing the hydrophobic S10 subsite of thermolysin-like proteases on substrate specificity. Eur J Biochem 2001;268:4985–4991. 41. Marie-Claire C, Ruffet E, Antonczak S, Beaumont A, O’Donohue M, Roques BP and Fournie-Zaluski MC. Evidence by site-directed mutagenesis that arginine 203 of thermolysin and arginine 717 of neprilysin (neutral endopeptidase) play equivalent critical roles in substrate hydrolysis and inhibitor binding. Biochemistry 1997; 36:13938–13945. 42. O’Donohue MJ and Beaumont A. The roles of the prosequence of thermolysin in enzyme inhibition and folding in vitro. J Biol Chem 1996;271:26477–26481. 43. Marie-Claire C, Roques BP and Beaumont A. Intramolecular processing of prothermolysin. J Biol Chem 1998;273:5697–5701. 44. Marie-Claire C, Ruffet E, Beaumont A and Roques BP. The prosequence of thermolysin acts as an intramolecular chaperone when expressed in trans with the mature sequence in E. coli. J Mol Biol 1999;285:1911–1915. 45. Mansfeld J, Petermann E, Du¨rrschmidt P and Ulbrich-Hofmann R. The propeptide is not required to produce catalytically active neutral protease from Bacillus stearothermophilus. Protein Expr Purif 2005;39:219–228. 46. Choi JH and Lee SY. Secretory and extracellular production of recombinant proteins using Escherichia coli. Appl Microbiol Biotechnol 2004;64:625–635. 47. Yamada M, Kubo M, Miyake T, Sakaguchi R, Higo Y and Imanaka T. Promoter sequence analysis in Bacillus and Escherichia: construction of strong promoters in E. coli. Gene 1991;99:109–114. 48. Inouye K, Mazda N and Kubo M. Need for aromatic residues at position 115 for proteolytic activity found by site-directed mutagenesis of tryptophan 115 in thermolysin. Biosci Biotechnol Biochem 1998;62:798–800. 49. Inouye K, Minoda M, Takita T, Sakurama H, Hashida Y, Kusano M and Yasukawa K. Extracellular production of recombinant thermolysin expressed in Escherichia coli and its purification and enzymatic characterization. Protein Expr Purif 2006;46: 248–255. 50. Kusano M, Yasukawa K, Hashida Y and Inouye K. Engineering of the pH-dependence of thermolysin activity as examined by site-directed mutagenesis of Asn112 located at the active site of thermolysin. J Biochem 2006;139:1017–1023. 51. Fassina G, Vita C, Dalzoppo D, Zamai M, Zambonin M and Fontana A. Autolysis of thermolysin. Isolation and characterization of a folded three-fragment complex. Eur J Biochem 1986;156:221–228. 52. Pangburn MK, Burstein Y, Morgan PH, Walsh KA and Neurath H. Affinity chromatography of thermolysin and of neutral proteases from B. subtilis. Biochem Biophys Res Com 1973;54:371–379. 53. Walsh KA, Burstein Y and Pangburn MK. Affinity chromatography of thermolysin and other neutral metalloendopeptidases. Methods Enzymol 1974;34:435–440. 54. Yasukawa K, Kusano M, Nakamura K and Inouye K. Characterization of Gly-D-Phe, Gly-L-Leu, and D-Phe as affinity ligands to thermolysin. Protein Expr Purif 2006;46:332–336. 55. Stepanov VM, Rudenskaya GN, Gaida AV and Osterman AL. Affinity chromatography of proteolytic enzymes on silica-based biospecific sorbents. J Biochem Biophys Methods 1981;5:177–186. 56. Stepanov VM and Rudenskaya GN. Proteinase affinity chromatography on bacitracinSepharose. J Appl Biochem 1983;5:420–428.

64 57. van den Burg B, Eijsink VG, Stulp BK and Venema G. One-step affinity purification of Bacillus neutral proteases using bacitracin-silica. J Biochem Biophys Methods 1989;18:209–219. 58. Mock WL and Stanford DJ. Arazoformyl dipeptide substrates for thermolysin. Confirmation of a reverse protonation catalytic mechanism. Biochemistry 1996;35:7369–7377. 59. Kitagishi K and Hiromi K. Binding between thermolysin and its specific inhibitor, phosphoramidon. J Biochem 1984;95:529–534. 60. Kitagishi K, Hiromi K, Oda K and Murao S. Equilibrium study on the binding between thermolysin and Streptomyces metalloprotease inhibitor, talopeptin (MKI). J Biochem 1983;93:47–53. 61. Wells JA, Powers DB, Bott RR, Graycar TP and Estell DA. Designing substrate specificity by protein engineering of electrostatic interactions. Proc Natl Acad Sci USA 1987;84:1219–1223. 62. Russell AJ and Fersht AR. Rational modification of enzyme catalysis by engineering surface charge. Nature 1987;328:496–500. 63. Arai M and Iwakura M. Probing the interactions between the folding elements early in the folding of Escherichia coli dihydrofolate reductase by systematic sequence perturbation analysis. J Mol Biol 2005;347:337–353. 64. Matthews BW. Structural and genetic analysis of the folding and function of T4 lysozyme. FASEB J 1996;10:35–41. 65. Inouye K, Lee SB and Tonomura B. Effect of nitration and amination of tyrosyl residues in thermolysin on its hydrolytic activity and its remarkable activation by salts. J Biochem 1998;124:72–78. 66. Kamo M, Inouye K, Nagata K and Tanokura M. Preliminary X-ray crystallographic analysis of thermolysin in the presence of 4 M NaCl. Acta Cryst 2005;D61:710–712. 67. Sidhu SS and Borgford TJ. Selection of Streptomyces griseus protease B mutants with desired alterations in primary specificity using a library screening strategy. J Mol Biol 1996;257:233–245. 68. Holland DR, Tronrud DE, Pley HW, Flaherty KM, Stark W, Jansonius JN, McKay DB and Matthews BW. Structural comparison suggests that thermolysin and related neutral proteases undergo hinge-bending motion during catalysis. Biochemistry 1992;31:11310–11316. 69. Langmuir I. The adsorption of gases on plane surfaces of glass, mica and platinum. J Am Chem Soc 1918;40:1361–1403. 70. Hagihara B, Matsubara H, Nakai M and Okunui K. Crystalline bacterial proteinase. I. Preparation of crystalline proteinase of B. subtilis. J Biochem 1958;45:185–194.

65

Preparation of recombinant vaccines Eric Soler1, and Louis-Marie Houdebine2 1

Cell Biology Department, Erasmus MC, dr. Molewaterplein 50, 3015 GE, Rotterdam, The Netherlands 2 Biologie du De´veloppement et Reproduction; Institut National de la Recherche, Agronomique, 78350 Jouy en Josas, France Abstract. Vaccination is one of the most efficient ways to eradicate some infectious diseases in humans and animals. The material traditionally used as vaccines is attenuated or inactivated pathogens. This approach is sometimes limited by the fact that the material for vaccination is not efficient, not available, or generating deleterious side effects. A possible theoretical alternative is the use of recombinant proteins from the pathogens. This implies that the proteins having the capacity to vaccinate have been identified and that they can be produced in sufficient quantity at a low cost. Genetically modified organisms harboring pathogen genes can fulfil these conditions. Microorganisms, animal cells as well as transgenic plants and animals can be the source of recombinant vaccines. Each of these systems that are all getting improved has advantages and limits. Adjuvants must generally be added to the recombinant proteins to enhance their vaccinating capacity. This implies that the proteins used to vaccinate have been purified to avoid any immunization against the contaminants. The efficiency of a recombinant vaccine is poorly predictable. Multiple proteins and various modes of administration must therefore be empirically evaluated on a case-by-case basis. The structure of the recombinant proteins, the composition of the adjuvants and the mode of administration of the vaccines have a strong and not fully predictable impact on the immune response as well as the protection level against pathogens. Recombinant proteins can theoretically also be used as carriers for epitopes from other pathogens. The increasing knowledge of pathogen genomes and the availability of efficient systems to prepare large amounts of recombinant proteins greatly facilitate the potential use of recombinant proteins as vaccines. The present review is a critical analysis of the state of the art in this field. Keywords: vaccine, recombinant proteins, adjuvants, epitope carrier, VLP, transgenic animals.

Introduction The pioneering work of E. Jenner, L. Pasteur and others made it possible the eradication of smallpox from the earth by vaccinating a large number of people. Other diseases like hepatitis B and gastroenteritis induced by rotavirus might also be markedly reduced using vaccination. The method commonly used to prepare vaccines consists of obtaining sufficient amount of attenuated or inactivated pathogens and administering this material to humans or animals. Attenuated forms of the pathogen are generally obtained by natural mutation followed by a selection. The number of random mutants may be increased by using mutagenic chemicals or Corresponding author. Tel: 31-10-408-7352.

E-mail: [email protected] (E. Soler). BIOTECHNOLOGY ANNUAL REVIEW VOLUME 13 ISSN 1387-2656 DOI: 10.1016/S1387-2656(07)13004-0


66 irradiation. Alternatively, known virulent genes may be removed from the pathogen genome using genetic engineering. The native pathogen may also be inactivated by physicochemical treatment. This approach suffers from several limitations. Attenuated exploitable forms of the pathogen may not be obtained in all cases. The tools commonly used for that purpose are animal cell lines and chicken eggs. The live vaccines obtained in this way are generally potent but their composition is complex and they may induce severe deleterious effects precluding their use. This was the case for a live-attenuated vaccine against rotavirus, which induced severe intestine inflammation (intussusceptions) [1]. New vaccines still containing attenuated retrovirus are under development with expected reduced side effects [2,3]. Even in case of success, possible unknown side effects may persist with live vaccines. One is that the vaccinated persons are effectively protected but still shedding wild active viruses contributing to support the epidemic. Another problem generated by the use of attenuated or inactivated pathogen is that it is difficult to make a distinction between animals or humans who are vaccinated and those who are infected. Indeed, the same antibodies against the pathogens are present in the blood of both categories of animals or people. The absence of one gene of the pathogen may make the distinction possible between vaccinated and infected individuals. Alternatively, serum antibody markers resulting from the immunization by a foreign antigen added to the vaccine may also distinguish vaccinated and infected individuals. Viral vectors can be used to express genes coding for vaccinating proteins from another pathogen. This system cumulates the advantage of using the efficiency of the viral vector to transfer and express the foreign gene. Several strains of vaccinia virus and adenoviruses from different origins are being used successfully to vaccinate animals. Naked DNA under the form of plasmids and harboring genes coding for vaccinating proteins is also a simple, versatile and safe tool to vaccinate animals. This method still needs to be improved before being approved for animals and humans. A possible alternative consists of using subunits of the pathogens containing one or a few proteins organized as in the pathogens and forming viruslike particle (VLP) in the case of viruses [4]. This approach is expected to be safe as the material does not contain nucleic acids from the pathogen or from the vectors used to carry the gene coding for vaccinating proteins, which may be plasmids or viral vectors. This approach also makes it possible the distinction between vaccinated and infected individuals as antibodies are raised against most of the pathogen proteins after infection and only against a few of them after vaccination. The preparation of recombinant vaccines may be efficient but not easy to implement. Indeed, a long study may be necessary to define which proteins have a sufficient vaccinating capacity and ideally are efficient against most if not all the forms of the pathogens. The proteins cannot generally be obtained from the pathogen in sufficient quantity. Systems capable of providing large amounts of recombinant proteins at a low

67 cost must then be implemented. The isolated proteins are also generally less potent to induce an immune response than the whole pathogens, mainly if they are living. This implies the use of potent adjuvants, which must be devoid of deleterious side effects. The vaccinating proteins must therefore be purified to prevent any immune response against their contaminants. The vaccinating potency of recombinant proteins is largely unpredictable and this obliges experimenters to evaluate the efficiency of various modes of administration with different adjuvants. The increasing knowledge of pathogen genomes offers multiple possibilities to identify proteins and even epitopes capable of inducing a protection against the pathogens. It has become possible to test one by one the different antigens of a pathogen. This approach implies the systematic cloning of the genes coding for putative vaccinating proteins, the preparation of the corresponding proteins and the evaluation of their capacity to be used as vaccines. This brute-force method already resulted in an unprecedented burst of new antigen discovery. A less laborious approach recently met a great success to identify new antigens from group A Streptococcus. This method consists of releasing fragments of the surface antigens by treating the bacteria with proteases. The peptides were identified by mass spectrometry and the corresponding genes were cloned to prepare and evaluate the corresponding antigens. This method allowed the fast identification of antigens for vaccination [5]. This method should be applied for a number of pathogens [6]. Several systems are becoming efficient to produce large amount of recombinant proteins including vaccines. Among these systems are transgenic animals and plants. A pathogen protein having potent vaccination properties may theoretically be used as carriers for epitopes from other pathogens. In practice, the generation of fusion proteins harboring the epitopes and capable of inducing a protection against the pathogen is not an easy task. The present review examines the different steps in the preparation and the evaluation of recombinant proteins to be used as vaccines. The different systems to produce recombinant vaccines Different systems are being implemented to produce recombinant proteins for experimental use or for biotechnological applications. They include peptide chemical synthesis, microorganisms, animal cells, plant cells, transgenic plants and transgenic animals (Table 1). Peptide chemical synthesis A number of peptides covering a pathogen protein known to induce vaccination may be chemically synthesized and tested for their capacity to induce a protection against a pathogen. This was achieved with the fragments of

68

Table 1. Comparison of the different systems for the production of recombinant proteins. Production systems (Points to consider)

Bacteria

Yeast

Insect Animal cells cells+baculovirus (CHO cells)

Transgenic plants

Transgenic animals

Theoretical production level Practical production level Investment cost Production cost Flexibility Line conservation Line stability Delay for the first production Scaling up Collection Effect on organism Post translational modifications Glycosylation Stability of product Purification Contaminant pathogens Dissemination in environment Intellectual property Products on the market

+++++

+++++

+++

+

+++++

+++++

++ (+)

++ (+)

+

+

++

++++

+++++ +++++ +++++ +++++ +++++ +++++

+++++ +++++ +++++ +++++ +++++ +++++

++ ++ ++ +++ ++++ +++

+ ++ + +++ +++ +++++

++++ +++++ +++++ +++++ +++++ ++++

+++ ++++ ++++ +++++ +++++ +++ (+)

+++++ +++++ +++ (+) +

+++++ +++++ +++ (+) ++

++ +++++ +++ (+) +++

+ +++++ +++ (+) ++++

+++++ +++++ +++ (+) +++

++++ ++++ +++ ++++

+ +++++ +++ +++++

++ +++++ +++ +++++

+++ +++ +++ +++++

++++ +++ ++++ ++++

++ ++++ +++ +++++

++++ ++++ +++ ++++

+++++

+++++

+++++

+++++

++

+++++

++++ ++++

+++ +++

+++ +++

++ +++++

+++ +

+++ ++

Note: The best parameters have the largest cross number.

69 VP6 proteins from rotavirus [7,8]. Essential epitopes of a protein for vaccination may thus be determined. This also makes it possible the identification of the mechanisms of the immune response induced by each epitope. The peptides containing relevant epitopes may then be chemically synthesized and chemically linked to carrier proteins. The resulting material may be used as vaccine. Alternatively, fusion recombinant proteins containing the epitopes of interest and a carrier protein known to induce vaccination may be prepared (see below). Microorganisms Microorganisms were the first to be used to produce recombinant proteins. Human insulin has been prepared for the last twenty years by bacteria. Although highly efficient for some proteins, bacteria show limited possibilities due to the fact that they cannot fold properly a number of proteins and proceed to posttranslational modifications [9]. Bacteria may produce so high amount of recombinant proteins that they form inclusion bodies precluding an easy purification. Some proteins are toxic for bacteria and cannot be prepared in this way. Interestingly, VP6 protein prepared from bacteria proved to have vaccinating capacity almost similar to the protein prepared from insect Sf9 cells infected by recombinant baculovirus harboring the corresponding viral gene [7,8–10,11]. Yeast may be easily transformed but they often produce limited amount of recombinant proteins, which are not glycosylated or unduly glycosylated. Interestingly, several genes coding for glycosylating enzymes have been transferred into yeast, which has become capable of adding several of the carbohydrates present in human proteins [12]. It is interesting to mention that a part of the hepatitis B vaccine is prepared from recombinant yeast. It is also important to note that the viral protein prepared from yeast does not form correctly polymers by disulfide bridges. The protein must therefore be chemically reduced to allow an appropriate formation of disulfide cross-links and VLPs having full vaccinating potency. Animal cells Various animal cells are currently being used to prepare recombinant proteins for experimental studies or for biotechnical applications [13]. It is interesting to note that most of the recombinant proteins used as pharmaceuticals are being prepared from animal cells. One of the cell systems frequently used in laboratories to prepare viral proteins is the baculovirus-Sf9 cell system. This system is relatively simple to use and it proved efficient to prepare well-assembled viral proteins forming VLPs [4]. To reach this goal, the viral genes are introduced into baculovirus by homologous recombination in insect Sf9 cells. The resulting viral particles

70 are used to infect a large number of Sf9 cells that produce high amounts of viral proteins, which are not secreted but stored in cytoplasm to form spontaneously well-shaped VLPs. The VLPs can be isolated from cell lysate and purified using different protocols. One of them consists of fractionating VLPs in cesium chloride gradients. Several VLPs prepared in this way show structure similar to native corresponding viral complex as judged by electron microscopy and biochemical analysis. The baculovirus system makes it possible the preparation of VLPs from a broad variety of viruses having or not a simple or a double capsid and an envelope [4]. This tool allows the preparation of VLPs in sufficient quantity to determine their structure and to evaluate their vaccinating properties. The baculovirus-Sf9 system cannot be easily scaled up to prepare vaccines at an industrial scale. Mammalian cells can be used to prepare recombinant proteins. CHO (Chinese Hamster Ovary) cells are most frequently used to prepare pharmaceutical proteins. One of the advantages of these cells is that they proceed to most of the posttranslational modifications of proteins. However, glycosylation of recombinant proteins secreted by CHO may be incomplete due to a saturation of the glycosylating enzymes. The extremity of the carbohydrate moiety of the secreted proteins does not contain quantitatively the terminal sialic acid. The addition of genes coding for glycosylating enzymes improves the quality of the secreted proteins [14,15]. Moreover, human cells synthesize sialic acid under the NANA (N-acetylneuraminic acid) form as do rabbit and chicken cells [16,17] whereas ruminant cells synthesize also the N-glycosylneuraminic acid [18]. The glycosylation of proteins is essential for the activity of some proteins. Non-glycosylated proteins have a short half-life in vivo. Unexpectedly, a peptide which is a candidate to become a vaccine against malaria looses its capacity to vaccinate mice under a glycosylated form [19]. This exemplifies the necessity to control glycosylation of recombinant vaccines in some cases. In one case, for the preparation of a vaccine against hepatitis B, animal cells are used and provide a vaccine essentially similar to this obtained with yeast. Although efficient, CHO cells remain a costly and poorly flexible system to prepare recombinant proteins. Indeed, a 100,000 l fermentor costs 400 million dollars and five years are needed to build such a tool.

Transgenic plants The first transgenic plants were obtained in 1983. Apart from their use for basic studies, transgenic plants are increasingly used to improve food production. The idea of using transgenic plants as the source of recombinant proteins has become a reality. A number of enzymes used for research or for diagnosis are currently being produced at an industrial scale.

71 Producing pharmaceuticals in plants is a more ambitious project. This system offers several advantages but also serious limits [20]. Various plant species can be obtained as transgenics. Two essential methods are implemented to transfer genes to plants. One of these tools is the Agrobacterium tumefaciens system which contains a natural vector able to transfer a foreign gene into the plant genome. The other known as biolistics consists of projecting minute bullets covered by DNA into plant cells. In both cases, viable plants are developed from transformed somatic cells. A large number of transgenic plants can be obtained making it possible the selection of those in which the transgene is intact and functional. Foreign proteins may be stored in leaves, in seeds, or both according to the promoter used. Leaves are very abundant but it may be difficult to purify the protein of interest from them due to the presence of proteases or substances like polyphenol, which are not well-tolerated by patients. The amount of recombinant proteins which can be prepared in plants is virtually unlimited and the production cost is low. Moreover, agriculture techniques offer a great flexibility for scaling up. Leaves or seeds containing the proteins of interest can be stored easily. It is also simple to rescue the plant lines and establish master banks allowing a reproducible production of proteins. Plant cells are able to fold proteins and associate subunits as those forming antibodies essentially as efficiently as animal cells. On the contrary, plants cells add carbohydrates to protein chains but not as animal cells do. Proteins synthesized in plant cells have no terminal sialic acid and they contain xylose, which may induce deleterious immune response. Experiments are in progress to modify protein glycosylation by transferring various genes responsible for the addition of sugars to proteins in a way similar to mammalian cells [21]. Proteins prepared from plants have very little chance to contain pathogens for humans or animals. Using transgenic plants to prepare recombinant proteins raise little ethical problems. One major concern is the uncontrolled dissemination of the proteins thus of the antigens when plants are cultured in open fields [22]. Low amount of antigens might induce a tolerance in humans or a basal unknown vaccination. This problem cannot be solved easily. Plants may be sterile to prevent any dissemination of the transgene. Another proposition which has been retained by companies involved in the production of recombinant proteins by plants is to limit the gene transfer to plants not used for human feeding such as tobacco or alfalfa [23]. This does not stop completely the uncontrolled diffusion of the antigen. One possibility to suppress the problem consists of keeping the plants in greenhouses. This is technically possible but would enhance markedly the production cost reducing the attractiveness of plants for this purpose. A satisfactory approach could be to use plants, which can be cultured easily in large quantity and at a low cost in confined areas. Encouraging

72 experiments have shown that duckweed and microalgae could provide humans with large amount of proteins produced in perfectly well-controlled conditions [24]. Another possibility would be to use cultured plant cells. Recent studies suggest that this perspective offers attractive alternative in some cases [25]. Several antigens potentially to be used for vaccination have been produced in transgenic plants. The capsid protein VP6 of rotavirus has been found in tubers and leaves of transgenic potatoes [26,27]. The amount of VP6 was 0.01% of soluble tuber proteins in the first case and 0.02% and 0.06% in tubers and leaves in the second case. Oral immunization with tuber tissues generated measurable titers of both anti-VP6 IgG in serum and IgA in intestine. This proof of concept is insufficient to conclude that this method may contribute to a vaccine. A fragment of S protein from SARS virus (severe acute respiratory syndrome) was found in transgenic tomato and nicotine-free tobacco. Oral administration of transgenic tomatoes to mice induced synthesis of IgA antibodies suggesting that mucosal immune response was triggered after oral administration. Parenteral administration of transgenic tobacco to mice was followed by the presence of IgG antibodies in serum [28]. The protein G of the rabies virus was obtained in tobacco at the concentration of 0.38% of soluble proteins. Intraperitoneal injection of tobacco extract in mice in the presence of complete Freund adjuvant induced a total protection against the virus [29]. To produce anti-hepatitis B vaccine at a lower price, the antigen was produced in transgenic potatoes. The viral protein was directed to the endoplasmic reticulum by adding to the cDNA a signal peptide and the KDEL signal. Retention of the antigen in the reticulum was observed. Oral immunization of mice in the presence of cholera toxin induced the secretion of a high-antibody titer, which was still increased by boosting with parenteral administration of the potato extract [30,31]. The retention of the antigen in the reticulum may have played the role of a bioencapsulation and favored the immune response. The synthesis in transgenic rice of epitopes known to induce a tolerance toward Japanese cedar antigen was achieved. The rice extract given orally to mice inhibited Th2-mediated IgE responses to the antigen [32,33]. Recently, a system called magnifection was shown to allow the rapid production (within two weeks) of gram of functional antibodies in plants [34]. This system involves the transient high-level co-expression of the transgenes (for example immunoglobulin heavy- and light-chains) through the use of plant viruses vectors delivered by Agrobacterium to the plant body. Although encouraging, these results cannot predict when or if recombinant vaccines prepared from transgenic plants will be able to reach the market.

73 Transgenic animals The first transgenic animals were generated in 1980 and the idea of using these animals as the source of recombinant proteins was proposed two years later when the giant mice having high concentration of growth hormone were obtained. In 1985, it was shown that the DNA microinjection used to generate transgenic mice could be extrapolated to rabbits, sheep and pigs. The use of farm transgenic animals to produce recombinant proteins appeared realistic and the choice of milk as the vehicle was made in 1986. One year later, two proteins were produced in the milk of transgenic mice. This proof of concept was followed by the industrial development of the method. Only in 2006 one protein, human antithrombin III, has been approved by the European agency EMEA to be on the market. In the mean time a large part of the technical obstacle has been crossed. These problems are (i) the establishment of transgenic lines (ii) the secretion of the foreign proteins at a high level (iii) the purification of the recombinant proteins and (iv) the validation of the proteins as therapeutics on a case by case basis.

Generation of transgenic animals The generation of transgenic farm animals may be achieved according to species by DNA microinjection into embryo pronuclei, by using lentiviral vectors or transposons, by incubating sperm with DNA followed by in vitro fertilization using ICSI (Intracytoplasmic Sperm Injection), by transferring the foreign gene into pluripotent cells (embryonic stem cells or primordial germ cells) followed by the generation of chimeric animals harboring normal and transformed cells, by transferring the foreign gene into somatic cells and by the generation of cloned animals using nuclear transfer. These methods have been described in recent reviews [35–37]. They are summarized in Fig. 1. Microinjection into pronuclei is very poorly efficient in ruminants and some other species. It is still being used successfully in mice, rats, rabbits, pigs and fish. To increase the integration frequency, foreign genes can be introduced in integrating vectors such as transposons and lentiviral vectors. The latter proved highly efficient in ruminants and pigs. This technique is being adopted by experimenters even if these vectors have limited capacity to harbor foreign DNA and if the integration number is presently difficult to control. DNA transfer via sperm has been developed mainly in pigs and mice. It may simplify transgenesis in some cases. The utilization of cells as carrier for the foreign genes has been used in mice for almost twenty years. In this case, pluripotent cells capable of participating to the development of chimeric transgenic animals are being used. This method is laborious and used only for gene targeting and in practice essentially to inactivate genes (gene knockout).

74

Fig. 1. Different methods to generate transgenic animals: (1) DNA transfer via direct

microinjection into pronucleus or cytoplasm of embryo; (2) DNA transfer via a transposon: the gene of interest is introduced in the transposon, which is injected into a pronucleus; (3) DNA transfer via a lentiviral vector: the gene of interest is a lentiviral vector, which is injected between zona pellucida and membrane of oocyte or embryo; (4) DNA transfer via sperm: sperm is incubated with the foreign gene and injected into oocyte cytoplasm for fertilization by ICSI (intracytoplasmic sperm injection); (5) DNA transfer via cloning: the foreign gene is introduced into a somatic cell, the nucleus of which is introduced into the cytoplasm of an enucleated oocyte to generate a transgenic clone; (6) DNA transfer via pluripotent cells: DNA is introduced into pluripotent cell lines (ES: embryonic stem cells: lines established from early embryo, EG: embryonic germ cells: lines established from the primordial germ cells of fetal gonads). The pluripotent cells containing DNA are injected into an early embryo to generate chimeric animals harboring the foreign gene. Methods 4, 5 and 6 allow random gene addition and targeted gene integration via homologous recombination for gene addition or gene replacement including gene knockout and knockin.

For unknown reasons, it has not been possible to obtain and use pluripotent cells from embryos (ES cells: embryonic stem cells) in species other than mice. A recent study has shown that in chicken and quails it was possible to establish pluripotent cell lines (EG cells) from the pluripotent cells which are present in fetal gonads (PGC: primordial germ cells). This made it

75 possible the generation of transgenic birds, which are candidates to produce recombinant proteins in egg white. The cloning technique used to generate Dolly the sheep is being used to generate transgenic ruminants and pigs. This technique allows gene addition but also gene targeting by homologous recombination. This makes it possible gene knockout. Gene targeting is also a way to integrate foreign genes in genomic sites known to favour their expression. The generation of transgenic animals remains relatively laborious and costly but it is no more a hurdle to the production of recombinant proteins. The different sources of recombinant proteins Milk is presently the most mature system to produce recombinant proteins from transgenic organisms [38]. Blood, milk [38], egg white [39,40], seminal plasma [41], urine and silk gland [42] and insect larvae hemolymph [43] are other theoretical systems (Table 2). Silk gland is a promising system in particular cases. Preliminary results indicate that active human factor VII can be found in different tissues of a transgenic fish (tilapia). It is not known if this system may be improved and scaled up (McLean unpublished data). Blood cannot store high levels of recombinant proteins most of the time. Moreover, proteins in blood may alter the health of the animals. Milk avoids essentially these problems. Several mammalian species (rabbits, pigs, sheep, goats and cows) are currently being used to produce recombinant proteins in their milk. Rabbits offer a number of advantages: easy generation of transgenic founders and offspring, high fertility, relatively high milk production, insensitivity to prion diseases, and no transmission of severe diseases to humans. Pigs are more costly but produce higher amounts of milk than rabbits. Ruminants are potentially the most appropriate species to produce large amount of proteins but they need cloning or lentiviral vectors to integrate foreign genes, their reproduction is relatively slow, they do not glycosylate proteins as well as rabbits and pigs and they are sensitive to prion diseases. Until recently, egg white was considered as a promising system strongly limited by the great difficulty of generating transgenic birds. This difficulty appears now surmounted. Lentiviral vectors proved efficient in chicken. More impressively, pluripotent cell lines have been established in chicken and quail. These cells harboring foreign genes can be reintroduced in early embryos and participate to the development of chimeric transgenic animals [40]. In a previous experiment, the same group showed that chimeric transgenic chicken generated by using non-pluripotent cells was able to secrete a monoclonal antibody in egg white. This antibody was functional but a reduced half-life due to the lack of sialic acid in the terminal end of the carbohydrate chain [39]. These experiments validate egg white as a source of foreign proteins including recombinant vaccines.

76

Table 2. Comparison of the different sources of recombinant proteins from transgenic animals. Production systems (Points to consider)

Blood

Milk

Egg white

Seminal plasma

Urine

Silk gland

Drosophila larva

Theoretical production level Practical production level Investment cost Production cost Flexibility Line conservation Line stability Delay for the first production Scaling up Collection Effect on animal Post translational modifications Glycosylation Stability of product Purification Contaminant pathogens Dissemination in environment Intellectual property Products on the market

+++++

+++++

+++++

+++

++

++

++

++

++++

+++

+

+

++

+

+++ ++++ +++++ +++++ +++++ +++

+++ ++++ +++++ +++++ +++++ +++

++ ++++ +++++ +++++ +++++ +++

+ ++ ++ +++++ +++++ ++

+ + + +++++ +++++ +

+++ +++++ +++++ +++++ ++++ ++++

+++ ++++ ++++ +++++ +++++ ++++

++++ +++++ ++ +++++

++++ ++++ +++ ++++

++++ +++++ +++ (+) +++ (+)

++ +++ +++ (+) +++ (+)

+ +++ +++ (+) +++ (+)

+++++ ++++ ++ (+) + (+)

+++ +++++ ++++ ++ (+)

++++ (+) +++ ++ ++

++++ ++++ +++ +++

+++ ++++ +++ +++

+++ (+) +++ (+) ++ (+) +++

+++ (+) +++ (+) ++ (+) ++

++ (+) +++ (+) +++ +++

++ +++ (+) ++ (+) ++++

+++++

+++++

+++++

+++++

+++++

++++

+++++

++++ +

+++ ++++

+++ ++

+++ +

+++ +

+++ ++

+++ +

Note: The best parameters have the largest cross number.

77 Optimization of transgene expression To be expressed in a reliable manner, a transgene must ideally contain a promoter, enhancers, insulators, introns and a transcription terminator [36,44]. Expression in milk is achieved successfully with promoters from milkprotein genes. Expression in egg white is possible using the potent promoter of ovalbumin gene. Using long-genomic DNA fragments containing the promoter of interest generally enhances greatly the expression of foreign cDNA. This proved to be the case for the promoter of one milk-protein gene, WAP gene (Whey Acidic Protein) [45]. This suggests that elements from longDNA fragments will be used in future to construct compact vectors expressing transgene in a reliable manner. Constructing an efficient expression vector to produce a therapeutic protein is not a standard operation. Two examples may illustrate this point. Recombinant vaccines against malaria are presently under study [46]. One of the proteins was initially obtained in mouse milk [19]; it is now being produced in goat milk. Unexpectedly, the antigen produced in mouse milk lost its vaccinating properties when glycosylated. The second example is the production of VP2 and VP6 proteins from rotavirus in transgenic rabbit milk [47]. Rotavirus has a genome formed of several independent RNA fragments. This virus is replicated in cytoplasm and its proteins are not individually secreted. The following modifications of the VP2 and VP6 nucleotide sequence were performed: elimination of the slicing sites and of several N-glycosylation sites, addition of a peptide signal and adaptation of codons to optimize the expression of the two cDNAs in the mammary gland of the animals. The modified cDNAs were introduced into a vector designed according to the criteria defined above [44]. These gene constructs made it possible the co-secretion in milk of the two viral proteins at concentration up to 500 mg/ml. These proteins were able to protect mice against the virus completely or partially according to the mode of administration (see below). A number of experiments have shown that the posttranslational modifications of recombinant proteins secreted in milk may be incomplete. This indicates that the cellular machinery of mammary gland is not sufficient to mature completely proteins when they are secreted at a high level. Experiments carried out several years ago showed that human protein C found in mouse milk was only partly cleaved. This maturation process was complete in transgenic mice expressing furin gene coding for a cleavage enzyme [48]. This pioneer work indicates that living fermentors such as mammary gland can be engineered to perform the posttranslational modifications of recombinant proteins.

Vaccine adjuvants, formulation and delivery Identifying and producing vaccinating proteins can be a long and difficult task. But once it is done, other important challenges need to be achieved.

78 Choosing a suitable adjuvant to enhance the immune response against vaccine antigens together with choosing the right way to deliver the vaccine in recipients are critical for its efficiency. Most of the recombinant proteins or subunit vaccines are poorly immunogenic by themselves compared to whole killed or live-attenuated pathogens. They lack important features commonly present in pathogens like lipopolysaccharides (LPS) or unmethylated CpG-containing-DNA that are able to activate the innate immune system and shape the adaptive immune response. For toxicity reasons, whole killed or attenuated pathogens cannot be used in many cases. Adjuvants are then needed to increase the immunogenicity of the subunit vaccines. The common role of adjuvants is to enhance the immune response to weak antigens, and they also are implicated in the orientation of the response to a defined type: cellular or humoral, Th1- or Th2-biased response. The field of adjuvant research is very active and several new candidates are being developed and tested in animals and humans. In addition, the choice of formulation and vaccine delivery is crucial to induce an appropriate protective immune response (local or systemic). It should also be easy to handle and in the best case be needle-free and non-invasive to avoid pain and requirement for sterile material and trained medical workers (this is especially important for vaccines targeting developing countries). The following section summarizes the recent advances in these fields. Vaccine adjuvants Aluminum Despite extensive evaluation of several candidates over the past few years, the aluminum-based mineral salts (also called alum) are the only adjuvants approved by the US Food and Drug Administration (FDA) for human use. Alum is well tolerated and presents a good safety record. However, it is a relatively weak adjuvant for antibody induction against recombinant vaccines. It induces mainly a Th2 immune response and is not efficient for activation of cellular immunity (Th1) [49]. Thus alum adjuvant is suitable when antibody-based protective immunity is required (for example induction of neutralizing antibodies), but alum lacks the ability to induce mucosal IgA. This can impede efficiency of several vaccines where a strong mucosal immunity is needed to prevent pathogen entry and replication into host. This is for example the case for rotaviruses that replicate in the intestine causing severe gastroenteritis, and for which intestinal IgA were shown to protect against disease [50]. Other limitations of alum adjuvants are increased IgE production, allergenicity and neurotoxicity [49,51,52]. Alum also cannot be effective in some vaccine formulations [49]. Despite its extensive use for many years, alum mechanism of action is not completely understood. Adsorption of antigens onto alum results in the formation of a depot at the site of injection. The particulate structure of the alum/antigen complex may

79 facilitate uptake by antigen presenting cells and alum could activate complement and macrophages [49]. The saponin Quil A, derived from the bark of a Chilean tree, Quallaja saponaria, or purified extracts none as QS-21 have been evaluated as alternatives to alum for cell-mediated responses activation. The observed toxicity (local reactions, hemolysis) associated with these adjuvants renders their use in humans limited to life threatening diseases like cancer or HIV infection [53]. ISCOMS Immunostimulatory complexes (ISCOMS) are adjuvants composed of hydrophobically associated cholesterol, phospholipids and quillaja saponins that form a stable cage-like structure in which the antigens can be enclosed [54]. ISCOM-based vaccines are able to induce strong antibody and cellular immune response. It has been shown with a number of different antigens in several animal models including non-human primates (reviewed in [55]). In mouse, ISCOM-based vaccines were shown to be potent inducers of Th1 immune responses, contrarily to aluminum-based vaccines. In non-human primates, strong long-lasting CD4+ and CD8+ responses were observed following immunization with the core protein of the hepatitis C virus (HCV) complexed to the ISCOMATRIX (a preformed ISCOM preparation) in addition to humoral responses [56]. The mechanism of action of ISCOM is not fully understood. It is believed that because of their particulate structure their uptake by antigen-presenting cells is more efficient. The saponin component also has potent adjuvanticity (see above), and it has been shown that ISCOM activate the innate immune system through an IL-12-dependent mechanism [57]. In humans, a number of clinical studies were conducted with different vaccine-based ISCOM (reviewed in [55]). Antibody and/or cellular responses were induced in most of the recipients, and faster antibody responses of higher intensity were observed in people vaccinated with an influenza/ ISCOM-based vaccine [58]. ISCOM-based vaccines have been administered to several recipients and showed to be safe with low reactogenicity. Common adverse events were reaction at the site of injection and myalgia of mild intensity and of short duration. ISCOMS appear to be interesting candidates for human use. In particular, the ISCOMATRIX adjuvant has been well characterized and appears to be stable and easy to handle [55]. Finally, ISCOMS benefit from robust and reproducible manufacturing procedures that can be scaled up for industrial production. CpG Oligodeoxynucleotides CpG dinucleotides-containing oligodeoxynucleotides (CpG ODN) possess adjuvant activity and were shown to be efficient in different vaccine formulations in animals and humans. CpG ODN are currently evaluated in clinical

80 trials in humans in the field of infectious diseases, cancer treatment and asthma/allergy. CpG ODN are very potent at orienting the immune system toward a Th1-biased response and can therefore be of primary interest for vaccines where a Th1-biased reaction is needed to achieve protective immunity. Furthermore, CpG ODN are able to stimulate mucosal immunity. CpG ODN even showed greater efficiency when administered with other adjuvants like alum or in formulation like lipid emulsions or nanoparticules, which can be necessary to induce a protective response when the antigen is weak. Studies in mice showed that CpG ODN can boost both humoral and cell-mediated immune responses against a broad range of proteins or vaccines. For example inclusion of CpG ODN in a SARS coronavirus subunit vaccine composed of a fragment of the spike protein in alum, increased IgG2a titers (representative of a Th1-like response) and interferon-g (INF-g) secreting cells [59]. The same observations were reported with several other subunit vaccines against different pathogens (hepatitis A and B virus [60–63], herpes virus [64] and rotavirus [65]). The exact mechanism of action of CpG ODN is not precisely elucidated, but it is known that CpG ODN act mainly through activation in the innate immunity. The innate compartment of the immune system evolved to recognize general structures commonly found on a broad range of pathogens. These include the structure of the bacterial and of many viruses DNA, which unlike vertebrate genomic DNA, contain a high proportion of unmethylated CpG dinucleotides. Bacterial and other pathogens DNA can be recognized directly by the innate immune system through the interaction with the Toll-like receptor 9 (TLR9) which, in humans, is present in B-cells and plasmacytoid dendritic cells (pDC). In mice, TLR9 is also expressed in monocytes and in myeloid dendritic cells. The effect of TLR9 activation is the induction of a proinflammatory (IL-1, IL-6, IL-18, TNF-a) and a Th1-biased cellular and humoral immune response (reviewed in [66]). CpG ODN mimic the presence of bacterial DNA and primarily trigger activation of the innate immune system. CpG ODN are rapidly internalized by immune cells where they are bound by TLR9. The TLR9 activation caused by CpG ODN administration can enhance antigen-specific humoral or cellular immune response against co-administered antigens. Contrarily to humans, TLR9 in mice is not only expressed in B-cells and pDC but also in monocytes and myeloid dendritic cells. This observation renders difficult the extrapolation of the encouraging results obtained in mice to humans because these cells may play important roles in vaccination efficiency. However, data obtained from clinical trials in humans showed efficacy of CpG ODN adjuvants. Coadministration of CpG ODN with hepatitis B vaccine (Engerix-B) to healthy adult volunteers, either alone or in combination with alum, resulted in increased IgG titers compared to the control group receiving Engerix-B alone [67]. Furthermore, hepatitis B-specific surface antigen antibodies appeared

81 earlier when immunizations were carried out with CpG [67,68]. Inclusion of CpG adjuvant also increased the proportion of antigen-specific high-avidity antibodies [69]. An accelerated antibody response combined with increased magnitude and avidity was also observed when healthy volunteers were immunized with the anthrax vaccine adsorbed (AVA) when CpG ODN were included [70]. So far, treatments with CpG ODN were well tolerated, and the adverse effects observed among recipients only included pain and erythema at the site of injection, and mild to moderate flu-like symptoms that did not last and did not impede daily life activities [66]. Taken together, these results underline the potential of CpG ODN adjuvants both in animals and humans. Although more studies are needed and important points remain to be addressed (like the possibility to induce autoimmune diseases in recipients [66]), CpG ODN appear to be promising tools. Interestingly, CpG ODN could benefit from the large-scale – good manufacturing practices – industrial production technologies developed during the past few years for the antisens drug development (which have been approved by the US FDA) [71]. Bacterial toxins Two bacterial toxins were identified as powerful mucosal adjuvants: the cholera toxin (CT) and the related heat labile enterotoxin (LT) produced by Escherichia coli. Both toxins consist of a catalytic subunit A (CTA or LTA) associated with a pentameric cell-binding B subunit. CTA and LTA subunits possess an ADP-ribosyl transferase enzymatic activity resulting in permanent adenylate cyclase activation in targeted cells, increased cAMP production and hypersecretion of salt and water into the bowel [72,73]. The CTB and LTB parts allow the binding to cell surface through their association with GM1 gangliosides, which result in the internalization of the toxic A subunit. These toxins are internalized by polarized epithelial cells and it is thought that co-administered antigens may follow the same route. These toxins induce strong systemic and mucosal immune responses and increased responses against co-administered antigens. Vaccinations with CT and LT as adjuvants produced Th1 and Th2 responses. They showed excellent efficacy in inducing protective immunity when administered via the nasal and rectal route and to a lesser extent via the oral route [65,74,75]. However, the strong toxicity of these molecules precludes their use in humans (ingestion of 5 mg of CT in human would result in the induction of a 5-l watery diarrhea). Several less toxic derivatives that retain adjuvanticity were generated by site-directed mutagenesis. These mutants comprise the LT K63, LT R72 and LT R192G forms of LT. The LT K63 and LT R72 bear single amino acid substitutions in the catalytic A subunit. Both mutants differ in the residual enzymatic activity, which positively correlates with their adjuvanticity. LT R192G contains a single amino acid substitution in a protease sensitive portion of the catalytic A subunit [76]. This mutant with reduced enterotoxicity shows great adjuvanticity when delivered mucosally either by the nasal, oral or rectal

82 route [10,74,75,77]. Interestingly, intranasal delivery of antigens in combination with CT, LT or its derivatives induces mucosal responses even at distant sites. When rotavirus virus-like particles (VLP) were administered intranasally, strong mucosal and systemic responses were induced together with intestinal IgA production [74,78]. Encouraging results were obtained using a strategy consisting of fusing the enzymatically active A subunit of CT to a B-cell-targeting moiety (D) of Staphylococcus aureus protein A. This adjuvant, called CTA1-DD, is far less toxic than the intact CT and contrarily to CT produces a balanced Th1/Th2 response [79]. It was also shown to give comparable protection against rotavirus infection when compared with LT R192G or CpG [80]. Furthermore, in mice receiving a nasal administration of the universal influenza vaccine M2e-HBc combined with CTA1-DD, a complete protection against a lethal infection was observed, together with a reduction of morbidity, in the context of a Th1-type immunity [81]. The B subunits of CT (CTB) and LT (LTB) could also serve as mucosal adjuvants. CTB and LTB contain adjuvant activity when administered by the nasal route. Mice vaccinated with an influenza virus vaccine with LTB showed higher systemic and mucosal antibody responses than mice receiving the vaccine alone [82]. Interestingly, recent study showed that the fusion of CTB to CpG ODN (CpG-CTB) resulted in an enhancement of the immunostimulatory effect of CpG ODN, with a more potent stimulatory effect of pro-inflammatory cytokine and chemokine responses in human and mouse splenocytes [83]. It is worth being mentioned that in addition to CT and LT (and their derivatives), a third toxin called Zonula occludens toxin (Zot) showing adjuvant activity has been identified [84]. Zot is a single polypeptide chain encoded by the filamentous bacteriophage CTXF and expressed by Vibrio cholerae. Zot binds a receptor on intestinal epithelial cells and increases mucosal permeability by acting on the structure of epithelial tight junctions. This phenomenon is believed to allow penetration of antigens into the tissue where they can interact with immune cells. It is also possible that Zot does not only act as a co-delivery system for antigens but may also have immunomodulatory properties by activating antigen-presenting cells. Interestingly, the effect that Zot exerts on tight junctions is reversible and does not cause tissue damage. Several other bacterial toxins having adjuvant activity have also been identified and studied by different groups (reviewed by [85]) but their mechanisms of action still need to be clarified. Monophosphoryl lipid A LPS is a major constituent of the Gram-negative bacteria. LPS are considered to be endotoxins and induce strong pro-inflammatory reaction. LPS have strong adjuvant properties but excess production of pro-inflammatory

83 cytokines linked to repeated administration of LPS leads to endotoxin shock characterized by inflammation, profound hypotension and organ failure [86]. Because of this elevated toxicity, LPS cannot be used in humans. An LPSmimetic compound called monophosphoryl lipid A (MPL), exhibiting adjuvanticity and low toxicity has been generated. MPL, like LPS act by interacting with Toll-like receptor 4 (TLR4) on antigen presenting cells resulting in the release of pro-inflammatory cytokines like TNF-a, IL-6, IL-10 and INF-g, which will ultimately enhance the adaptive immune response (humoral and cellular). In preclinical studies, MPL has been shown to generate Th1-type immune response to antigens [87]. The molecular mechanisms resulting in the lower toxicity of MPL versus LPS are not clear; but recently, Okemoto and collegues [88] showed that contrarily to LPS, MPL activation of macrophages does not result in the release of IL-1b (a pleiotropic proinflammatory cytokine involved in the endotoxin shock [89]), nor the activation of caspase-1 (also involved in the induction of endotoxin shock). MPL adjuvant, or synthetic analogue components (RC-529) formulations have often been used in clinical trials in combination with alum and QS21 [90,91]. The adjuvant designated AS04 composed of an association of alum salts with MPL has been shown to increase antibody responses against a papillomavirus subunit vaccine in humans [92]. This formulation also led to a long-lasting immunity to the vaccine (at least 3.5 years), and an increase of memory B-cells when compared to alum salt only formulations [91–93]. More than 12,000 subjects received MPL-formulated vaccines for herpes virus [94], hepatitis B virus (HBV) [95], papillomavirus [91–93]) and extensive clinical data are available for this adjuvant. In addition, MPL is presently approved in Europe for use in combination with allergy vaccines [96]. Formulation and delivery At present different strategies are developed to optimize antigen stability and bioavailability in the host. Most of them rely on the entrapment of the antigens into polymer-based particles in the case of microspheres, or into lipid-based membranous vesicles in the case of liposomes. Microspheres are composed of biodegradable polymers, mainly polylactide (PLA) or poly(DL-lactide-co-glycolide) (PLGA). The polymers degrade in vivo to form non-toxic lactic and glycolic acids. Administered microspheres allow controlled antigen release: it may form a depot at the site of injection, allowing the slow release of the antigen for extended periods. It can thus minimize the number of doses required for immunization. Liposomes are bilayered vesicles composed of phospholipids and other sterols surrounding an aqueous center where the antigens can be entrapped. Liposomes allow for prolonged release times of antigens. Microspheres and liposomes present several advantages like increased resistance to degradation of the antigens in the gastro-intestinal tract, controlled

84 antigen release minimizing the number of doses, particle uptake by immune cells, and ability to induce cytotoxic T-lymphocytes responses. Adjuvants can also be entrapped in the particles to enhance immune responses against delivered antigens and one may include this type of formulation to increase vaccine efficiency. The choice of site of vaccine delivery is particularly important. Usually, vaccines are delivered by the parenteral route (either by subcutaneous or intra-muscular injection). This immunization regimen often leads to induction of systemic immune responses and circulating antibodies but a poor mucosal immunity. This type of immunization is suitable when serum neutralizing antibody induction is needed to prevent pathogens to replicate or to reach their target cells in the host. This is for example the case for HBV vaccine delivered parenterally by injection, where neutralizing antibodies mediate protection. However, it is generally considered that in order to produce protective immunity it is best to vaccinate via the natural route of infection of pathogens. Most pathogens infect hosts via the mucosal epithelium that represents 90% of the body surface: respiratory tract (respiratory syncitial virus), gastrointestinal (enterotoxigenic E. coli, rotavirus), vaginal (papillomavirus, HIV) or rectal mucosa (HIV). At present, a great challenge for vaccination is to stimulate a strong mucosal immunity to prevent pathogen entry into host. The easiest way to administer a vaccine is through oral delivery. However some limitations do exist. These include degradation of the antigens in the harsh gastrointestinal environment (acidity, bile salts and pancreatic secretions), and induction of oral tolerance to the antigens. One major feature of the mucosa-associated lymphoid tissue is the homing of circulating activated B-cells at distant effector sites from the site of induction [97]. This feature allows, for example, for the production of intestinal or vaginal IgA after intranasal immunization [74,78,98]. Intranasal immunization has been widely used in mouse and is recognized as a very potent induction site for protective immunity in a number of cases. However, this immunization strategy may not be well adapted for humans. Indeed, the nasal epithelium is in close contact with the olfactory bulb and the central nervous system (CNS). The close vicinity of these structures renders the intranasal delivery of bacterial toxin-based adjuvants a dangerous approach for mass vaccination since toxins and co-administered antigens could penetrate the CNS [99,100], (and see also NIAID July 9, 2001 meeting summary at http://www3.niaid.nih.gov/ research/topics/enteric/meetings.htm). Alternative immunization sites could be used to overcome this problem, for example the vaginal or the intrarectal delivery of antigens. The latter has recently been shown to be efficient for vaccination against the enteric pathogen rotavirus [65,75,119]. Recently the transcutaneous route has been shown to stimulate mucosal responses [101,102].

85 Mucosal immunization offers a number of important advantages including non-invasive (needle-free) easy administration (intranasal, oral or intra-rectal/vaginal) of vaccines. It can also be conceived that mucosal vaccines could be self-delivered without the use of sterile equipment (syringes) and trained medical workers, which may be a real advantage for vaccination in developing countries.

The use of virus-like particles as foreign antigen carrier systems Virus-like particles (VLPs) are non-infectious, non-replicating analogues of pathogenic viruses. VLPs are formed in vitro by the self-assembly of viral capsid proteins. A number of VLPs from different viruses have been described to date like papillomaviruses, rotaviruses, Norwalk viruses, hepatitis B and E viruses, and parvoviruses to name a few. Some of them are used as vaccines (papillomavirus and hepatitis B virus). The repetitive structure of the arranged capside proteins in VLPs (as in native virus particles) favors activation of B-cells and antibody production [103–105]. Some VLPs can also efficiently activate cytotoxic T-cell responses in the absence of infection and intracellular replication [106–109]. VLPs are attractive tools to present foreign epitopes to the immune system. Some chimeric VLPs have been described for hepatitis B virus [110–112], hepatitis E virus [113], rotavirus [75,114], and parvovirus [115] among others. HBV VLPs consisting of the fusion of HCV epitopes to the HBV core protein have been used in mice immunization. Both anti HBV and anti-HCV epitope responses were observed [116]. HBV VLPs were also used to carry large polypeptides like GFP (Kratz) or the ectodomain of the outer surface protein A (OspA) from Borrelia burgdorferi, the causative agent of Lyme disease. HBV/OspA hybrid VLPs immunization could protect mice against challenge with Borrelia burgdorferi [111]. In another study, inclusion of a B-cell epitope tag into hepatitis E virus (HEV) VLPs induced specific antibody responses against both the VLP and the B-cell epitope. Sedlik et al. showed that porcine parvovirus VP2 capsid protein carrying a CD8+ T-cell epitope from the lymphocytic choriomeningitis virus nucleoprotein retain its capacity to assemble into VLPs. Immunization of mice with these hybrid VLPs resulted in strong cytotoxic T-lymphocytes responses against the CD8+ epitope and protected mice against a lethal challenge with the lymphocytic choriomeningitis virus [117]. It is important to mention that in some of these experiments, vaccinations were successfully conducted without the use of adjuvant, underlining the immunostimulatory effect of VLPs on the foreign epitopes. Thus, combining the presentation of antigens in an immunogenic repetitive structure (like VLPs) with the use of powerful adjuvants should result in

86 increased efficiency of immune system activation against otherwise poorly immunogenic soluble antigens. This approach could be a nice strategy for the elaboration of combined multivalent vaccines, presenting the advantage of vaccinating against both the carrier (VLP) and the introduced epitopes. Conclusion Recombinant vaccines have well-identified theoretical advantages over conventional live vaccines. Yet, recombinant vaccines remain scarce. Vaccine against hepatitis B is one of them. A vaccine against RHDV (rabbit hemorrhagic disease virus) is also used to vaccinate rabbits. An efficient vaccine against poultry Newcastle disease has been prepared in transgenic plant and approved but not put on the market so far [118]. No more than four plantderived recombinant vaccines have reached clinical development [20]. The vaccine against malaria produced in goat milk is under clinical study whereas the vaccine anti-rotavirus produced in rabbit milk is under preclinical study. Identifying a relevant antigen capable of becoming efficient is the result of a relatively long-term study. Yet, such antigens have been characterized and could be prepared. Validating a mode of administration and determining the valuable adjuvant require specific studies on animal models. Such models are not always relevant. Mice are most frequently used species for this purpose. These animals often give only limited information. Infection by rotavirus is not followed by diarrheas. Other species not so easily used such as pigs or monkeys are then required. The different systems for the production of recombinant vaccines have been markedly improved during the last decades. Additional progress is expected but the state of the art in this field is no more a hurdle. About 475,000 l of animal cell fermentors are available and could contribute more extensively to the production of recombinant vaccines. Production in the yeast Pichia pastoris is getting more and more efficient and reliable. Transgenic plants are still facing important problems. The production level remains often low. The glycosylation problem is not expected to be solved in a near future and the uncontrolled dissemination of antigens may not find solution other than the implementation of confined areas. The only marketed proteins produced in plants are enzymes for industrial applications. Two proteins only are under clinical study, dog lipase for patients suffering from cystic fibrosis and a monoclonal antibody directed against Streptococcus mutans and preventing tooth decay [118] and none of them has been approved yet. The production in milk is the most mature system and is available to produce reliable recombinant vaccines at a low cost. The spectacular advance for generating transgenic chicken and for expressing monoclonal antibodies in egg white (3 mg per egg) suggest that this system will soon contribute to boost the production of recombinant proteins in transgenic animals. The fact

87 that Atryn (human antithrombin III) produced in goat milk has been approved by EMEA contributes companies and investors to be more confident in transgenesis to produce biopharmaceuticals. Technical gaps cannot therefore account for the slow development of recombinant vaccines. Economical reasons are the major limitation in this field. It is important to note that the vaccines all included are at the eighth position in the classification of the biopharmaceuticals [118]. The vaccine business is in the hand of five major companies, which focus their effort on influenza and childhood diseases. The demand of vaccines including recombinant vaccines remains relatively modest as these biopharmaceuticals require relatively a high investment in research. The amount of product to be prepared is relatively low. Vaccination is a preventive operation. This implies a very low level of risk. The price of vaccines is expected to be low, especially when they are to be used in developing countries. The recent rotavirus vaccines are being used in several countries despite the risk of intussusceptions as the risk due to the vaccination is significantly lower than the risk of infection. Recombinant vaccines appear to be a better tool than conventional vaccines in a number of cases. Their development might become more rapid during the coming decade as, in an increasing number of countries, governments recommend or require systematic vaccination for entry of children into schools. Recent world epidemics such as SARS or influenza incline government to support the development of new vaccines. The threat of bioterrorism is going in the same direction. The development of recombinant vaccines thus depends on political decision but technical improvements are still needed to improve the efficiency of recombinant vaccines and to lower their production cost. References 1.

2. 3. 4. 5.

6. 7.

Wood D. WHO informal consultation on quality, safety and efficacy specifications for live attenuated rotavirus vaccines Mexico City, Mexico, 8–9 February 2005. Vaccine 2005;23:5478–5487. Roberts L. Vaccines: rotavirus vaccines’ second chance. Science 2004;305:1890–1893. Glass RI. New hope for defeating rotavirus. Sci Am 2006;294:46–51 54–55. Noad R and Roy P. Virus-like particles as immunogens. Trends Microbiol 2003; 11:438–444. Rodriguez-Ortega MJ, Norais N, Bensi G, Liberatori S, Capo S, Mora M, Scarselli M, Doro F, Ferrari G, Garaguso I, Maggi T, Neumann A, Covre A, Telford JL and Grandi G. Characterization and identification of vaccine candidate proteins through analysis of the group A Streptococcus surface proteome. Nat Biotechnol 2006;24:191–197. Musser JM. The next chapter in reverse vaccinology. Nat Biotechnol 2006;24:157–158. Choi AH, Basu M, McNeal MM, Flint J, VanCott JL, Clements JD and Ward RL. Functional mapping of protective domains and epitopes in the rotavirus VP6 protein. J Virol 2000;74:11574–11580.

88 8.

9. 10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20. 21.

22.

Choi AH, McNeal MM, Basu M, Bean JA, VanCott JL, Clements JD and Ward RL. Functional mapping of protective epitopes within the rotavirus VP6 protein in mice belonging to different haplotypes. Vaccine 2003;21:761–767. Baneyx F and Mujacic M. Recombinant protein folding and misfolding in E. coli. Nat Biotechnol 2004;22:1399–1408. Choi AH, McNeal MM, Basu M, Flint JA, Stone SC, Clements JD, Bean JA, Poe SA, VanCott JL and Ward RL. Intranasal or oral immunization of inbred and outbred mice with murine or human rotavirus VP6 proteins protects against viral shedding after challenge with murine rotaviruses. Vaccine 2002;20:3310–3321. Choi AH, McNeal MM, Flint JA, Basu M, Lycke NY, Clements JD, Bean JA, Davis HL, McCluskie MJ, VanCott JL and Ward RL. The level of protection against rotavirus shedding in mice following immunization with a chimeric VP6 protein is dependent on the route and the coadministered adjuvant. Vaccine 2002;20: 1733–1740. Hamilton SR, Bobrowicz P, Bobrowicz B, Davidson RC, Li H, Mitchell T, Nett JH, Rausch S, Stadheim TA, Wischnewski H, Wildt S and Gerngross TU. Production of complex human glycoproteins in yeast. Science 2003;301:1244–1246. Fries S, Glazomitsky K, Woods A, Forrest G, Hsu A, Olewinski R, Robinson D and Chartrain M. Evaluation of disposable bioreactors. Bio Process Int 2005;3 (Suppl. 6):36–44. Umana P, Jean-Mairet J, Moudry R, Amstutz H and Bailey JE. Engineered glycoforms of an antineuroblastoma IgG1 with optimized antibody-dependent cellular cytotoxic activity. Nat Biotechnol 1999;17:176–180. Weikert S, Papac D, Briggs J, Cowfer D, Tom S, Gawlitzek M, Lofgren J, Mehta S, Chisholm V, Modi N, Eppler S, Carroll K, Chamow S, Peers D, Berman P and Krummen L. Engineering Chinese hamster ovary cells to maximize sialic acid content of recombinant glycoproteins. Nat Biotechnol 1999;17:1116–1121. Raju TS, Briggs JB, Borge SM and Jones AJ. Species-specific variation in glycosylation of IgG: evidence for the species-specific sialylation and branch-specific galactosylation and importance for engineering recombinant glycoprotein therapeutics. Glycobiology 2000;10:477–486. Koles K, van Berkel PH, Pieper FR, Nuijens JH, Mannesse ML, Vliegenthart JF and Kamerling JP. N- and O-glycans of recombinant human C1 inhibitor expressed in the milk of transgenic rabbits. Glycobiology 2004;14:51–64. Edmunds T, Van Patten SM, Pollock J, Hanson E, Bernasconi R, Higgins E, Manavalan P, Ziomek C, Meade H, McPherson JM and Cole ES. Transgenically produced human antithrombin: structural and functional comparison to human plasma-derived antithrombin. Blood 1998;91:4561–4571. Stowers AW, Chen Lh LH, Zhang Y, Kennedy MC, Zou L, Lambert L, Rice TJ, Kaslow DC, Saul A, Long CA, Meade H and Miller LH. A recombinant vaccine expressed in the milk of transgenic mice protects Aotus monkeys from a lethal challenge with Plasmodium falciparum. Proc Natl Acad Sci USA 2002;99:339–344. Ma JK, Chikwamba R, Sparrow P, Fischer R, Mahoney R and Twyman RM. Plant-derived pharmaceuticals–the road forward. Trends Plant Sci 2005;10:580–585. Gomord V, Chamberlain P, Jefferis R and Faye L. Biopharmaceutical production in plants: problems, solutions and opportunities. Trends Biotechnol 2005;23: 559–565. Kirk DD, McIntosh K, Walmsley AM and Peterson RK. Risk analysis for plant-made vaccines. Transgenic Res 2005;14:449–462.

89 23. Abranches R, Marcel S, Arcalis E, Altmann F, Fevereiro P and Stoger E. Plants as bioreactors: a comparative study suggests that Medicago truncatula is a promising production system. J Biotechnol 2005;120:121–134. 24. Leon-Banares R, Gonzalez-Ballester D, Galvan A and Fernandez E. Transgenic microalgae as green cell-factories. Trends Biotechnol 2004;22:45–52. 25. Hellwig S, Drossard J, Twyman RM and Fischer R. Plant cell cultures for the production of recombinant proteins. Nat Biotechnol 2004;22:1415–1422. 26. Matsumura T, Itchoda N and Tsunemitsu H. Production of immunogenic VP6 protein of bovine group A rotavirus in transgenic potato plants. Arch Virol 2002;147: 1263–1270. 27. Yu J and Langridge W. Expression of rotavirus capsid protein VP6 in transgenic potato and its oral immunogenicity in mice. Transgenic Res 2003;12:163–169. 28. Pogrebnyak N, Golovkin M, Andrianov V, Spitsin S, Smirnov Y, Egolf R and Koprowski H. Severe acute respiratory syndrome (SARS) S protein production in plants: development of recombinant vaccine. Proc Natl Acad Sci USA 2005;102: 9062–9067. 29. Ashraf S, Singh PK, Yadav DK, Shahnawaz M, Mishra S, Sawant SV and Tuli R. High level expression of surface glycoprotein of rabies virus in tobacco leaves and its immunoprotective activity in mice. J Biotechnol 2005;119:1–14. 30. Richter LJ, Thanavala Y, Arntzen CJ and Mason HS. Production of hepatitis B surface antigen in transgenic plants for oral immunization. Nat Biotechnol 2000;18: 1167–1171. 31. Kong Q, Richter L, Yang YF, Arntzen CJ, Mason HS and Thanavala Y. Oral immunization with hepatitis B surface antigen expressed in transgenic plants. Proc Natl Acad Sci USA 2001;98:11539–11544. 32. Ma S and Jevnikar AM. Transgenic rice for allergy immunotherapy. Proc Natl Acad Sci USA 2005;102:17255–17256. 33. Takagi H, Hiroi T, Yang L, Tada Y, Yuki Y, Takamura K, Ishimitsu R, Kawauchi H, Kiyono H and Takaiwa F. A rice-based edible vaccine expressing multiple T cell epitopes induces oral tolerance for inhibition of Th2-mediated IgE responses. Proc Natl Acad Sci USA 2005;102:17525–17530. 34. Giritch A, Marillonnet S, Engler C, van Eldik G, Botterman J, Klimyuk V and Gleba Y. Rapid high-yield expression of full-size IgG antibodies in plants coinfected with noncompeting viral vectors. Proc Natl Acad Sci USA 2006;103:14701–14706. 35. Houdebine LM. Use of transgenic animals to improve human health and animal production. Reprod Domest Anim 2005;40:269–281. 36. Houdebine LM. Transgenic Animal Models and Target Validation, Totowa, Humana Press Inc., 2006. 37. Soler E, Thepot D, Rival-Gervier S, Jolivet G and Houdebine L. Preparation of recombinant proteins in milk to improve human and animal health. Reprod Nutr Dev 2006;1–10. 38. Houdebine LM. Antibody manufacture in transgenic animals and comparisons with other systems. Curr Opin Biotechnol 2002;13:625–629. 39. Zhu L, van de Lavoir MC, Albanese J, Beenhouwer DO, Cardarelli PM, Cuison S, Deng DF, Deshpande S, Diamond JH, Green L, Halk EL, Heyer BS, Kay RM, Kerchner A, Leighton PA, Mather CM, Morrison SL, Nikolov ZL, Passmore DB, Pradas-Monne A, Preston BT, Rangan VS, Shi M, Srinivasan M, White SG, WintersDigiacinto P, Wong S, Zhou W and Etches RJ. Production of human monoclonal antibody in eggs of chimeric chickens. Nat Biotechnol 2005;23:1159–1169.

90 40. van de Lavoir MC, Diamond JH, Leighton PA, Mather-Love C, Heyer BS, Bradshaw R, Kerchner A, Hooi LT, Gessaro TM, Swanberg SE, Delany ME and Etches RJ. Germline transmission of genetically modified primordial germ cells. Nature 2006; 441:766–769. 41. Dyck MK, Lacroix D, Pothier F and Sirard MA. Making recombinant proteins in animals- – different systems, different applications. Trends Biotechnol 2003;21:394–399. 42. Royer C, Jalabert A, Da Rocha M, Grenier AM, Mauchamp B, Couble P and Chavancy G. Biosynthesis and cocoon-export of a recombinant globular protein in transgenic silkworms. Transgenic Res 2005;14:463–472. 43. Markaki M, Drabek D, Livadaras I and Craig RK. Stable expression of human Growth Hormone over 50 generations in transgenic insect larvae. Transgenic Res 2007; 16:99–107. 44. Houdebine LM. The methods to generate transgenic animals and to control transgene expression. J Biotechnol 2002;98:145–160. 45. Rival-Gervier S, Viglietta C, Maeder C, Attal J and Houdebine LM. Positionindependent and tissue-specific expression of porcine whey acidic protein gene from a bacterial artificial chromosome in transgenic mice. Mol Reprod Dev 2002; 63:161–167. 46. Giersing B and Dubovsky F. Malaria Vaccine Initiative, Bioforum Europe, 2006, p. 34. 47. Soler E, Le Saux A, Guinut F, Passet B, Cohen R, Merle C, Charpilienne A, Fourgeux C, Sorel V, Piriou A, Schwartz-Cornil I, Cohen J and Houdebine LM. Production of two vaccinating recombinant rotavirus proteins in the milk of transgenic rabbits. Transgenic Res 2005;14:833–844. 48. Lubon H. Transgenic animal bioreactors in biotechnology and production of blood proteins. Biotechnol Annu Rev 1998;4:1–54. 49. Gupta RK. Aluminum compounds as vaccine adjuvants. Adv Drug Deliver Rev 1998;32:155–172. 50. Corthesy B, Benureau Y, Perrier C, Fourgeux C, Parez N, Greenberg H and SchwartzCornil I. Rotavirus anti-VP6 secretory IgA contributes to protection via intracellular neutralization but not via immune exclusion. J Virol 2006;80:10692–10699. 51. Audibert FM and Lise LD. Adjuvants: current status, clinical perspectives and future prospects. Immunol Today 1993;14:281–284. 52. Goto N, Kato H, Maeyama J, Eto K and Yoshihara S. Studies on the toxicities of aluminium hydroxide and calcium phosphate as immunological adjuvants for vaccines. Vaccine 1993;11:914–918. 53. Petrovsky N. Novel human polysaccharide adjuvants with dual Th1 and Th2 potentiating activity. Vaccine 2006;24(Suppl 2):S2-26-9. 54. Morein B, Sundquist B, Hoglund S, Dalsgaard K and Osterhaus A. Iscom, a novel structure for antigenic presentation of membrane proteins from enveloped viruses. Nature 1984;308:457–460. 55. Sanders MT, Brown LE, Deliyannis G and Pearse MJ. ISCOM-based vaccines: the second decade. Immunol Cell Biol 2005;83:119–128. 56. Polakos NK, Drane D, Cox J, Ng P, Selby MJ, Chien D, O’Hagan DT, Houghton M and Paliard X. Characterization of hepatitis C virus core-specific immune responses primed in rhesus macaques by a nonclassical ISCOM vaccine. J Immunol 2001; 166:3589–3598. 57. Smith RE, Donachie AM, Grdic D, Lycke N and Mowat AM. Immune-stimulating complexes induce an IL-12-dependent cascade of innate immune responses. J Immunol 1999;162:5536–5546.

91 58. Rimmelzwaan GF, Nieuwkoop N, Brandenburg A, Sutter G, Beyer WE, Maher D, Bates J and Osterhaus AD. A randomized, double blind study in young healthy adults comparing cell mediated and humoral immune responses induced by influenza ISCOM vaccines and conventional vaccines. Vaccine 2000;19:1180–1187. 59. Zakhartchouk AN, Sharon C, Satkunarajah M, Auperin T, Viswanathan S, Mutwiri M. Petric G, See RH, Brunham RC, Finlay BB, Cameron C, Kelvin DJ, Cochrane A, Rini JM and Babiuk LA. Immunogenicity of a receptor-binding domain of SARS coronavirus spike protein in mice: implications for a subunit vaccine. Vaccine 2007; 25:136–143. 60. Brazolot Millan CL, Weeratna R, Krieg AM, Siegrist CA and Davis HL. CpG DNA can induce strong Th1 humoral and cell-mediated immune responses against hepatitis B surface antigen in young mice. Proc Natl Acad Sci USA 1998;95:15553–15558. 61. Davis HL, Weeratna R, Waldschmidt TJ, Tygrett L, Schorr J and Krieg AM. CpG DNA is a potent enhancer of specific immunity in mice immunized with recombinant hepatitis B surface antigen. J Immunol 1998;160:870–876. 62. Moldoveanu Z, Love-Homan L, Huang WQ and Krieg AM. CpG DNA, a novel immune enhancer for systemic and mucosal immunization with influenza virus. Vaccine 1998;16:1216–1224. 63. Mitchell LA, Joseph A, Kedar E, Barenholz Y and Galun E. Mucosal immunization against hepatitis A: antibody responses are enhanced by co-administration of synthetic oligodeoxynucleotides and a novel cationic lipid. Vaccine 2006;24:5300–5310. 64. Tengvall S, Lundqvist A, Eisenberg RJ, Cohen GH and Harandi AM. Mucosal administration of CpG oligodeoxynucleotide elicits strong CC and CXC chemokine responses in the vagina and serves as a potent Th1-tilting adjuvant for recombinant gD2 protein vaccination against genital herpes. J Virol 2006;80:5283–5291. 65. Agnello D, Herve CA, Lavaux A, Darniot M, Guillon P, Charpilienne A and Pothier P. Intrarectal immunization with rotavirus 2/6 virus-like particles induces an antirotavirus immune response localized in the intestinal mucosa and protects against rotavirus infection in mice. J Virol 2006;80:3823–3832. 66. Klinman DM. Adjuvant activity of CpG oligodeoxynucleotides. Int Rev Immunol 2006;25:135–154. 67. Halperin SA, Van Nest G, Smith B, Abtahi S, Whiley H and Eiden JJ. A phase I study of the safety and immunogenicity of recombinant hepatitis B surface antigen coadministered with an immunostimulatory phosphorothioate oligonucleotide adjuvant. Vaccine 2003;21:2461–2467. 68. Cooper CL, Davis HL, Morris ML, Efler SM, Adhami MA, Krieg AM, Cameron DW and Heathcote J. CPG 7909, an immunostimulatory TLR9 agonist oligodeoxynucleotide, as adjuvant to Engerix-B HBV vaccine in healthy adults: a double-blind phase I/II study. J Clin Immunol 2004;24:693–701. 69. Siegrist CA, Pihlgren M, Tougne C, Efler SM, Morris ML, AlAdhami MJ, Cameron DW, Cooper CL, Heathcote J, Davis HL and Lambert PH. Co-administration of CpG oligonucleotides enhances the late affinity maturation process of human anti-hepatitis B vaccine response. Vaccine 2004;23:615–622. 70. Klinman DM. CpG oligonucleotides accelerate and boost the immune response elicited by AVA, the licensed anthrax vaccine. Expert Rev Vaccines 2006;5:365–369. 71. Krieg AM. Therapeutic potential of Toll-like receptor 9 activation. Nat Rev Drug Discov 2006;5:471–484. 72. Foss DL and Murtaugh MP. Mechanisms of vaccine adjuvanticity at mucosal surfaces. Anim Health Res Rev 2000;1:3–24.

92 73. Eriksson K and Holmgren J. Recent advances in mucosal vaccines and adjuvants. Curr Opin Immunol 2002;14:666–672. 74. Bertolotti-Ciarlet A, Ciarlet M, Crawford SE, Conner ME and Estes MK. Immunogenicity and protective efficacy of rotavirus 2/6-virus-like particles produced by a dual baculovirus expression vector and administered intramuscularly, intranasally, or orally to mice. Vaccine 2003;21:3885–3900. 75. Parez N, Fourgeux C, Mohamed A, Dubuquoy C, Pillot M, Dehee A, Charpilienne A, Poncet D, Schwartz-Cornil I and Garbarg-Chenon A. Rectal immunization with rotavirus like particles induces systemic and mucosal humoral immune responses and protects mice against rotavirus infection. J Virol 2006;80:1752–1761. 76. Dickinson BL and Clements JD. Dissociation of E. coli heat-labile enterotoxin adjuvanticity from ADP-ribosyltransferase activity. Infect Immun 1995;63:1617–1623. 77. Yuan L, Geyer A, Hodgins DC, Fan Z, Qian Y, Chang KO, Crawford SE, Parreno V, Ward LA, Estes MK, Conner ME and Saif LJ. Intranasal administration of 2/6rotavirus-like particles with mutant E. coli heat-labile toxin (LT-R192G) induces antibody-secreting cell responses but not protective immunity in gnotobiotic pigs. J Virol 2000;74:8843–8853. 78. O’Neal CM, Clements JD, Estes MK and Conner ME. Rotavirus 2/6 viruslike particles administered intranasally with cholera toxin, E. coli heat-labile toxin (LT), and LT-R192G induce protection from rotavirus challenge. J Virol 1998;72: 3390–3393. 79. Eriksson AM, Schon KM and Lycke NY. The cholera toxin-derived CTA1-DD vaccine adjuvant administered intranasally does not cause inflammation or accumulate in the nervous tissues. J Immunol 2004;173:3310–3319. 80. Lycke N. From toxin to adjuvant: the rational design of a vaccine adjuvant vector, CTA1-DD/ISCOM. Cell Microbiol 2004;6:23–32. 81. De Filette M, Ramne A, Birkett A, Lycke N, Lowenadler B, Min Jou W, Saelens X and Fiers W. The universal influenza vaccine M2e-HBc administered intranasally in combination with the adjuvant CTA1-DD provides complete protection. Vaccine 2006;24:544–551. 82. Haan L, Verweij WR, Holtrop M, Brands R, van Scharrenburg GJ, Palache AM, Agsteribbe E and Wilschut J. Nasal or intramuscular immunization of mice with influenza subunit antigen and the B subunit of E. coli heat-labile toxin induces IgA- or IgG-mediated protective mucosal immunity. Vaccine 2001;19:2898–2907. 83. Adamsson J, Lindblad M, Lundqvist A, Kelly D, Holmgren J and Harandi AM. Novel immunostimulatory agent based on CpG oligodeoxynucleotide linked to the nontoxic B subunit of cholera toxin. J Immunol 2006;176:4902–4913. 84. De Magistris MT. Zonula occludens toxin as a new promising adjuvant for mucosal vaccines. Vaccine 2006;24(Suppl 2):S2-60-1. 85. De Magistris MT. Mucosal delivery of vaccine antigens and its advantages in pediatrics. Adv Drug Deliv Rev 2006;58:52–67. 86. Villa P and Ghezzi P. Animal models of endotoxic shock. Methods Mol Med 2004;98:199–206. 87. Gustafson GL and Rhodes MJ. Bacterial cell wall products as adjuvants: early interferon gamma as a marker for adjuvants that enhance protective immunity. Res Immunol 1992;143:483–488 discussion 573–574. 88. Okemoto K, Kawasaki K, Hanada K, Miura M and Nishijima M. A potent adjuvant monophosphoryl lipid A triggers various immune responses, but not secretion of IL-1beta or activation of caspase-1. J Immunol 2006;176:1203–1208.

93 89. Li P, Allen H, Banerjee S, Franklin S, Herzog L, Johnston C, McDowell J, Paskind M, Rodman L, Salfeld J, et al. Mice deficient in IL-1 beta-converting enzyme are defective in production of mature IL-1 beta and resistant to endotoxic shock. Cell 1995;80:401–411. 90. Durrant LG and Spendlove I. Cancer vaccines entering Phase III clinical trials. Expert Opin Emerg Drugs 2003;8:489–500. 91. Harper DM, Franco EL, Wheeler C, Ferris DG, Jenkins D, Schuind A, Zahaf T, Innis B, Naud P, De Carvalho NS, Roteli-Martins CM, Teixeira J, Blatter MM, Korn AP, Quint W and Dubin G. Efficacy of a bivalent L1 virus-like particle vaccine in prevention of infection with human papillomavirus types 16 and 18 in young women: a randomised controlled trial. Lancet 2004;364:1757–1765. 92. Giannini SL, Hanon E, Moris P, Van Mechelen M, Morel S, Dessy F, Fourneau MA, Colau B, Suzich J, Losonksy G, Martin MT, Dubin G and Wettendorff MA. Enhanced humoral and memory B cellular immunity using HPV16/18 L1 VLP vaccine formulated with the MPL/aluminium salt combination (AS04) compared to aluminium salt only. Vaccine 2006;24:5937–5949. 93. Harper DM, Franco EL, Wheeler CM, Moscicki AB, Romanowski B, Roteli-Martins CM, Jenkins D, Schuind A, Costa Clemens SA and Dubin G. Sustained efficacy up to 4.5 years of a bivalent L1 virus-like particle vaccine against human papillomavirus types 16 and 18: follow-up from a randomised control trial. Lancet 2006;367:1247–1255. 94. Bernstein D. Glycoprotein D adjuvant herpes simplex virus vaccine. Expert Rev Vaccines 2005;4:615–627. 95. Dupont J, Altclas J, Lepetic A, Lombardo M, Vazquez V, Salgueira C, Seigelchifer M, Arndtz N, Antunez E, von Eschen K and Janowicz Z. A controlled clinical trial comparing the safety and immunogenicity of a new adjuvanted hepatitis B vaccine with a standard hepatitis B vaccine. Vaccine 2006;24:7167–7174. 96. Wheeler AW, Marshall JS and Ulrich JT. A Th1-inducing adjuvant, MPL, enhances antibody profiles in experimental animals suggesting it has the potential to improve the efficacy of allergy vaccines. Int Arch Allergy Immunol 2001;126:135–139. 97. Brandtzaeg P, Farstad IN and Haraldsen G. Regional specialization in the mucosal immune system: primed cells do not always home along the same track. Immunol Today 1999;20:267–277. 98. Dell K, Koesters R, Linnebacher M, Klein C and Gissmann L. Intranasal immunization with human papillomavirus type 16 capsomeres in the presence of non-toxic cholera toxin-based adjuvants elicits increased vaginal immunoglobulin levels. Vaccine 2006;24:2238–2247. 99. van Ginkel FW, Jackson RJ, Yuki Y and McGhee JR. Cutting edge: the mucosal adjuvant cholera toxin redirects vaccine proteins into olfactory tissues. J Immunol 2000;165:4778–4782. 100. Fujihashi K, Koga T, van Ginkel FW, Hagiwara Y and McGhee JR. A dilemma for mucosal vaccination: efficacy versus toxicity using enterotoxin-based adjuvants. Vaccine 2002;20:2431–2438. 101. Belyakov IM, Hammond SA, Ahlers JD, Glenn GM and Berzofsky JA. Transcutaneous immunization induces mucosal CTLs and protective immunity by migration of primed skin dendritic cells. J Clin Invest 2004;113:998–1007. 102. Glenn GM and Kenney RT. Mass vaccination: solutions in the skin. Curr Top Microbiol Immunol 2006;304:247–268. 103. Bachmann MF, Rohrer UH, Kundig TM, Burki K, Hengartner H and Zinkernagel RM. The influence of antigen organization on B cell responsiveness. Science 1993; 262:1448–1451.

94 104. Chackerian B, Lowy DR and Schiller JT. Induction of autoantibodies to mouse CCR5 with recombinant papillomavirus particles. Proc Natl Acad Sci USA 1999;96: 2373–2378. 105. Chackerian B, Lenz P, Lowy DR and Schiller JT. Determinants of autoantibody induction by conjugated papillomavirus virus-like particles. J Immunol 2002;169:6120–6126. 106. Jondal M, Schirmbeck R and Reimann J. MHC class I-restricted CTL responses to exogenous antigens. Immunity 1996;5:295–302. 107. Sedlik C, Dridi A, Deriaud E, Saron MF, Rueda P, Sarraseca J, Casal JI and Leclerc C. Intranasal delivery of recombinant parvovirus-like particles elicits cytotoxic T-cell and neutralizing antibody responses. J Virol 1999;73:2739–2744. 108. Storni T, Lechner F, Erdmann I, Bachi T, Jegerlehner A, Dumrese T, Kundig TM, Ruedl C and Bachmann MF. Critical role for activation of antigen-presenting cells in priming of cytotoxic T cell responses after vaccination with virus-like particles. J Immunol 2002;168:2880–2886. 109. Ohlschlager P, Osen W, Dell K, Faath S, Garcea RL, Jochmus I, Muller M, Pawlita M, Schafer K, Sehr P, Staib C, Sutter G and Gissmann L. Human papillomavirus type 16 L1 capsomeres induce L1-specific cytotoxic T lymphocytes and tumor regression in C57BL/6 mice. J Virol 2003;77:4635–4645. 110. Kratz PA, Bottcher B and Nassal M. Native display of complete foreign protein domains on the surface of hepatitis B virus capsids. Proc Natl Acad Sci USA 1999;96:1915–1920. 111. Nassal M, Skamel C, Kratz PA, Wallich R, Stehle T and Simon MM. A fusion product of the complete Borrelia burgdorferi outer surface protein A (OspA) and the hepatitis B virus capsid protein is highly immunogenic and induces protective immunity similar to that seen with an effective lipidated OspA vaccine formula. Eur J Immunol 2005;35:655–665. 112. Ruedl C, Schwarz K, Jegerlehner A, Storni T, Manolova V and Bachmann MF. Viruslike particles as carriers for T-cell epitopes: limited inhibition of T-cell priming by carrier-specific antibodies. J Virol 2005;79:717–724. 113. Niikura M, Takamura S, Kim G, Kawai S, Saijo M, Morikawa S, Kurane I, Li TC, Takeda N and Yasutomi Y. Chimeric recombinant hepatitis E virus-like particles as an oral vaccine vehicle presenting foreign epitopes. Virology 2002;293:273–280. 114. Charpilienne A, Nejmeddine M, Berois M, Parez N, Neumann E, Hewat E, Trugnan G and Cohen J. Individual rotavirus-like particles containing 120 molecules of fluorescent protein are visible in living cells. J Biol Chem 2001;276:29361–29367. 115. Rueda P, Moron G, Sarraseca J, Leclerc C and Casal JI. Influence of flanking sequences on presentation efficiency of a CD8+ cytotoxic T-cell epitope delivered by parvoviruslike particles. J Gen Virol 2004;85:563–572. 116. Mihailova M, Boos M, Petrovskis I, Ose V, Skrastina D, Fiedler M, Sominskaya I, Ross S, Pumpens P, Roggendorf M and Viazov S. Recombinant virus-like particles as a carrier of B- and T-cell epitopes of hepatitis C virus (HCV). Vaccine 2006;24:4369–4377. 117. Sedlik C, Saron M, Sarraseca J, Casal I and Leclerc C. Recombinant parvovirus-like particles as an antigen carrier: a novel nonreplicative exogenous antigen to elicit protective antiviral cytotoxic T cells. Proc Natl Acad Sci USA 1997;94:7503–7508. 118. Walsh G. Biopharmaceutical benchmarks 2006. Nat Biotechnol 2006;24:769–776. 119. Soler E, Parez N, Passet B, Dubuquoy C, Riffault S, Pillot M, Houdebine LM, Schwartz-Cornil I, Recombinant rotavirus inner core proteins produced in the milk of transgenic rabbits confer a high level of protection after intrarectal delivery. Vaccine (in preparation); manuscript number JVAC-D-07-00356R1.

95

Advances in antibody manufacturing using mammalian cells K. John Morrow Jr. Newport Biotechnology Consultants, 625 Washington Avenue, Newport, KY 41071, USA Abstract. In this review, we describe recent advances in antibody processing technology including: (1) development of proprietary cell lines; (2) improved expression systems optimized by selective technologies to boost underperformers; (3) improved protein-free and serum-free culture media; and (4) attention to glycosylation and other post-translational modifications. Advances in computer technology and sophisticated redesign of bioreactors have been major contributors to the dramatic improvements in antibody yields that have been documented in the last decade. Disposable bioreactor components are now widespread, resulting in improved yields, better quality product and lower costs for producers. Downstream innovations include (1) disposable devices for clarification and purification, (2) improved resins and ligands, and (3) new designs of hardware for improved performance. While there are numerous factors contributing to the increased yields that have been obtained, the most sustained of these is the introduction of disposable technologies on both the upstream and the downstream ends of the process. With the continuing introduction of improved computer technology and technological innovation, there is every reason to believe that quality and quantity of antibody products will continue to improve in the coming years, and supply will be adequate to meet the forthcoming needs of the industry. Keywords: recombinant antibodies, proprietary cell lines, protein-free and serum-free culture media, glycosylation, bioreactor design, disposable devices for clarification and purification, affinity ligands, proteins A G, and L, disposable bioreactor components, disposable protein purification devices.

Introduction Antibody processing has matured over the last three decades into a comprehensive discipline including bioengineering, equipment design, molecular biology, cell genetics, cell culture technology, and downstream purification technologies. These advances have been forced by the ever increasing demands for therapeutic antibodies, which has led to rapid expansion of global manufacturing capacity, increasing size of reactors (up to 20,000 L) and a drive for improved process efficiency to reduce manufacturing costs. The current dominance of drug development by antibodies is evidenced by the many approvals by the FDA in the last decade (see Table 1). Every point along the way, from production of the protein to the purification of the final product is under intense scrutiny, since small, incremental improvements represent large gains in production when taken in the aggregate. This strategy has produced impressive gains in the upstream part of the Corresponding author: Tel: 1-513-237-3303.

E-mail: [email protected]; Newportbiotech.com (K.J. Morrow Jr.). BIOTECHNOLOGY ANNUAL REVIEW VOLUME 13 ISSN 1387-2656 DOI: 10.1016/S1387-2656(07)13005-2


96 Table 1. Therapeutic antibodies approved in the USA and EU. Trade name

Company

Disease state

Date approved

Orthoclone OKT3 Reopro

J&J

Graft rejection

6/19/86 USA

Centocor

Reducing ischemic complications in PCI patients Colorectal cancer

12/22/94 USA

Panorex Rituxin

Centocor/Glaxo Smith Kline Genentech

Follicular, CD20positive, B-cell nonHodgkin’s and lowgrade, CD20positive, B-cell nonHodgkin’s lymphoma Organ rejection in kidney transplant patients Organ rejection in kidney transplant patients

Zenapax

Hoffman-La Roche

Simulect

Novartis

Synagis

Medimmune

Treatment of RSV

Remicade

Centocor

Anti-inflammatory

Herceptin

Genentech

Mylotarg Campath-1H

Wyeth–Ayerst Laboratories Genzyme

Zevalin

Biogen IDEC

Humira

Abbott

Xolair

Genentech

Metastatic breastcancer expressing Her-2 protein Relapsed AML patients Multiple sclerosis, organ transplant rejection and several types of leukemia Relapsed or refractory low-grade, follicular, or transformed B-cell non-Hodgkin’s lymphoma (NHL), Rheumatoid arthritis psoriatic arthritis ankylosing spondylitis Allergic asthma symptoms

95 (Germany only) 11/26/97 USA 6/2/98 EU

12/10/97 USA 10/9/98 EU 5/12/98 USA 10/9/98 6/19/98 8/13/99 8/24/98 8/13/99 9/25/98 8/28/00

EU USA EU USA EU USA EU

5/1/00 USA 11/00 EU 5/7/01 USA 7/6/01 EU 2/19/02 1/16/04

12/31/02 USA 9/1/03 EU 6/20/03 USA 10/05 EU

97 Table 1 (Continued ) Trade name

Company

Disease state

Date approved

Bexxar

Corixa

6/27/03 USA 2/03 EU

Raptiva

Genentech

Radioimmunotherapy for non-Hodgkin’s lymphoma Psoriasis

Erbitux

Imclone

Avastin

Genentech

Tysabria

Biogen Idec

Lucentis

Genentech Novartis

Vectibix

Amgen

EGFR-expressing, metastatic colorectal carcinoma EGFR positive Metastic colorectal cancer Relapsing forms of multiple sclerosis Wet age-related macular degeneration (AMD) EGFr expressing metastatic colorectal cancer

12/27/03 USA 9/20/04 EU 2/12/04 USA 6/29/04 EU 2/26/04 USA 1/12/05 EU 6/5/06 USA 6/29/06 EU 6/30/06 USA 3/2/06 submitted EU 9/27/2006 USA

Source: Modified from Das and Morrow KJ. Progress in Antibody Therapeutics. American Biotechnology Laboratory, 2006, pp. 8–10, and reprinted with permission. a Initially approved by FDA in 2004, but withdrawn from the market in 2005 due to adverse incidence in a few patients including one fatality, but subsequently re-approved in 2006.

process, where productivity of cell cultures has improved by orders of magnitude in the past 15 years. While various alternative production platforms are the subject of much investigation, all the approved antibodies currently employed in therapy are produced by mammalian cell cultures. This is the result of (1) improvements in productivity which have made alternatives such as bacteria, yeast and transgenic plants, and animals less appealing; (2) challenges of regulatory approval; and (3) the complexity of proper folding and appropriate post translational modifications in non-mammalian systems, which have yet to be resolved. It seems that biotech companies are still reluctant to abandon a proven winner in the race. Given the cost of changing over and validation, it is unlikely that this decision will be reversed for any of the therapeutic antibodies that are presently on the market. But this success at the upstream end of the train has resulted in a bottleneck at the downstream end, with the result that companies are focusing much of their effort toward improving this component of the overall process.

98 Whereas there was pessimism in some quarters that improvements at the downstream end could not stay abreast with the upstream component of the production picture, a surge of new advances in chemistry, bioengineering, and adaptation of new disposable purification technologies have put this view to rest. In this review I will describe these recent developments in both the upstream and downstream workings of monoclonal antibody production and suggest where the technology is headed in the future. Upstream considerations Mammalian cell culture expression systems Cell lines: manipulated for better performance In recent years, much of the basic science behind variations in protein synthetic rates in mammalian systems has been clarified. Freimuth [1] have reviewed the molecular interventions that allow greatly elevated recombinant protein production in mammalian cells. These include a variety of new expression vectors, engineering of more productive integration sites on the chromosome, and genetic enhancements to suppress apoptosis and augment proper folding and post-translational modifications. For instance, Barnes et al. [2] investigated the GS-NS0 mammalian expression system noting that it produces large amounts of protein from relatively few copies of recombinant genes. One strategy for increasing yields is by the engineering (genetic manipulation) of mammalian cell lines to generate greater levels of the target antibodies. This approach can be applied to both conventional cell lines (such as CHO; Chinese hamster ovary) as well as other cell types, including differentiated cell lines, such as long-term cultures of hematopoietic stem cells in bioreactors [3]. Cell lines: varieties. Antibody production in mammalian cell lines was initially carried out in myeloma lines derived from the NSO line, a mouse mineral oil induced plasmacytoma. These lines grow as suspension cultures, allowing cultivation in bioreactors. Over the years, these original strains have been extensively manipulated as the demands of the bioprocessing industry called for greater levels of productivity. Other cell lines, such as CHO were transfected with antibody genes and altered critically to improve their performance and growth properties in bioreactors (Table 2). Selection for overproducers. There are a variety of approaches by which antibody production can be optimized in cell lines. A conventional strategy involved the isolation and testing of many independent clones, but this approach is labor intensive, ties up resources and frequently fails to achieve the desired outcome [4].

99 Table 2. Properties of some antibody producing cell lines in common use today. Property under consideration

Productivity Product quality Impurities Ease of manufacture Economics Time to clinic Intellectual property Regulatory issues

Cell line CHO

Nonengineered Hybridomas

NSO

Sp 20

PER.C6

BHK

+++ ++ ++ +++

+ + +

+++ + + ++

++ + + ++

++ +++ ++ ++

+ + ++ ++

++ + +

+++ ++

++ +

++ + +

+ +

+ + +

++

++

+

+++

+++ ++

Notes: +++ ¼ favorable; – ¼ unfavorable.

Birch and his collaborators used selective procedures to obtain high productivity, stability, and growth in suspension culture using a chemically defined, animal component-free medium. The resultant lines were capable of performing appropriate post-translational modification of the antibodies, especially glycosylation. A number of enzymes such as glutamine synthetase have been adapted to selection protocols by coupling a strong promoter, next to which the target protein gene is inserted. Stringent and rapid selection of highly productive cells is possible without the need for gene amplification. It is used in combination with a variant of the CHO cell line, CHOKISV, which grows to high cell population densities in chemically defined medium, in suspension culture. Li and his colleagues [5] have improved expression of antibodies through the use of redesigned bicistronic cassettes encoded on two plasmids. Since there are few studies comparing types of vector design that are optimal for stable or transient production of recombinant IgG, they examined different expression cassette designs, including monocistronic and bicistronic expression with different promoters and cistron arrangements. Antibody production levels were evaluated in transiently transfected 293T and CHO-K1 cells. Compared to monocistronic expression, bicistronic expression constructs yielded similar antibody expression levels and showed long-term stability in CHO cell lines. In addition, Birch and his colleagues have investigated potential bottlenecks for productivity in the cell including the role of protein translation and secretory pathways where protein production may be bottlenecked at folding and assembly steps. Proteomic studies indicate that high productivity is correlated with increased expression of a range of proteins involved not just in protein secretion but in a number of other cellular activities as well [6].

100 Finally, the achievement of viable cell concentrations at very high densities is essential to reach maximum protein output. However in crowded cultures, apoptotic pathways may be activated by adverse environmental triggers, causing the population to crash. Once proteolytic caspases are activated, the cell is committed to a one-way route to self-destruction [7]. To prevent this outcome, resistance to apoptosis can be engineered into the cell lines. Frequently, anti- and pro-apoptotic genes belonging to the Bcl-2 family have been exploited in this regard. Lim et al. [8] have shown that RNAi suppression of the pro-apoptotic proteins Bax and Bak enhances viability in fedbatch cultures of CHO cells. In recent years, there has been much consideration given to scaffold/ matrix attachment regions (S/MARs), since these genetic elements could be used to stabilize and augment mammalian gene expression [9]. S/MARs protect transgenes from external repressive genes. They are essential for structural organization of the nuclear chromatin and anchor chromatin loop domains [10]. Papapetrou et al. [11] have studied the ability of S/MARs to function as vehicles for gene therapy applications. They employed the vector, pEPI-eGFP, in gene transfer protocols in hematopoietic progenitor cell lines and primary human cells. This S/MAR-based plasmid was episomally maintained and conferred sustained green fluorescent protein (eGFP) expression in the human cell line, K562, as well as in primary human fibroblast-like cells. However, in a murine erythroleukemia cell line, transgene expression was silenced through histone deacetylation, despite the vector’s episomal persistence. These data demonstrate that S/MAR-based plasmids can function as stable episomes in primary human cells, albeit that, they do not behave in this fashion in all cell types. Cell culture media Media formulation has been addressed by numerous academic and commercial investigators over the years. Classically, the cultivation of animal tissues ex vivo goes back to the turn of the 20th century, but modern investigations began in the 1950s as investigators sought to recreate an artificial animal serum environment. These strategies became more and more sophisticated over the years, as serum free formulations were introduced, by optimizing amino acid and vitamin levels one at a time. Since animal serum contains so many trace contaminants, it is not surprising that its presence was essential in early, crude media formulations. Today there are many commercial and peerreviewed media formulations available, but there are yet many shortcomings and concerns that need to be taken into account. Optimization strategies. Schroder et al. [12] have introduced a modified formulation for cultivation of CHO cell lines in both a serum- and protein-free variety that supports excellent growth of recombinant cell lines expressing

101 various amounts of human antithrombin ATIII (ATIII) to serum-free conditions. Xcellerex (Marlborough, MA) has developed technology to create an inhouse, defined medium formulation comparable in growth and productivity to the best commercially available proprietary growth medium formulation. The platform employs high-throughput screening, allowing very large numbers of sample combinations to be evaluated. The technology was designed for increasing antibody production, but can be adapted for transfected cell lines producing other engineered proteins. The high-throughput approach gives the best chance of finding the process optima within timelines that are significantly shorter than companies typically spend on process development. Pham et al. [13] evaluated the effect of nine protein hydrolysates on cell proliferation, transfection efficiency, and volumetric productivity using green fluorescent protein and human placental secreted alkaline phosphate as reporter genes. The authors produced over 15 r-proteins using this serum-free system and purified them to 495% purity using immobilized metal affinity chromatography. The use of a serum-free adapted cell line grown with the peptone-enriched serum-free medium significantly improves protein expression and purification efficiency compared to serum-containing medium. However, the advantages of economy of the simple, protein-free medium described here must be weighted against the consideration that the peptone mixture may introduce unknown trace materials into the system, and a natural product such as peptone will always display variability from lot to lot. In addition to optimization strategies involving amino acids, protein supplements and other metabolites, there is a substantial body of data supporting the use of sodium propionate and sodium butyrate for the enhancement of mammalian cell production of recombinant proteins [14]. Chen and Harcum [15] evaluated a range of amino acids in the growth medium of CHO cells as a means of mitigating the negative effects of ammonium, a toxic and inhibitory byproduct of mammalian cell. While little progress has been made in reducing ammonium accumulation, the authors demonstrated that threonine, proline, and glycine additions improved CHO cell growth and recombinant protein levels. Even today, with the availability of automated systems for screening vast numbers of media options, substantial challenges exist. Zhang et al. [16] have investigated a novel function of selenite as an effective carrier for delivery of iron for cell growth and function. When transferrin is eliminated from culture medium, its iron-transporter function is usually replaced by chelators capable of delivering iron for cell respiration and metabolism. Selenium, an essential trace element for cell growth and development, detoxifies free radicals by activating glutathione peroxidase and in the form of selenite it can protect the cells from oxidative damage. Using an iron-selenite compound, they achieved markedly improved cell growth and antibody production. It should be noted that without a knowledge of trace element interactions and their role in cell

102 metabolism, it would have been impossible to predict these results or develop a scheme for the optimization of these molecules. Moreover, trace elements function in many roles, and it may be necessary to follow their interactions as other cellular metabolites are varied. Online sources of information. Falkner et al. [17] have presented an online summary of media options and these investigators have issued warnings concerning the time-consuming and often difficult task of gathering of product information. They stress that information provided by the industry focuses on proprietary products neglecting rival developments and may be outdated and of limited value. They further note that online sources such as Bioresearch Online, BioSupplyNet, Biotech-Register.com, Hum–Molgen, Biocompare – The Buyer0 s Guide for Life Scientists or Google Science Products (all 2004) proved to be incomplete, not specific enough and sometimes even contained false or misleading data. In view of the overwhelming complexity of cellular metabolism and the difficulties posed by medium development, these problems are not surprising. To deal with these shortcomings and failures, Falkner et al. [17] have compiled a data bank of commercially available formulations, searchable for products, applications, cell lines, and manufacturers. It is accessible free of charge in HTML format and as PDF download, and the data are frequently checked and updated. The website also offers a forum for discussion of problems and issues concerning serum-free cell culture. The reader would be cautioned to carefully evaluate any commercial media to confirm their ability to support growth in any given system. Post-translational modifications Protein characterization has advanced substantially in the past 10 years and it is presently economical and convenient to detect extremely subtle modification of antibodies, and correlate these small changes with the biological behavior of these molecules. Because of these advances, regulatory agencies have looked more critically on the role of post-translational modifications in determining antibody performance and comparability, especially glycosylation, as it is known to exert a profound effect on antibody effector function. Glycosylation outlook. A promising approach for the development of more effective therapeutic glycosylated proteins is the concept of protein remodeling using synthetic oligosaccharides and in vitro enzymatic modification. Since glycosylation patterns vary from species to species and from cell line to cell line, engineered cell lines have figured prominently in the development of effective therapeutic products. In addition to the in vitro enzymatic remodeling of glycoproteins, the development of ‘‘knock-out’’ or ‘‘knock-in’’ mammalian cell lines with appropriately engineered glucosyltransferases could serve as an alternative approach to producing large amounts of

103 homogeneous product. These steps include knocking out the alpha(2–3) sialyltransferase and knocking in the alpha(2–6) sialyltransferase gene. This may be a more promising approach than reengineering glycosylation processes in an intact organism, in which disruption of organ systems and survival functions might occur. One example is the completion of terminal sialic acid additions to proteins having insufficient sialic acid. Such incompletely sialylated proteins experience drastically reduced circulatory half-life. By treating these molecules with sialyltransferase, it is possible to produce a properly glycosylated product. The reduction in half-life is the result of terminal galactose residues on the oligosaccharide structure, which allows binding to the asialoglycoprotein receptor, present in the liver. The protection of the terminal group with sialic acid prevents binding and subsequent clearance from the circulation by the liver. The widespread availability of sophisticated technology for the identification of subtle molecular differences, combined with numerous options for the production of correctly glycosylated proteins, provides a mechanism for understanding the role of glycosylation in complex biological processes. Moreover, this understanding guides the development of new genetically engineered therapies and will ensure their optimal performance. It has been demonstrated that an engineered antineuroblastoma antibody could be optimized for its antibody-dependent cellular cytotoxicity by increasing its content of complex, bisected oligosaccharides. It will be of great interest to observe the performance of these optimized antibodies in clinical trials. Given that subtle modifications of culture conditions can alter the glycosylation process, the level of fidelity of glycosylation has been investigated in microcarrier cultures [18]. Microcarriers such as CytoporeTM beads are widely used to increase cell densities and to adapt stationary cultures to suspension. In these investigations, it was observed that there was no substantial modification of glycosylation patterns of beta interferon produced by cells cultivated on microcarrier beads. Cell densities could be achieved that were three fold higher than ordinary suspension cultures, and the yield of interferon, in specific units per milliliter was up to four times as high as the suspension cultures. With improvements in media formulation and better technology for characterization of glycosylation, it has become evident that the presence of serum can affect the glycosylation profiles of proteins. Serrato et al. [19] studied glycosylation profiles of antibodies produced by hybridomas grown in medium containing fetal bovine serum, serum-free medium or chemically defined medium. Capillary electrophoresis analyses demonstrated broad microheterogeneity of glycosylation patterns in all the media, ranging from complex to high mannose and paucimannosidic glycans. Antibodies generated from hybridomas grown in serum-containing medium displayed 26 glycan structures, whereas a lower glycan microheterogeneity was found for cultures in the serum-free and chemically defined media.

104 Given the range and unpredictability of recombination antibody glycosylation, it is evident that these modifications much be carefully monitored in order for therapeutic antibodies to meet comparability levels demanded by regulatory agencies. While it is not clear to what extent variable glycosylation profiles affect the performance of a recombinant antibody, regulatory agencies will certainly push companies to provide the greatest possible level of detail, and the findings described here will only raise the bar higher. Phosphorylation and other modifications. After glycosylation, phosphorylation is the most common post-translational protein modification that occurs in animal cells. However, most phosphorylations occur as a regulatory mechanism and as such are transient, so the phosphate is added and later removed. Phosphorylation is catalyzed by protein kinases and its regulatory effects are profound and widespread throughout the cell. Phosphorylation does not play a role in the regulatory behavior of antibodies, so it has not generated the level of concern brought about by glycosylation. Bioreactors: design and function Computer modeling of bioreactor performance Increasing the production levels of biologics may result in unforeseen problems as bioreactor sizes are scaled-up over orders of magnitude. Among various companies, Amgen (www.amgen.com) has used complex mathematical modeling to simulate patterns of hydrodynamic forces. These predictive models are an invaluable tool for process development and can save time and resources by avoiding unsatisfactory designs. Frequently, the Monte Carlo simulation technique is used to analyze problems of this sort. This approach is a useful means of imitating the randomness inherent in certain industrial applications. Using a sampling process a range of possible outcomes is determined, which can be used to arrive at the likelihood of achieving a certain critical value. These data can allow the investigator to make decisions in the design of antibody production facilities, such as allocation of resources [20]. Previous scale-up models for bioreactor performance were unsatisfactory so Amgen adopted an alternative approach known as Computational Fluid Dynamics taking into account the 3-dimensional dynamic properties of the system under investigation. The modeling results indicate that the flow in the 1 L bioreactor is quite irregular and turbulent kinetic energy is generated only in the region of strong vortexes, near the impeller. While the different bioreactor sizes vary in terms of flow patterns, shear rates, and mixing efficiencies, the overall impact of the hydrodynamic forces to the cells is similar over four orders of magnitude when shear rates and residence times at those shear rates are combined. This is far from an intuitive conclusion, and demonstrates the power of the methodology.

105 The problem of controlling cylindrical tank bioreactor using computer modeling has been considered from a flow dynamics perspective [21]. Simple laminar flows in the vortex breakdown region appear to be a suitable alternative to turbulent spinner flask flows and horizontally oriented rotational flows. Dusting and his colleagues measured vortex breakdown flows using three-dimensional stereoscopic particle image velocimetry, and non-dimensionalized velocity and stress distributions. They determined that regions of stress occur in the vicinity of the impeller and the lower sidewall, while relatively large stresses occurred along the edge of disks impinging with the boundary of the vortex breakdown region. Fed-batch and perfusion culture dominate mammalian cell culture production processes [8]. These investigators studied the economic feasibility of both culture technologies by modeling large-scale production of monoclonal antibodies by both the processes. There was an insignificant difference (3%) between the costs of goods per gram (COG/g) values. When Monte Carlo simulations were used to account for uncertainties in titer and yield, as well as the risks of contamination and filter fouling, the frequency distributions for the output metrics revealed that neither process route offered the best of both NPV or product output. The perfusion option was no longer feasible as it failed to meet the product output criterion and the fed-batch option had a 100% higher reward/risk ratio. Their analysis showed that contamination and fouling in the perfusion option need to be reduced from 10% to 3%, to achieve the higher reward/risk ratio. Disposable bioreactor modules In recent years, disposable modules have become more and more widely accepted throughout the biotechnology community [22]. While permanent, stainless-steel bioreactors were the only available option for many years, disposables are gaining acceptance in the biotech industry because of their convenience and ease of maintenance and operation. Disposable bioreactors are difficult to design and for larger scale the problems are compounded. Ozturk, (personal communications, 2006) and his colleagues at Amgen have carried out investigations comparing large (1,000 L) disposable bioreactors with conventional stainless steel containers. When considered overall, disposable bioreactors have many obvious advantages that are mirrored by opposite disadvantages in the permanent, non-disposable units (see Table 3). Instrumentation for both designs has grown in complexity, and now includes monitors for process parameters, gas controllers and probes and transmitters. This improved instrumentation allows more precise and accurate monitoring and control, leading to better outcome in terms of yield. Ozturk and his co-workers (personal communications) compared performance of a 250 L reactor containing a disposable liner with a conventional bioreactor of the same size. The run used the CHO cell line grown in a chemically defined medium with a fed-batch, multiple bolus feeding protocol.

106 Table 3. Positive and negative features of disposable and reusable bioreactor technology. Pros and cons of disposable and hardwired bioreactors Disposable Pro

Con

Pre-sterile, ready to use Easy set-up Eliminates CIP/SIP Low capital outlay Low validation requirement Increased flexibility New technology Volume limitations

Stainless steel Con

Pro

Hard to clean and maintain Large capital investment Extensive CIP/SIP cycle Extensive validation Expensive installation Accessories also stainless steel Proven technology Scalability virtually unlimited

Source: Reprinted courtesy of Dr. Sadettin Ozturk.

There was no statistically significant difference in cell growth, cell viability, expression level of the product, glucose consumption, or lactate production. HPLC profiles showed identical, overlapping peaks on chromatograms. The major peak was isolated and analyzed by mass spectrometry with comparable results. Isoelectric focusing gels were indistinguishable between products generated by the two systems. These results were encouraging but because a 250 L bioreactor is hardly adequate for large-scale protein production projects, Ozturk and his team initiated a collaboration with HyClone to design a 1,000 L disposable bioreactor for fed-batch processes. Evaluation of the 1,000 L system provided maximum cell densities of ca. 107 cell/mL and levels of expression comparable to that enjoyed with the smaller vessels. The performance of the system allowed successful implementation in the GLP pilot plant and now points the way to production of clinical materials according to cGMP regulations. The principle advantage of the conventional bioreactors is their proven performance. As Ozturk’s group demonstrate, this benefit may fade away as more and more comparability data are generated on disposable systems. Scale-up of suspension cultures is usually accomplished by an increase in reactor volume, however microcarrier particles can increase density without some of the problems of increased shear and oxygen requirement associated with larger culture vessels. Du¨rrschmid et al. [23] used Cytoline macroporous microcarriers (Amersham Bioscience, Uppsala) to scale-up production of arylsulfatase B by a transfected CHO cell line. The system guarantees a stable production environment with no degradation of product quality and allows companies seeking to expand their flexibility and avoid capital outlays for costly hardware. The standardized designs meet the demand for predictable, reliable, and cost effective disposable systems, which can be scaled-up to as much as 2,000 L, although these are not as widely used as those in the 500–1,000 L range.

107 Wave Biotech is one of the largest producers of disposable bioreactor equipment, but other companies, including Lampire Biological Laboratories and New Brunwick Scientific are marketing disposable cell culture systems. The Lampire culture bags are permeable allowing for air exchange, whereas the New Brunswick product uses protein coated disks, permitting very high buildup of cellular mass. With the increased interest in disposable culture vessels, the availability of sensors, mixers, heating devices, and other support equipment has increased dramatically. Yet another option is the disposable shaking bioreactor, composed of polypropylene or transparent polycarbonate [24]. These are round vessels, manufactured by Nalgene, which can be mounted on a standard shaking machine. Such bioreactors can be used to cultivate insect or mammalian cells in large volumes of 20 L or 50 L, and they provide good yields of recombinant proteins. A shaking mode provides much more gentle agitation than that obtained with propeller blades, an important consideration for mammalian cells, which are extremely fragile. It has not been determined whether they can be up-scaled to even larger volumes, comparable to the wave system, since this would require investigation of heat and oxygen transference in the larger vessels. Downstream considerations The downstream component of the antibody purification train from crude protein soup to a final bottled pharmaceutical has traditionally been the most challenging phase of recombinant antibody development. This is because, in the first place, the reagents, such as Protein A, are expensive and substitutes are difficult to produce. Secondly, as volumes are up-scaled from initial pilot programs, larger size columns assume different properties, not necessarily predictable from small-scale experiments. Finally, the chemistry of the resins and support materials used to separate proteins is quite complex and may not be sufficiently understood when options are being investigated. However in the last few years, there has been a serious focus on this issue and many advances have been made, especially in the area of disposable purification systems. Disposable modules may be quite costly, but they more than compensate for their high prices by virtue of their simplicity of use, reliability, low failure rate, and avoidance of cross-contamination. With fewer preparative steps there is a commensurate reduction of process time, floor space, buffer usage, labor cost, and the added benefit of increased manufacturing flexibility. Disposables purification technology Schlaeppi et al. [25] report the development of an integrated large-scale process for continuous and partially automated protein production in the

108 Baculovirus system. Their system employs tangential flow filtration using two custom-made filtration units and automated purification by multidimensional chromatography. Prominently featured are disposable materials (bags, filters, and tubing), automated cleaning cycles and column regeneration. The preparation of the clear cell lysate takes less than 2 h and represents considerable time saving compared to standard cell harvesting and lysis by sonication and ultra-centrifugation. Tangential flow filtration, because the fluid stream is pumped across the membrane rather than forced through it, has the advantage of much lower rates of clogging and blocking of the filter pores. So this allows longer life and greater reliability for the filters. Phillips et al. [26] document the use of disposable membrane adsorbers as an alternative to bead-based chromatography columns for removal of trace impurities in protein purification protocols. They achieved exceptional removal of mammalian viruses, endotoxin, DNA, and host-cell protein through use of their linearly scalable membrane adsorber family. These single use, disposable membrane adsorbers provide the benefits of disposables, including no costly and time-consuming column packing or cleaning validation, in addition to much more efficient use of buffers. The rigid microporous structure of the membrane layers allows for high process flux operation and uniform bed consistency at all processing scales. Sartorius, Millipore, and Pall are among the major corporations offering a range of disposable purification technologies. Sartorius products include filtration, purification, and separation technologies. Prominent items are Sartocell filter modules and Sartoclear filter sheets for cell harvesting and clarification after fermentation and cell culture and Sartobind SingleSep and MultiSep membrane adsorbers, featuring disposable or process scale membrane chromatography technology. Protein purification products are based on membrane adsorber technology including centrifugal devices using metal chelate, ion exchange, C18 and Protein-A functional groups. The company markets spin columns as ready-to-use kits and as stand-alone devices. The various products cover a range of volume capacities and a variety of pore sizes. Sartorius process filtration technology covers additional disposables for media preparation, handling and storage, pre-filtration, sterile filtration, mycoplasma removal, and integrity in all scales. Millipore Corporation offers various disposable devices for use in protein purification. The single-head Acerta Disposable Filling System is an example. It consists of a reusable hardware unit and a disposable, presterilized filling assembly that provides complete product containment. The process of protein extraction includes using a number of buffers. Ultimately, ultrapure proteins are needed because salt contamination can affect the kinetic studies, which often follow purification. Contamination also affects gel electrophoresis; in particular, the presence of the salts has a negative impact on 2D

109 gels. The Centricon filters can also be used as a final step in protein purification, to enhance protein concentration and remove unwanted buffer. Millipore produces a line of disposable protein concentrators for use in protein purification, referred to as Millipore, Microcon, Centriplus, and Centricon. The latter Centrifuge Filter Units are designed for rapid processing of aqueous biological samples in a volume of ca. 20 mL, and can concentrate solutions down to 150 mL. Millipore markets tangential flow filtration systems, designed for filtering larger volumes. In contrast to typical filtration systems, in which fluid is directed toward the membrane with applied pressure in tangential flow filtration systems, fluid is pumped tangentially along the surface of the membrane. At the same time, applied pressure forces the smaller molecules through the membrane, while larger molecules are swept along the surface of the membrane, thus avoiding membrane fouling. Referred to as Pellicon 2 MiniCassettes, they are available with either regenerated cellulose, or polyethersulfone membranes for processing up to 10 L of protein-containing solution. Larger cassette systems are available for processing 250 L or more of solution. Pall Corporation produces a disposable depth filter capsule for pilot through large-scale biotech drug manufacturing for use in clarification and prefiltration of biological solutions. The SUPRAclean depth filter capsule is designed to increase safety by decreasing operator exposure to the product. The capsule design also provides a lower hold-up volume than standard module housings, allowing for greater recovery of the product. In addition to the SUPRAclean capsule, the company markets five other new products for various types of biopharmaceutical manufacturing. The Supors EKV sterilizing grade filter removes contaminants from buffer solutions, tissue culture media, and other process fluids used in drug manufacturing. Also Pall offers disposable manifolded filtration systems. Disposable filter capsules can be pre-bundled as pre-sterilized single use filtration system. The Palltronics Flowstar MUX filter integrity test system is designed for testing of multiple discrete filters in a single-set-up. It can integrity test up to eight filters sequentially for various filter size and test type. Its internal memory capacity provides an electronic audit of test results, programs, user information, and system settings.

New purification tools As the US Food and Drug Administration (FDA) and European Agency for the Evaluation of Medicinal Products (EMEA), have tightened regulations, manufacturers have been forced to improve their production and purification processes for recombinant proteins. A clear movement from the classical sucrose gradient centrifugation approach toward more sophisticated

110 purification methods including tangential flow filtration and liquid chromatography is rapidly taking place [27]. In addition, there is a trend toward the use of membrane affinity absorption technology and away from columns. Zhou and Tressel [28] reviewed the Q membrane adsorber as an alternative to a Q-packed-bed column in a flowthrough mode. Their cost analysis establishes that Q membrane chromatography is a viable alternative to Q column chromatography as a polishing step in a flow-through mode for process-scale antibody production. Because antibody molecules are glycosylated, the bacterial protein A is widely used (in both native and recombinant configurations) in antibody purification, but Protein G and Protein L (which binds to kappa chains) are also reagents of choice. Cossins et al. [29] describe the expression of a kappa light chain antibody fragment in the periplasm of Escherichia coli, its production in a defined medium and its purification using peptostreptococcal protein L-Sepharose affinity columns. This approach is highly effective, but expensive. Because of the astronomical cost of gargantuan quantities of purification reagents such as Proteins A, L, and G, some companies have elected to use MABsorbents A2P (Prometic Biosciences, UK) as a feasible alternative. The compound is composed of a di-substitued phenolic derivative of trichlorotriazine and is commercially available coupled to a 6% cross-linked agarose base matrix. The ligand is thought to mimic the structure of two critical amino acids side chains of Protein A, essential for the complexing of Protein A to the Fc region of the IgG molecule. Francis’ team found that the material could be reused over 100 cycles with no drop in IgG purity. It is important to consider GMP (good manufacturing practices) when developing a purification process and A2P provides a feasible alternative for largescale antibody fragment purification. Whereas naturally occurring bacterial proteins with high affinity for antibodies have served as effective purification aids for many years, proteome screening protocols are now being investigated for their potential to uncover new antibody purification reagents. Platis et al. [30] have discussed the lockand-key motif as a generalized strategy for developing affinity reagents. This structural moiety occurs in subunit interfaces of glutathione S-transferases (GSTs) and constitutes a possible recognition and purification tool. They produced through combinatorial synthesis a 13-membered solid-phase ligand library, employing as a lead ligand the Phe-Trz-X structure, mimicking the LAK motif. 1,3,5-Triazine (Trz) was used as the scaffold for assembly, substituted with different LAK-mimetic amino acids. The library of affinity adsorbents was assessed for its ligand-binding specificity. One LAK-mimetic adsorbent was integrated in a single-step purification protocol for a human monoclonal antibody from spiked corn extract, affording high recovery and purity. The findings demonstrate that the principle of natural recognition found in the lock-and-key motif can be investigated and analyzed using proteomic

111 screening. This approach may open the door to entire new types of synthetic affinity ligands, which could revolutionize the field in terms of utility and cost effectiveness. Conclusions The striking success of therapeutic antibodies in the last decade has forced demand for vast amounts of these materials, as noted on various drug reporting services (http://www.drugresearcher.com/news/ng.asp?id=59935-therapeuticantibodies-realise). This happy circumstance has caused concern that a worldwide shortage of antibody production capability could develop. But these needs have proven in the past to be powerful drivers of innovation in antibody expression technologies. The last two years have seen notable advances in all phases of expression technology, though downstream technology continues to lag behind upstream technology. This imbalance reflects the extreme difficulty encountered on the engineering side in design of purification systems and from the standpoint of the analytical chemist, the challenges in developing the chemical entities needed for rapid and effective separation. A reengineering of the rate limiting components of the protein synthesizing machinery (chaperones and other proteins) promising large-scale increases in bacterial output of transgenic lines [31]. However, the challenge of improving the downstream side of the equation is proving to be extremely difficult. Moreover, the cost of reagents such as protein A is proving to be astronomical, and more cost effective systems are in great demand. Development of synthetic peptide ligands may in the future replace protein A, with attendant cost benefits [32]. It is certain that in the near future, regulatory approval will be gained for plant, animal (including chickens), and insect systems, which will generate competition with the classical bacterial and mammalian protein expression technologies. This will provide a robust rivalry and if we accept classic models of free enterprise, the result will be superior products at lower prices. However, so much progress has been made in mammalian cell protein expression that there seems little chance that alternative systems will overtake the 80–90% dominance of the market currently enjoyed by mammalian cells. Moreover, the rapidity with which a mammalian cell line can be engineered for a new protein production system, combined with the long-standing acceptance of mammalian cells by the FDA, constitute formidable hurdles for any competitor. The predicted shortfall in manufacturing capability has so far not materialized, due to the rise of many small contract manufacturers and because of the construction of several massive mammalian cell facilities by large biotech and pharma companies. Large pharma companies are at present undergoing complex and dramatic changes in their corporate strategy. With huge outlays in unproductive

112 research programs and fewer and fewer innovative small molecules progressing to new products, the pharmaceutical companies are following the biotech model, acquiring companies, buying new products, and outsourcing research programs. Biologicals, especially antibodies have been the most successful new products introduced in recent years, and it appears that this trend will continue. Therefore, powerful incentives will drive the bioprocessing sectors forward on all fronts for the foreseeable future. References 1. 2.

3. 4. 5.

6.

7.

8.

9.

10.

11.

12. 13.

14.

Freimuth P. Protein overexpression in mammalian cell lines. Genet Eng 2007;95–100. Barnes LM, Bentley CM, Moy N and Dickson AJ. Molecular analysis of successful cell line selection in transfected GS-NS0 myeloma cells. Biotechnol Bioeng 2006; Sep 25 [Epub ahead of print]. Kwon J, Kim BS, Kim MJ and Park HW. Suspension culture of hematopoietic stem cells in stirred bioreactors. Biotechnol Lett 2003;25(2):179–182. Birch JR and Racher AJ. Antibody production. Adv Drug Deliv Rev 2006;58:671–685. Li J, Menzel C, Meier D, Zhang C, Dubel S and Jostock T. A comparative study of different vector designs for the mammalian expression of recombinant IgG antibodies. J Immunol Methods 2006; Nov 13 [Epub ahead of print]. Dinnis DM, Stansfield SH, Schlatter S, Smales CM, Alete D, Birch JR, Racher AJ, Marshall CT, Nielsen LK and James DC. Functional proteomic analysis of GS-NS0 murine myeloma cell lines with varying recombinant monoclonal antibody production rate. Biotechnol Bioeng 2006;94(5):830–841. Hale AJ, Smith CA, Sutherland LC, Stoneman VE, Longthorne V, Culhane AC and Williams GT. Apoptosis: molecular regulation of cell death. Eur J Biochem 1996;236(1):1–26. Lim SF, Chuan KH, Liu S, Loh SO, Chung BY, Ong CC and Song Z. RNAi suppression of Bax and Bak enhances viability in fed-batch cultures of CHO cells. Metab Eng 2006;6:509–522. Girod PA, Zahn-Zabal M and Mermod N. Use of the chicken lysozyme 50 matrix attachment region to generate high producer CHO cell lines. Biotechnol Bioeng 2005;91(1):1–11. Tetko IV, Haberer G, Rudd S, Meyers B, Mewes HW and Mayer KF. Spatiotemporal expression control correlates with intragenic scaffold matrix attachment regions (S/MARs) in Arabidopsis thaliana. PloS Comput Biol 2006;2(3):e21. Papapetrou EP, Ziros PG, Micheva ID, Zoumbos NC and Athanassiadou A. Gene transfer into human hematopoietic progenitor cells with an episomal vector carrying an S/MAR element. Gene Ther 2006;13(1):40–51. Schroder M, Matischak K and Friedl P. Serum- and protein-free media formulations for the Chinese hamster ovary cell line DUKXB11. J Biotechnol 2004;108(3):279–292. Pham PL, Perret S, Doan HC, Cass B, St-Laurent G, Kamen A and Durocher Y. Largescale transient transfection of serum-free suspension-growing HEK293 EBNA1 cells: peptone additives improve cell growth and transfection efficiency. Biotechnol Bioeng 2003;84(3):332–342. Chun BH, Park SY, Chung N and Bang WG. Enhanced production of recombinant B-domain deleted factor VIII from Chinese hamster ovary cells by propionic and butyric acids. Biotechnol Lett 2003;25(4):315–319.

113 15. Chen P and Harcum SW. Effects of amino acid additions on ammonium stressed CHO cells. J Biotechnol 2005;117(3):277–286. 16. Zhang J, Robinson D and Salmon P. A novel function for selenium in biological systems: selenite as a highly effective iron carrier for Chinese hamster ovary cell growth and monoclonal antibody production. Biotech Bioeng 2006;95(6):1188–1197. 17. Falkner, E, Appl H, Ecler, C, Losert UM, Schoffl H and Pfaller W. Serum free cell culture: the free access online database. Toxicol In Vitro 2006;20(3):395–400. [Epub 2005 Oct 25]. 18. Spearman M, Rodriguez J, Huzel N and Butler M. Production and glycosylation of recombinant beta-interferon in suspension and cytopore microcarrier cultures of CHO cells. Biotechnol Prog 2005;21(1):31–39. 19. Serrato JA, Hernandez V, Estrada-Mondaca S, Palomares LA and Ramirez OT. Differences in the glycosylation profiles of a monoclonal antiobody produced by hybridomas cultured in serum supplemented, serum free or chemically defined media. Biotechnol App Biochem 2007; Jan 25 [E-Pub ahead of print]. 20. Lim AC, Zhou Y, Washbrook J, Sinclair A, Fish B, Francis R, Titchner-Hooker NJ and Farid SS. Application of a decision-support tool to assess pooling strategies in perfusion culture processes under uncertainity. Biotechnol Prog 2005;21(4):1231–1242. 21. Dusting J, Sheridan J and Hourigan K. A fluid dynamics approach to bioreactor design for cell and tissue culture. Biotechnol Bioeng 2006;94(6):1196–1208. 22. DePalma A. Bright sky for single use bioprocess products. Genet Eng News 2006;26:1. 23. Durrschmid M, Landauer K, Simic G, Bluml G and Doblhoff-Dier O. Scalable inoculation strategies for microcarrier-based animal cell bioprocesses. Biotechnol Bioeng 2003;83(6):681–686. 24. Raval K, Liu CM and Buchs J. Large scale disposable shaking bioreactors. Bioprocess Int 2006;4(1):46–49. 25. Schlaeppi JM, Henke M, Mahnke M, Hartmann S, Schmitz R, Pouliquen Y, Kerins B, Weber E, Kolbinger F and Kocher HP. A semi-automated large scale process for the production of recombinant tagged proteins in the Baculovirus syste. Protein Expr Purif 2006;50(2):185–195. 26. Phillips M, Cormier J, Ferrence J, Dowd C, Kiss R, Lutz H and Carter J. Performance of a membrane absorber for trace impurity removal in biotechnology manufacturing. J Chromatogr A 2005;1078(1–2):74–82. 27. Morenweiser R. Downstream processing of viral vectors and vaccines. Gene Ther 2005;12(Suppl 1):S103–S110. 28. Zhou JX and Tressel T. Basic concepts in Q membrane chromatography for large-scale antibody production. Biotechnol Prog 2006;22(2):341–349. 29. Cossins AJ, Harrison S, Popplewell AG and Gore MG. Recombinant production of a V(L) single domain antibody in E. coli and analysis of its interaction with peptostreptococcal protein L. Protein Expr Purif 2007;51(2):253–259. 30. Platis D, Sotriffer CA, Clonis Y and Labrou NE. Lock-and-key motif as a concept for designing affinity adsorbents for protein purification. J Chromatogr A 2006; 1128(1–2):138–151. 31. Georgiou G and Segatori L. Preparative expression of secreted proteins in bacteria: status report and future prospects. Curr Opin Biotechnol 2005;16:538–545. 32. Roque AC, Lowe CR and Taipa MA. Antibodies and genetically engineered related molecules: production and purification. Biotechnol Prog 2004;20(3):639–654.

This page intentionally left blank

115

Pharming and transgenic plants David Lieńard1, Christophe Sourrouille1, Ve´ronique Gomord and Loı¨ c Faye Universite´ de Rouen, CNRS UMR 6037, IFRMP 23, GDR 2590, Faculte´ des Sciences, Baˆt. Ext. Biologie, 76821 Mont-Saint-Aignan cedex, France Abstract. Plant represented the essence of pharmacopoeia until the beginning of the 19th century when plant-derived pharmaceuticals were partly supplanted by drugs produced by the industrial methods of chemical synthesis. In the last decades, genetic engineering has offered an alternative to chemical synthesis, using bacteria, yeasts and animal cells as factories for the production of therapeutic proteins. More recently, molecular farming has rapidly pushed towards plants among the major players in recombinant protein production systems. Indeed, therapeutic protein production is safe and extremely cost-effective in plants. Unlike microbial fermentation, plants are capable of carrying out post-translational modifications and, unlike production systems based on mammalian cell cultures, plants are devoid of human infective viruses and prions. Furthermore, a large panel of strategies and new plant expression systems are currently developed to improve the plant-made pharmaceutical’s yields and quality. Recent advances in the control of post-translational maturations in transgenic plants will allow them, in the near future, to perform human-like maturations on recombinant proteins and, hence, make plant expression systems suitable alternatives to animal cell factories. Keywords: glycosylation, molecular farming, plant-made pharmaceutical, recombinant protein, transgenic plant, therapeutic protein.

Introduction From 60,000 BC to the 19th century, plants were the main source for human drugs. For instance, when sick and obliged to stay in his cave, the Neanderthal man already used centaury to fight his fever. The first known text on medicinal plants, the Pen Tsao, was written more than 4,500 years ago under the direction of emperor Shen-Nung in China, and describes 365 medicinal plants, including opium, ephedra and hemp. More recently, around 1500 BC, the Ebers papyrus describes 700 remedies made from plants, including mandrake, castor bean and hemp, illustrating that plants had a major place in Egyptian medicine. In the Middle ages, places such as Salagon abbaye became famous for their specialization in the culture of medicinal plants and universities were created in Montpellier or Salerne to improve plant therapeutics, extraction and characterization. There was a great turn in medicament history, starting at the beginning of the 19th century until the early 1970s, when pharmacy turned to be dominated Corresponding author: Tel: 33-2-35-14-66-92. Fax: 33-2-35-14-67-87.

E-mail: [email protected] (L. Faye). David Lienard and Christophe Sourrouille have equal contributions to this work

1

BIOTECHNOLOGY ANNUAL REVIEW VOLUME 13 ISSN 1387-2656 DOI: 10.1016/S1387-2656(07)13006-4


116 by scientific chemistry with both the development of more and more sophisticated processes for extraction, purification and the synthesis of active pharmaceutical compounds. The 20th century became a triumph for drugs produced at an industrial level by chemical synthesis. This evolution probably started with the production of aspirin, a synthetic analogue of salicylic acid previously extracted from willow bark. In parallel, more and more sophisticated extraction and purification procedures were developed resulting, for example, with the first extraction of morphine from poppy in 1815 or extraction of insulin from pig pancreas in 1922. As a complement of synthesis and extraction chemistry, modern biology enters the world of pharmaceutical industry with the development of genetic engineering in the early 1970s, allowing biosynthesis of complex molecules too difficult to extract and purify from living material and inaccessible to synthesis chemistry. In the last decades, genetic engineering has offered an alternative to chemical synthesis and extraction procedures with the production of therapeutic molecules in transgenic bacteria, yeast and animal cells. After a temporary decrease in interest, plants are rapidly moving back into human pharmacopoeia, with the recent development of plant-based recombinant protein production systems offering a safe and extremely cost-effective alternative to microbial and mammalian cell culture. In this short review, we will illustrate that current improvements of plant expression systems for biopharmaceutical production are making them suitable as alternative factories for the production of either simple or highly complex therapeutic proteins. Plants have a high potential for pharmaceutical protein production The need for cheap and efficient production systems emerges as a critical factor in therapeutic protein production. To satisfy the more and more rigorous industrial standards of performance, a heterologous system of production has to fulfill several requirements. Pharmaceutical industry needs large-scale methods using simple and inexpensive purification techniques to obtain recombinant proteins with high-production rates, reproducible quality and for a low cost. Moreover, the system of production must be able to carry out co- and post-translational modifications (PTMs), including signal peptide cleavage, pro-peptide processing, protein folding, disulfide bond formation and glycosylation [1]. Currently, no heterologous expression system of production satisfies all of these requirements. For instance, complex therapeutic proteins produced in prokaryotes are not always properly folded or processed to provide the desired degree of biological activity. Consequently, microbial expression systems have generally been used for the expression of relatively simple therapeutic proteins that do not require folding or extensive post-translational processing to be biologically active such as insulin, interferon or human

117 growth hormone [2]. Due to the limitations of prokaryotes for production of complex therapeutic proteins, the pharmaceutical industry had focused efforts towards optimization of two main eukaryotic expression systems, yeasts and mammalian cell cultures. These production systems, however, suffer from many disadvantages such as inappropriate PTMs for yeast, or high operating costs, difficulties in scaling up to large volumes and potential contamination by virus or prion for cultured mammalian cells. Altogether, the biochemical, technical and economic limitations on existing prokaryotic and eukaryotic expression systems, the growing clinical demand for complex therapeutic proteins and the lack of bioreactor capacity have created substantial interest in developing new expression systems for the production of therapeutic proteins. To that end, plants have emerged in the past decade as a suitable alternative to the current production systems of therapeutic proteins and today their capacity in low-cost production of high quality, much safer and biologically active mammalian proteins is largely documented (for recent reviews see [1–4]). For instance, the use of transgenic plants could be a solution to the need for a rapid increase in production capacity of therapeutic antibodies. Indeed, even with relatively low expression levels for therapeutic proteins [5,6], the production capacity of recombinant antibodies in transgenic plants is almost unlimited, as it only depends on the surface dedicated to the plant culture. A plant ‘‘bioreactor’’ will allow the production of recombinant proteins up to 20 kg/ha, regardless of the plant material considered: tobacco, corn, soybean or alfalfa [7,8]. Another major advantage of transgenic plants over other production systems available for large-scale and low-cost production, such as E. coli or yeasts, is their ability to perform most PTMs required for therapeutic protein’s bioactivity and pharmacokinetics [4,9]. This is illustrated from their capacity to produce complex functional mammalian proteins including plasma proteins, antigens, growth factors, hormones, cytokines, enzymes and antibodies (Tables 1 and 2). The vast majority of therapeutic proteins undergoes several PTMs, which are the final steps in which genetic information from a gene directs the formation of a functional gene product. The term PTM covers covalent modifications of individual amino-acid residues (e.g., glycosylation, phosphorylation, methylation, ADP–ribosylation, oxidation and glycation); proteolytic processing and non-enzymatic modifications, such as deamidation and racemization. Most therapeutic proteins require at least proteolytic cleavage(s), oligomerization and glycosylation for their bioactivity, pharmacokinetics, stability and solubility. The production of immunoglobulins in plant cells is a good illustration of plant capacity to produce complex proteins. Indeed, transgenic plant cells are able to correctly synthesize, mature and assemble, via disulfide bridges, the light and heavy polypeptide chains constitutive of an antibody. Since the first production of a functional antibody in plant [5], many antibodies, or

118 Table 1. Therapeutic proteins produced in plants. Products

Proteins

Transgenic plants

References

Blood and plasma proteins

Albumin Aprotinin Collagen I Encephalin Hemoglobin Human a1 antitrypsin Bet v 1 Cholera toxin B subunit Glycoprotein B from human cytomegalovirus (CMV) Cholera toxin B subunit-insulin fusion protein D2 peptide of fibronectin binding protein B of S. aureus VP1

Potato, tobacco Maize Tobacco Tobacco Tobacco Rice Tobacco Potato

[10–12] [13] [14] [11] [15–17] [18,19] [20] [21]

Tobacco

[22]

Potato

[23]

Black bean

[24]

Medicago sativa, Black bean Arabidopsis thaliana Medicago sativa Tobacco Tobacco and Potato Soybean Potato, tobacco

[25–27]

Potato

[38]

Tobacco

[39]

Tobacco, potato Tobacco, spinach, tomato Potato Tobacco, candy cane

[40,41] [42]

Tobacco

[47]

Vaccines

VP2 VP4 Hemagglutinin Hepatitis antigen gp41 glycoprotein Enterotoxine B of E. coli Cholera toxine B of V. cholera Epitope of P. falciparum Norwalk virus capsid G protein of rabies virus Hormones, cytokins and growth factors

Autoantige`ne GM-CSF (Granulocyte Macrophage-Colony Stimulating Factor) Interferon b

[28] [29] [30,31] [32,33–35] [36] [37]

[33] [43–46]

119 Table 1 (Continued ) Products

Enzymes

Others

Proteins

Transgenic plants

References

Interferon a Interferon g Somatotropin hGH (human Growth Hormon) Erythropoitin Epidermal Growth Factor (EGF) Vascular endothelial growth factor (VEGF) Interleukin 2 Human interleukin 6 Interleukin 10 Interleukin 12 Insulin like Growth factor (IGF)

Tobacco Rice Tobacco (chloroplasts)

[48] [49] [50]

Tobacco (cells) Tobacco

[51] [11]

Moss

[52]

Potato Tobacco Tobacco Tobacco Rice

[53] [54] [55] [56] [57]

Tobacco Tobacco, tomato

[58]

Tobacco

[59]

Tobacco (plant and seed) Tobacco Tomato

[59,60,61]

Tobacco Tobacco

[64] [65]

Tobacco, Colza Tobacco Tobacco, rice

[66,67]

Converting enzyme of angiotensin Protein c (seric proteas) Glucocerebrosidase Alpha-trichosantin Human acetylcholinesterase Dog gastric lipase Human transglutaminase Hirudin Endostatin Human lactoferrin

[62] [63]

[68] [69–71]

antibody fragments, have been produced for therapeutic or diagnostic purposes in various plant expression systems (Table 2). Antibodies produced in plants are correctly assembled, proteolytically matured and glycosylated. Indeed, antibodies produced in plants bear both high-mannose and biantennary complex type N-glycans [72,73]. The highmannose-type N-glycans have the same structure in plant and mammalian glycoproteins. But complex-type N-glycans are structurally different in plants and mammals.

120

Table 2. Recombinant antibodies produced in transgenic plants. Adapted from [1]. Antigen

Type of antibodies

Indications

Transformed plant

Targeting signal

Reference

Human carcinoembryonic antigen

Mouse/human chimeric IgG1 antibody (cT84.66) scFv T84.66

Antibody-mediated cancer therapy (colon cancer, breast cancer and tumour with epithelial origin)

N. tabacum cv petit Havana SR1 (transient expression)

MSP [secreted] MSP+KDEL [ER] MSP [secreted] Plant codon optimized SP [secreted] SP+KDEL [ER] MSP [secreted] MSP+KDEL [ER]

[74]

MSP+KDEL [ER]

[77]

N. tabacum cv petit Havana SR1

T84.66/G68 diabody

scFvT84.66 Rabies virus protein Human IgG monoclonal antibody

Streptococcal surface antigen SAI/II Colon cancer surface antigen

Monoclonal antibody (mAb SO57) C5-1 IgG

Guy’s 13 IgG IgA/G slgA/G CO-17 A IgG

Rabies virus neutralization

Triticum aestivum L. cv bobwhite O. sativa L. Indica cv M12 and Bengal N. tabacum cv Xanthi

[75] [76]

Anti-human globulin reagent for phenotyping and cross-matching red blood cells of receivers and donors Tooth decay

M. sativa

MSP [secreted]

[7]

N. tabacum

MSP [secreted]

[78]

Antibody-mediated cancer therapy

N. benthamiana

MSP [secreted]

[79] [80]

Immunoprotection against genital herpes and transmission of HSV vaginal

O. sativa

[81]

G. max

[82]

MSP+KDEL [ER] HSV-2, protein from herpes simplex virus (HSV)

Zearalenone (mycotoxin)

IgG, IgA, DigA or slgA

IgG1 Fab and F(ab’)2 scfv

Passive humanization of animals in their feed

A. thaliana (ecotype Columbia)

No SP [cytosol] Plant PR1-b SP

[83]

Human creatine kinase-MM

MAK33 IgG1

Fab fragment

MAK33 scFv

MAK33 Fab fragment

Cardiac disease, mitochondrial disorders, inflammatory myopathies, myasthenia, polymyositis, McArdle’s disease NMJ disorders, muscular dystrophy, ALS, hypoand hyperthyroid disorders Central core disease, acid maltase deficiency, myoglobinuria, rhabdomyolysis, motor

A. thaliana

N. tabacum

N. tabacum SR1

Neurone diseases, rheumatic diseases, and others that create elevated or reduced levels of creatine kinases B-cell lymphoma treatment

A. thaliana

Chlamydomonas reinhardtii

Tumour’s surface Ig

38C13 scFv

Herpes simplex virus (HSV) glycoprotein D Hepatitis B virus

HSV8 lsc (large single chain of IgA) scFv

Antibody-mediated herpes therapy Immunoaffinity purification of recombinant HBsAg

Human CD30

mAb

Treatment of Hodgkin lymphoma and anaplastic large cell lymphoma

A. thaliana 2S2 [seed storage protein SP]

N. benthamiana

N. tabacum cv petit Havana SR1

Lemna minor

[84]

[85]

No SP [cytosol]

A. thaliana 2S2 [seed storage protein SP] A. thaliana 2S2 [seed storage protein SP]

Rice a-amylase SP [secreted] [Chloroplastic transformation] No SP [cytosol] Sweet potato Sporamin SP [secreted] Sporamin SP+PP [PSV] Sporamin SP+KDEL [ER]

[86]

[87]

[88] [89] [90]

[91]

KDEL, endoplasmic reticulum retention signal; MSP, murine signal peptide; SP, signal peptide; PP, propeptide; symbol [ ], organs or subcellular compartments targeting of the recombinant protein.

121

122 Despite these differences in the N-glycan structures, antibodies produced in plants have similar antigen-binding capacity as their homologs produced in mammalian cells. Furthermore, an antibody half-life in the bloodstream as well as its ability to be recognized by Fc receptors, which are both determined by heavy chains N-glycosylation are not strongly affected when a plantN-glycan is present instead of a mammalian N-glycan [7,12,92]. Which are the current limitations of the plant expression system? Current limitations of plant expression systems are the low yields observed for some therapeutic proteins and the impact of non-mammalian glycosylation on the activity, immunogenicity and allergenicity of glycosylated plant made pharmaceuticals (PMPs). To achieve higher yields, different stages of therapeutic protein expression in plants can be optimized from transcription to protein stability. Adaptation of codon usage Beside a long quest for stronger constitutive or inducible promoters, with up to now little success, it seems that in many cases where low therapeutic protein expression levels are observed, adaptation of codon usage could increase the yields. The genetic code which defines a mapping between amino acids and nucleotides is redundant: 18 of the 20 amino acids are coded by several codons known as synonyms. Most of the time, organisms use synonymous codons in a non-random way, i.e., some synonymous code being more frequently used than others. Codon usage is thus biased, and this bias varies according to the species and genes within a same species [93]. Thus each organism has a subset of synonymous code mainly used: the codon bias. The codon bias is adapted to the abundance of ARNt for a more effective translation of the ARNm. Among different ARNt iso-acceptors for one amino acid, one is more abundant, this is the main ARNt. The use of the main ARNt by a gene allows its faster translation rate and higher fidelity [94]. According to this model, the more a gene is expressed, the more it undergoes the selection for the effectiveness of his translation. This brings to a preferential use of optimal codon, causes a strong bias in the use of synonymous codons and results in a low number of synonymous substitutions per site [95]. So, in order to improve the rate and fidelity of translation in a plant expression system, depending on its origin, it can be important to adapt the coding sequence of the gene of interest to the codon bias of the host plant. Few information are yet available in plants, but a 5–100 times increase in protein expression was observed after codon optimization [96–99]. As an example, expression in tobacco and tomato of a bacterial insecticide gene (cryIA) partially modified and (3% nucleotide difference) fully modified (21% nucleotide difference) were compared. Plants transformed with either

123 the partially or fully codon optimized gene respectively expressed 10 and 100 times more insecticidal protein than plants transformed with the wildtype gene [96]. Suppressor of RNA silencing RNA silencing was discovered by Andrew Hamilton and David Baulcombe [100]. This is an evolutionarily conserved control system that occurs in many eukaryotic organisms, and was named RNA interference in animals, quelling in fungi and post-transcriptional gene silencing (PTGS) in plants. RNA silencing plays a key antiviral defence role by influencing virus replication in cells. Viruses, in turn, produce proteins capable of suppressing host cell RNA silencing [101,102]. This mechanism of defence can be initiated not only by the presence of virus RNA but also by the presence of exogenous genes. In transformed plants, RNA silencing is targeted against transcripts of the transgene and any similar endogenous genes, so that corresponding gene products accumulate at a low level [103]. This phenomenon can be avoided by expressing simultaneously the gene of interest and a suppressor of silencing. This provides a new tool for molecular farming in plants to obtain high-level expression of transgenes. Each plant-virus seems to produce its own suppressor of silencing and the characterization of a large number of suppressors such as HC-Pro, 2b, p25 is currently in progress. Today, the better-characterized suppressor is the p19 protein, encoded by Tomato Bushy Stunt Virus (TBSV) [104]. This suppressor was co-expressed with recombinant proteins and it was shown to dramatically enhance expression of a broad range of these proteins, allowing up to 50-fold increase in yield [103,105]. Targeted expression of recombinant proteins Targeted expression of PMPs into specific organs and subcellular compartments represent a plant-specific strategy to increase yields and simplify the first steps of purification. In this way, different plant organs (leaves, seeds, root) and plant cell compartments (endoplasmic reticulum, chloroplast, vacuole and oil body) have been efficiently used to express many therapeutic proteins (Tables 2 and 3) [106,107]. Generally, recombinant proteins are targeted into plant organs, which allows high-biomass yield. For example, in plants with large foliage volume such as tobacco, alfalfa and some other legume plants, expression is performed in leaves, whereas, in potato, corn, rapeseed, safflower, soybean, wheat or rice, the production and accumulation of recombinant proteins occur in tubers or in seeds [108,109]. Both systems have their own advantages and drawbacks. Leaves present an active and complex metabolism which

124

Table 3. Examples of plant-based expression systems used for pharmaceutical protein production [110]. System

Protein

Expression

Companya

References

Stable nuclear transformation systems Whole plant (cytosolic)

HbsAg, vaccine

0.007% TSP

[111]

Collagen

1 mg/g DW

scFv, hepatitis B

0.032% TSP

AltaGen Bioscience Inc. (potato) CropTech Corp. (tobacco) Medicago Inc. (alfalfa) Meristem Therapeutics (tobacco) PlantGenix Inc. (not reported)

ER

sIgA/G scFv, cutinase

Not reported 1% TSP

Apoplast

scFv, T84.66 scFv, ABA IgG1

29 mg/g FW 6.8% TSP 1.3% ISP

IgA/G IgG1, Fab avidin hirudin

500 mg/g FW 13% ISP 6% TSP 1% FW

Cellular compartment

Tissue-specificity

[14]

[90]

Vacuole

Seed

Novoplant GmbH (tobacco) Epicyte Pharmaceutical Inc. (tobacco) ProdiGene Inc. (corn) SemBioSys Genetics Inc. (canola)

[112] [113] [114] [115] [5] [6] [84] [116] [66]

Tuber

scFv, oxalozone

2% TSP

Root Fruit

IgM, RKN RSV-F protein human SEAP human SEAP

0.003% TSP not reported 20 mg/g DW/day 2.8% TEP

somatotropin

7% TSP

a-trichosanthin

2% TSP

Exudate

Stable plastid transformation system Chloroplast Transient transformation system Viral

Applied Phytologics Inc. (rice) Epicyte Pharmaceutical Inc. (corn) IPT, Monsanto (corn) Meristem Therapeutics (rape) Meristem Therapeutics (potato) Phytomedics Inc. (tobacco roots) Phytomedics Inc. (tobacco leaves) Biolex Inc. (duckweed)

[117] [118] [119] [120] [121]

[122] Large Scale Biology Corp. (tobacco)

[62]

DW, dry weight; ER, endoplasmic reticulum; FW, fresh weight; ISP, intercellular soluble protein; TEP, total exuded protein; TSP, total soluble protein. a Companies sharing a row with a protein and reference are sources of this information.

125

126 offers many possibilities, but they also contain significant protease activities limiting the accumulation of some PMPs [1]. The low-water content of the seeds allows accumulation of recombinant proteins into compact biomass site for long periods of time at relatively high concentrations [123]. Beside organ-specific storage of PMPs, many subcellular compartments are available for accumulation of large amounts of recombinant therapeutic proteins and thus greatly simplifying their purification. Targeted expression of therapeutic proteins into the secretory pathway In plants and animals, the endoplasmic reticulum (ER) compartment allows entry of proteins into the secretory pathway and ensures folding and correct assembly of newly synthesized secretory and resident proteins [124]. Most recombinant proteins produced so far in plants have been secreted into the intercellular space or apoplast [125,126]. This targeting is only dependent on the presence of an N-terminal signal peptide cleaved during the co-translational insertion of the nascent protein in the ER [1] (see Table 2). It has been shown in many plant expression systems and for many PMPs that plant and human signal peptides are recognized with the same efficiency. Interestingly, recombinant proteins targeted to the secretory pathway, can be secreted by the roots in the culture medium (rhizosecretion) and it was shown that these proteins were accumulated in this medium in higher amounts than in the root tissues [120,127]. This technology, which avoids cropping, brings a great simplification to the purification process [128]. Recently, multimeric proteins such as immunoglobulins have been produced in their active form by rhizosecretion in transgenic tobacco [129,130]. While soluble protein secretion in the extracellular compartment is a default pathway only depending on the presence of a signal peptide, targeting to other compartments of the secretory pathway such as ER or vacuoles, needs additional signals. Many examples illustrate that the H/KDEL-mediated protein retention in the ER could strongly increase the stability and consequently the yield of recombinant proteins as compared with secretion [32,77,87,90,131–136]. For instance, accumulation level of the pea vacuolar storage protein, vicilin, was increased by up to 100-fold in transgenic alfalfa leaves when the ER-retention signal, KDEL, was fused to its C-terminus [137]. Likewise, fusing the ER-retention signal HDEL in C-terminal onto sporamin, a storage protein from sweet potato showing antitrypsin activity, significantly increased its accumulation level, presumably by preventing its progression to the vacuole [138]. Interestingly, as developed below, retention into the ER also prevents addition of complex N-glycans on plant-made proteins, which are potentially immunogenic in humans. The protein storage vacuole (PSV) is an intracellular organelle where proteins are stored in seed cells and also in many different types of plant cells, including leaf and root cells [139,140]. Compared to vegetative vacuoles, seed PSVs exhibit slightly higher pH and lower hydrolytic activity. So that, the

127 PSV is an attractive compartment for recombinant protein accumulation [141]. For example, when expressed in the endosperm of rice seeds, human lysozyme (a naturally secreted protein) was stored under a biologically active form in PSVs [142–144]. Another secreted protein; human serum albumin has been expressed and delivered into the PSVs of wheat endosperm where it shows a good stability [145]. In the objective of a better exploitation of PSVs as a storage compartment for therapeutic proteins, a better knowledge of the signals and mechanisms responsible for protein targeting to these organelles will help further investigation on the advantages and limitations of storage in PSVs [146]. Oilseeds accumulate lipids to supply the energy required for seedling development in organelles arising from the ER: the oilbodies. Seed oilbodies are limited by a protein-rich phospholipids monolayer. Oleosins, the major proteins at the periphery of oilbody membrane, are anchored by their hydrophobic domain exposing their N- and C-terminal ends to the cytoplasm. Targeting to oilbodies enables both high levels of expression and costeffective recovery of recombinant therapeutic proteins. In this technology, therapeutic proteins are covalently targeted to oilbodies as oleosin fusions [147]. By combining this fusion with an expression targeted to the seed, researchers have established a simple expression/purification system in which the recombinant protein is recovered with oilbodies from other seed components by liquid–liquid phase separation. This mild process reduces the number of chromatography steps required to obtain a purified PMP and thereby significantly reduces their purification cost. This strategy has been used for the production of hirudin, an anticoagulant from leach salivary glands and antibodies in different oilseed plants [66,67,148]. Recently human insulin expressed Arabidopsis seed oilbodies was recovered as an active molecule at commercially relevant levels [149]. Production of therapeutic proteins in chloroplasts PMP(s) expression in the chloroplasts also offers several advantages including very high yield. Each cell from higher plants leaves contains as many as hundred chloroplasts with up to hundred chloroplast genome, resulting a total of about 10,000 genome copies per cell. The transgene is introduced into leaf chloroplasts by particule bombardment and directly integrated into the chloroplast genome by homologous recombination [60]. This stable transformation of chloroplasts allows amplification of transgene copies and accumulation of large amounts of recombinant proteins [150–152]. For example, transgenic tobacco chloroplasts produce 300-fold higher amounts of human somatotropin than their nuclear transgenic counterparts [122]. Resulting from high expression levels and low proteolytic activity, a protein expressed in this organelle could represent up to 20% of total leaf proteins [153]. In some cases, concentration of recombinant proteins expressed in chloroplasts is so high that they could form inclusion bodies thus, simplifying

128 purification and increasing resistance to proteolysis of these foreign proteins [154]. However, the need for refolding these therapeutic proteins, after solubilization from inclusion bodies could significantly increase their overall cost. With a limited protein maturation capacity, the chloroplast looks particularly well adapted for production of simple molecules [155], but quite surprisingly, tobacco chloroplasts are also capable to properly fold complex proteins with disulfide bridges, such as human somatotropin [122] and even full-size antibodies [156]. Ancestral plants like algae are also able to produce functional antibodies in their chloroplasts [89]. However, expression in the chloroplasts cannot be considered as a panacea for PMPs expression in planta, as a number of clinically useful proteins necessitate extensive posttranslational processing. For instance, oligosaccharides attached to polypeptide chains by N- or O-glycosylation, in particular, have a strong impact on the activity of several therapeutic proteins and unfortunately chloroplasts do not have the capacity to glycosylate proteins. Recently, an alternative pathway that mediates post-translational delivery of proteins to the chloroplast via the secretory pathway was described in A. thaliana [157]. This pathway provides new opportunities for complementation of the chloroplast protein maturation machinery with chaperones needing ER and/or Golgi typical maturations such as N-glycosylation for their biological activity or using chloroplasts as a storage compartment for glycoproteins [152]. Optimization of plant production platforms for lower proteolytic activity and humanized glycosylation In plants, like in any other heterologous expression system, recombinant protein yield not only depends on an efficient expression rate of the transgene, but also on the stability of the resulting protein during the whole expression/recovery process [158]. Transgenic plants with reduced protease activity Proteases found in the different compartments of plant cells may dramatically alter the stability of foreign proteins either in vivo, or in vitro during their recovery from plant tissues [158,159]. Vacuolar proteases active in mildly-acidic conditions, in particular, were readily identified as potentially damaging for the integrity of recombinant proteins expressed in vegetative organs of transgenic plants. As described above, targeting strategies based on the fusion of appropriate targeting signals to the therapeutic proteins have been used to avoid unwanted proteolysis in vivo by directing their accumulation in compartments such as ER [113,137,138] or chloroplasts where proteolytic activity is low [160].

129 Transgenic plant lines with reduced protease activity levels in vivo could also help to maximize protein yields by slowering cellular hydrolytic processes. In particular, recent evidence in the literature suggests that hindering endogenous protease activities in planta with recombinant protease inhibitors could help enhance protein levels in vegetative organs without compromising growth and development of the host plant. The rice cysteine proteinase inhibitor, oryzacystatin I, for instance, was shown to increase total soluble protein levels by 40% in leaves of transgenic tobacco lines expressing this inhibitor in the cytosolic compartment [161]. Similarly, transgenic potato lines ectopically expressing the aspartate proteinase inhibitor, tomato cathepsin D inhibitor, exhibited total leaf protein levels up to 35% higher than those of control plants, while showing no visible sign of altered growth or development [162]. More recently, transgenic lines of potato expressing either tomato cathepsin D inhibitor or bovine aprotinin, both active against trypsin and chymotrypsin, show a decrease in Rubisco hydrolysis by 30–40% relative to control plants [163]. Based on current knowledge and progress to come on plant cell proteolytic processes, the design of transgenic plant lines deficient in specific protease activities in the secretory pathway could provide plant production platforms optimized for the production of complex proteins in ‘‘mild’’ cellular environments. Current strategies to humanize glycans N-linked to PMPs Many therapeutic proteins are glycoproteins and glycosylation is often essential for their stability, solubility, folding and biological activity. When a mammalian glycoprotein is produced in a plant expression system it is glycosylated on the same Asn residues as it would be in mammals, but its N-glycan structures are different from that of its native counterpart. For instance, plant-made antibodies bear both high-mannose (Man5–Man9 glycans) and biantennary complex type N-glycans [72,73]. The high-mannosetype N-glycans have the same structure in plant and mammalian glycoproteins. But complex-type N-glycans are structurally different in plants and mammals. For instance, in plants, the proximal N-acetylglucosamine of the core is substituted by an a1,3-fucose in place of an a1,6-fucose in mammals, and the b-mannose of the core is substituted by a bisecting b1,2-xylose in plants, in place of a b1,4-N-acetylglucosamine in mammals. In addition, b1, 3-galactose and fucose a1,4-linked to the terminal N-acetylglucosamine of plant N-glycans form Lewis a oligosaccharide structures instead of b1, 4-galactose combined with sialic acids in mammals (Fig. 1). Together with Lewis a, bisecting b1,2-xylose and core a1,3-fucose residues are constitutive of three glycoepitopes described on complex plant N-glycans. Indeed, plant complex N-glycans are immunogenic in most laboratory mammals and elicit glycan-specific IgE-and IgG-antibodies in humans [164–167]. So that, as observed for any other eukaryotic system currently used for

130 Fig. 1. Addition and processing of N-linked glycans in the endoplasmic reticulum (ER) and Golgi apparatus of plant and mammalian cells. A precursor oligosaccharide assembled onto a lipid carrier is transferred on specific Asn residues of the nascent growing polypeptide. The N-glycan is then trimmed off with removal of glucosyl and most mannosyl residues. Differences in the processing of plant and mammalian complex N-glycans are late Golgi maturation events.

131 therapeutic protein production such as yeasts, insect cells or mammaliancultured cells, because of their structural differences with human N-glycans, glycans N-linked to PMPs would be immunogenic in humans when delivered parenterally. To fully exploit the potential of plants for the production of recombinant therapeutic glycoproteins, it is necessary to control the maturation of plantspecific N-glycans and thus prevent the addition of immunogenic glycoepitopes onto PMPs. One of the most drastic approaches is to prevent N-glycosylation, by inactivating N-glycosylation sites through the mutation of Asn or Ser/Thr residues. Generally, this strategy neither influences IgG folding and assembly in the plant ER nor the antigen-binding activity of an antibody [168]. However, many pharmaceuticals, including antibodies used for Fc-dependent functions require glycosylation for in vivo activity and longevity. This is why most efforts in glycoengineering of plant expression systems were focused on the production of glycosylated therapeutic proteins bearing non-immunogenic N-glycans. One of these strategies is based on the inhibition of plant-specific Golgi glycosyltransferases to prevent the addition of glyco-epitopes to PMPs. Knock-out a1,3 fucosyltransferase and b1,2-xylosyltransferase genes, to eliminate the plant-derived glyco-epitopes was successful in several plant expression systems using either insertional mutation in Arabidopsis mutants [169,170] or targeted gene inactivation in the moss Physcomitrella patens [171]. RNA interference was also used for a knock-out of a1,3-fucosyltransferase and b1,2-xylosyltransferase in two plant expression systems; Lemna minor and Medicago sativa [91,172]. The very high efficiency of this strategy has allowed the production in L. minor of a monoclonal antibody cumulating the advantages of homogeneous glycosylation with a single and non-immunogenic N-glycan species. In plants as in other eukaryotic cells, proteins that reside in the lumen of the plant ER contain high-mannose type N-glycans with structures common to mammals. We have recently shown that antibodies expressed in tobacco plants with a KDEL ER retention signal fused at the C-terminal ends of their heavy and light chains contain exclusively non-immunogenic high-mannose type N-glycans [173,174]. These different studies illustrate that several plant expression systems are already available for production of glycosylated PMPs without immunogenic glyco-epitopes. In addition to approaches involving glycosyltransferase inactivation, another attractive strategy to humanize plant N-glycans is to express mammalian glycosyltransferases in plants, which would complete and/or compete with the endogenous machinery of N-glycan maturation in the plant Golgi apparatus. As part of these complementation strategies, it has been shown that the human b1,4 galactosyltransferase, expressed in plant cells, transfers galactose residues onto the terminal N-acetylglucosamine residues of plant N-glycans [73, 175–177]. These results are very promising and several laboratories are currently working to increase the performance of heterologous glycosyltransferases

132 through better control of their targeting in the Golgi cisternae. Indeed, the analysis of several plant glycosyltransferases is currently providing a panel of specific signals sufficient for a targeted expression of heterologous glycosyltransferases within the different Golgi subcompartments of a plant cell [178,179]. The presence of sialic acid residues at the termini of N-glycan antennae is very important for the clearance of many mammalian plasma proteins of pharmaceutical interest. The absence of such residues on circulating proteins results in their rapid elimination from the blood by interactions with galactose-specific receptors on the surface of hepatic cells. Sialic acids are not detectable and thus more probably absent from plant glycoproteins [180,181]. The production of sialylated N-glycans is feasible in plants as previously shown in insect cells [182]. Indeed most of this complex biosynthetic pathway located both in the Golgi lumen and in the cytosol of mammalian cells was already rebuilt in plants with the expression of mammalian a2,6 sialyltransferase [183], human CMP-N-acetylneuraminic acid synthetase, CMP-sialic acid transporter [184] and recently two catalytically active microbial Nacetylneuraminic acid synthesizing enzymes [185]. The next and last steps to get PMP sialylation in planta will be to express an epimerase able to convert endogenous D-GlcNAc or UDP-GlcNAc into D-ManNAc in order to supply the heterologous N-acetylneuraminic acid synthesizing enzymes with the appropriate amino sugar, and to simultaneously express these different genes in a same plant expression system. Emerging plant expression systems for molecular farming As illustrated in Tables 1 and 2, tobacco, A. thaliana, maize, rice and alfalfa were very frequently used for therapeutic proteins production. However, some emerging plant expression systems, like Lemna minor, Physcomitrella patens, Chlamydomonas reinhardtii or higher plant cell suspension cultures are offering new opportunities for molecular farming. Indeed these expression systems have in common to be much more consistent with public demand for high containment of genetically modified plants and also more compliant with regulatory issues for the production of therapeutic proteins since they are grown in a completely controlled environment. Lemna Lemna gibba and Lemna minor, commonly named duckweeds, are free floating plants, which develop on water and are found all over the world. With their naturally simple growth conditions, duckweeds are well adapted for intensive culturing methods. Duckweed allows very high rates of biomass accumulation per unit of time – it can double in size every 24–48 h by a process. Recombinant proteins produced in duckweeds after Agrobacterium

133 tumefaciens-mediated or by biolistic transformation can be extracted and purified or the plant containing the protein can be used directly, dry or fresh. As for other plant expression systems, secretion into the extracellular media is dependent on the presence of a signal peptide. Lemna-recognizing plant and human signal sequences with the same efficiency [186]. The high capacity of this expression system for production of therapeutic proteins was recently illustrated by Cox et al., with the production of a human monoclonal antibody in a glycoengineered lemna. This antibody exhibited a single major N-glycan species without any detectable plant-specific N-glycans and shows a higher antibody-dependent cell-mediated cytotoxicity and effector cell receptor-binding activities than its homologs expressed in cultured Chinese hamster ovary (CHO) cells [91]. Moss Mosses are higher multicellular eukaryotes and therefore perform extensive post-translational processing of proteins including disulfide bridge formation and glycosylation. Transgenic Physcomitrella patens are generated via the polyethyleneglycol-mediated transfection of protoplasts. Generation of stable transgenic plants take about eight weeks after transformation [187,188] and cultivation of this moss in glass bioreactors is well established. As illustrated in Table 1, a therapeutic protein (grow factor VEGF) has been already produced in this expression system [53]. P. patens is unique among all multicellular plants analyzed to date in exhibiting a very effective homologous recombination process in its nuclear DNA. This allows targeted knock-outs and knock-in of genes, a highly attractive tool for production of strains designed for PMP production [187,189]. N-glycosylation in P. patens is very similar as in higher plants [190]. But this moss is currently the most advanced plant expression system for glycoengineering due to the ease with which knock-out and knock-in of glycosylation enzyme genes can be performed by homologous recombination in this system [171]. Thus, P. patens has been engineered to produce a strain that does not add b1,2 xylose or a1,3 fucose, but produces PMPs bearing a core heptasaccharide identical to that of a human IgG [171]. Algae Algae are currently emerging as alternative system for production of recombinant therapeutic proteins. Unicellular eukaryotic green algae, such as Chlamydomonas reinhardtii, Phaeodactylum tricornutum, Tetraselmis suecica, Odontella aurita, can produce a significant amount of recombinant proteins [191]. Freshwater algae C. reinhardtii is the best-studied for recombinant protein production via chloroplast transformation [192]. C. reinhardtii

134 contains a single large chloroplast that occupies approximately 40% of the cell volume and its transformation was first realized in 1988. Unlike nuclear transformation, plastid transformation occurs via homologous recombination. Hence, integration events can be targeted precisely to any region in the chloroplast genome that contains a so-called silent site for transgene integration [191]. The chloroplast contains its own genome, which is a circular molecule of approximately 200 Kb, and each chloroplast contains approximately 80 identical copies of the genome. As a consequence, stable transformation of the chloroplast requires that all 80 copies convert to the recombinant form [193]. C. reinhardtii can be grown in a cost-effective manner at a large scale, in 500,000-l containers. Compared to land plants, it grows at a much faster rate, doubling its cell number every 8 h [191]. Purification of recombinant proteins should be simpler in algae than in terrestrial plants. Indeed, the cellular population of algae is uniform in size and type, hence there is no gradient of recombinant protein distribution, which simplifies purification and reduces the loss of biomass. C. reinhardtii has also the ability to produce secreted proteins, a pathway which could still cut down the production costs [193]. A human mAb produced in transgenic algae was correctly assembled and has the same capacity to bind herpes virus proteins as its mammalian homolog [89]. But chloroplast-encoded proteins are not glycosylated and this mAb has shown no evidence for glycosylation required for the Fc-dependent functions. In addition, codon bias in algae constitutes an additional difficulty for foreign protein expression in this system due to the need of a extensive optimization of gene sequences. Higher plant suspension-cultured cells Higher plant cell cultures offer many advantages over field grown plants or even plants grown in greenhouses for PMP production. Among these advantages, plant cells are grown in highly contained and sterile in vitro conditions. Some plant cells grow very fast, for instance BY2 tobacco cells number is doubling every 12 h in optimal growth conditions, thus rapidly providing an important biomass. Many therapeutic proteins have already been successfully expressed in suspension-cultured plant cells. The potential of plant cells for biopharmaceutical production was recently illustrated with the use of suspension-cultured tobacco cells to synthesize correctly matured and highly immunoreactive recombinant house dust mite allergens that could be used for allergy diagnostic and immunotherapy [194]. This work perfectly exemplifies the high potential of plant suspension-cultured cells as bioreactors for the production of therapeutic proteins under controlled and environmentally safe conditions. In addition this production system allows for an efficient secretion of PMPs into an inorganic culture medium offering substantial cost advantages in downstream purification. This could

135 counterbalance an increased production cost due to the use of fermentor for production instead of field- or green-house production with whole plants. Another advantage for down-stream processing is that plant cells are uniform in size and types, which leads to a low PMP heterogeneity as compared to production in whole plants. For instance, it was reported that glycosylation patterns of an antibody expressed in tobacco plants, differ from young to old leaves [195]. In contrast, glycan patterns are reproducible from batch to batch in BY2 tobacco cell cultures and interestingly complementation of the culture medium could strongly reduce N-glycan heterogeneity (Faye et al., unpublished results). Conclusion Plants offer a safe and extremely cost-effective alternative to microbial or mammalian expression systems for the production of biopharmaceuticals. Current strategies to improve plant expression systems will rapidly result in increased yield and simplification of down-stream processing of plant-made therapeutic proteins. These promising results, together with the fast progresses in glycan humanization and reduced heterogeneity of PMPs are currently moving plants among the major expression systems, particularly when large quantities of multimeric recombinant proteins are required. References 1.

2. 3. 4. 5. 6.

7.

8.

9.

Gomord V, Sourrouille C, Fitchette A-C, Bardor M, Pagnt S, Lerouge P and Faye L. Production and glycosylation of plant-made pharmaceuticals: the antibodies as a challenge. Plant Biotechnol J 2004;2:83–100. Walsh G and Jefferis R. Post-translational modifications in the context of therapeutic proteins. Nat Biotechnol 2006;24:1241–1252. Twyman RM, Stoger E, Schillberg S, Christou P and Fischer R. Molecular farming in plants: host systems and expression technology. Trends Biotechnol 2003;21:570–578. Gomord V and Faye L. Posttranslational modification of therapeutic proteins in plants. Curr Opin Plant Biol 2004;7:171–181. Hiatt A, Cafferkey R and Bowdish K. Production of antibodies in transgenic plants. Nature 1989;342:76–78. Ma JK, Hiatt A, Hein M, Vine ND, Wang F, Stabila P, van Dolleweerd C, Mostov K and Lehner T. Generation and assembly of secretory antibodies in plants. Science 1995;268:716–719. Khoudi H, Laberge S, Ferullo JM, Bazin R, Darveau A, Castonguay Y, Allard G, Lemieux R and Vezina LP. Production of a diagnostic monoclonal antibody in perennial alfalfa plants. Biotechnol Bioeng 1999;64:135–143. Austin S, Bingham ET, Koegel RG, Mathews DE, Shahan MN, Straub RJ and Burgess RR. An overview of a feasibility study for the production of industrial enzymes in transgenic alfalfa. Ann NY Acad Sci 1994;721:234–244. Gomord V, Chamberlain P, Jefferis R and Faye L. Biopharmaceutical production in plants: problems, solutions and opportunities. Trends Biotechnol 2005;23:559–565.

136 10. Sijmons PC, Dekker BM, Schrammeijer B, Verwoerd TC, van den Elzen PJ and Hoekema A. Production of correctly processed human serum albumin in transgenic plants. Biotechnology (NY) 1990;8:217–221. 11. Goddjin OJM and Pen J. Plants as bioreactors. Trends Biotechnol 1995;13:379–387. 12. Farran I, Sanchez-Serrano JJ, Medina JF, Prieto J and Mingo-Castel AM. Targeted expression of human serum albumin to potato tubers. Transgenic Res 2002;11:337–346. 13. Zhong GY, Peterson DJ, Delaney D, Bailey M, Witcher DR, Register JC, Bond D, Li CP, Marshall L, Kulisek E, Ritland D, Meyer T, Hood EE and Howard J. Commercial production of aprotinin in transgenic maize seeds. Mol Breeding 1999;5(4):345–356. 14. Ruggiero F, Exposito JY, Bournat P, Gruber V, Perret S, Comte J, Olagnier B, Garrone R and Theisen M. Triple helix assembly and processing of human collagen produced in transgenic tobacco plants. FEBS Lett 2000;469:132–136. 15. Dieryck W, Pagnier J, Poyart C, Marden MC, Gruber V, Bournat P, Baudino S and Merot B. Human haemoglobin from transgenic tobacco. Nature 1997;386:29–30. 16. Halling-Sorensen B, Nors Nielsen S, Lanzky PF, Ingerslev F, Holten Lutzhoft HC and Jorgensen SE. Occurrence, fate and effects of pharmaceutical substances in the environment – a review. Chemosphere 1998;36:357–393. 17. Dieryck W, Gruber V, Baudino S, Lenee P, Pagnier J, Merot B and Poyart C. Expression of recombinant human hemoglobin in plants. Transfus Clin Biol 1995;2:441–447. 18. Terashima M, Murai Y, Kawamura M, Nakanishi S, Stoltz T, Chen L, Drohan W, Rodriguez RL and Katoh S. Production of functional human alpha 1-antitrypsin by plant cell culture. Appl Microbiol Biotechnol 1999;52:516–523. 19. McDonald KA, Hong LM, Trombly DM, Xie Q and Jackman AP. Production of human alpha-1-antitrypsin from transgenic rice cell culture in a membrane bioreactor. Biotechnol Prog 2005;21:728–734. 20. Krebitz M, Wiedermann U, Essl D, Steinkellner H, Wagner B, Turpen TH, Ebner C, Scheiner O and Breiteneder H. Rapid production of the major birch pollen allergen Bet v 1 in Nicotiana benthamiana plants and its immunological in vitro and in vivo characterization. Faseb J 2000;14:1279–1288. 21. Arakawa T, Yu J and Langridge WH. Food plant-delivered cholera toxin B subunit for vaccination and immunotolerization. Adv Exp Med Biol 1999;464:161–178. 22. Tackaberry ES, Dudani AK, Prior F, Tocchi M, Sardana R, Altosaar I and Ganz PR. Development of biopharmaceuticals in plant expression systems: cloning, expression and immunological reactivity of human cytomegalovirus glycoprotein B (UL55) in seeds of transgenic tobacco. Vaccine 1999;17:3020–3029. 23. Arakawa T, Chong DK, Merritt JL and Langridge WH. Expression of cholera toxin B subunit oligomers in transgenic potato plants. Transgenic Res 1997;6:403–413. 24. Brennan FR, Bellaby T, Helliwell SM, Jones TD, Kamstrup S, Dalsgaard K, Flock JI and Hamilton WD. Chimeric plant virus particles administered nasally or orally induce systemic and mucosal immune responses in mice. J Virol 1999;73:930–938. 25. Wigdorovitz A, Perez Filgueira DM, Robertson N, Carrillo C, Sadir AM, Morris TJ and Borca MV. Protection of mice against challenge with foot and mouth disease virus (FMDV) by immunization with foliar extracts from plants infected with recombinant tobacco mosaic virus expressing the FMDV structural protein VP1. Virology 1999;264:85–91. 26. Tacket CO and Mason HS. A review of oral vaccination with transgenic vegetables. Microbes Infect 1999;1:777–783. 27. Dus Santos MJ, Wigdorovitz A, Trono K, Rios RD, Franzone PM, Gil F, Moreno J, Carrillo C, Escribano JM and Borca MV. A novel methodology to develop a foot and

137

28.

29.

30. 31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

mouth disease virus (FMDV) peptide-based vaccine in transgenic plants. Vaccine 2002;20:1141–1147. Wu H, Singh NK, Locy RD, Scissum-Gunn K and Giambrone JJ. Expression of immunogenic VP2 protein of infectious bursal disease virus in Arabidopsis thaliana. Biotechnol Lett 2004;26:787–792. Wigdorovitz A, Mozgovoj M, Santos MJ, Parreno V, Gomez C, Perez-Filgueira DM, Trono KG, Rios RD, Franzone PM, Fernandez F, Carrillo C, Babiuk LA, Escribano JM and Borca MV. Protective lactogenic immunity conferred by an edible peptide vaccine to bovine rotavirus produced in transgenic plants. J Gen Virol 2004;85:1825–1832. Beachy RN, Fitchen JH and Hein MB. Use of plant viruses for delivery of vaccine epitopes. Ann NY Acad Sci 1996;792:43–49. Huang Z, Dry I, Webster D, Strugnell R and Wesselingh S. Plant-derived measles virus hemagglutinin protein induces neutralizing antibodies in mice. Vaccine 2001;19: 2163–2171. Sojikul P, Buehner N and Mason HS. A plant signal peptide-hepatitis B surface antigen fusion protein with enhanced stability and immunogenicity expressed in plant cells. Proc Natl Acad Sci USA 2003;100:2209–2214. Ma SW, Zhao DL, Yin ZQ, Mukherjee R, Singh B, Qin HY, Stiller CR and Jevnikar AM. Transgenic plants expressing autoantigens fed to mice to induce oral immune tolerance. Nat Med 1997;3:793–796. Richter LJ, Thanavala Y, Arntzen CJ and Mason HS. Production of hepatitis B surface antigen in transgenic plants for oral immunization. Nat Biotechnol 2000;18: 1167–1171. Kong Q, Richter L, Yang YF, Arntzen CJ, Mason HS and Thanavala Y. Oral immunization with hepatitis B surface antigen expressed in transgenic plants. Proc Natl Acad Sci USA 2001;98:11539–11544. Buratti E, McLain L, Tisminetzky S, Cleveland S, Dimmock N and Baralle F. The neutralizing antibody response against a conserved region of human immunodeficiency virus type 1 gp41 (amino acid residues 731–752) is uniquely directed against a conformational epitope. Virology 1999;13:930–938. Mason HS, Haq TA, Clements JD and Arntzen CJ. Edible vaccine protects mice against Escherichia coli heat-labile enterotoxin (LT): potatoes expressing a synthetic LT-B gene. Vaccine 1998;16:1336–1343. Arakawa T, Yu J, Chong DK, Hough J, Engen PC and Langridge WH. A plant-based cholera toxin B subunit-insulin fusion protein protects against the development of autoimmune diabetes. Nat Biotechnol 1998;16:934–938. Turpen T, Reinl S, Charoenvit Y, Hoffman S, Fallarme V and Grill L. Malarial epitopes expressed on the surface of recombinant tobacco mosaic virus. Biotechnology (NY) 1995;13. Tacket CO, Mason HS, Losonsky G, Estes MK, Levine MM and Arntzen CJ. Human immune responses to a novel norwalk virus vaccine delivered in transgenic potatoes. J Infect Dis 2000;182:302–305. Huang Z, Elkin G, Maloney BJ, Beuhner N, Arntzen CJ, Thanavala Y and Mason HS. Virus-like particle expression and assembly in plants: hepatitis B and Norwalk viruses. Vaccine 2005;23:1851–1858. McGarvey PB, Hammond J, Dienelt MM, Hooper DC, Fu ZF, Dietzschold B, Koprowski H and Michaels FH. Expression of the rabies virus glycoprotein in transgenic tomatoes. Biotechnology (NY) 1995;13:1484–1487.

138 43. James EA, Wang C, Wang Z, Reeves R, Shin JH, Magnuson NS and Lee JM. Production and characterization of biologically active human GM-CSF secreted by genetically modified plant cells. Protein Expr Purif 2000;19:131–138. 44. Lee JS, Choi SJ, Kang HS, Oh WG, Cho KH, Kwon TH, Kim DH, Jang YS and Yang MS. Establishment of a transgenic tobacco cell suspension culture system for producing murine granulocyte-macrophage colony stimulating factor. Mol Cells 1997; 7:783–787. 45. Robinson A. Harvesting blood proteins from grain. Can Med Assoc J 1995;153: 427–429. 46. Wang ML, Goldstein C, Su W, Moore PH and Albert HH. Production of biologically active GM-CSF in sugarcane: a secure biofactory. Transgenic Res 2005;14:167–178. 47. Edelbaum O, Stein D, Holland N, Gafni Y, Livneh O, Novick D, Rubinstein M and Sela I. Expression of active human interferon-beta in transgenic plants. J Interferon Res 1992;12:449–453. 48. De Zoeten GA, Penswick JR, Horisberger MA, Ahl P, Schultze M and Hohn T. The expression, localization, and effect of a human interferon in plants. Virology 1989;172:213–222. 49. Chen TL, Lin YL, Lee YL, Yang NS and Chan MT. Expression of bioactive human interferon-gamma in transgenic rice cell suspension cultures. Transgenic Res 2004; 13:499–510. 50. Cramer CL, Boothe JG and Oishi KK. Transgenic plants for therapeutic proteins: linking upstream and downstream strategies. Curr Top Microbiol Immunol 1999;240:95–118. 51. Matsumoto S, Ikura K, Ueda M and Sasaki R. Characterization of a human glycoprotein (erythropoietin) produced in cultured tobacco cells. Plant Mol Biol 1995; 27:1163–1172. 52. Baur A, Reski R and Gorr G. Enhanced recovery of a secreted recombinant human growth factor using stabilizing additives and by co-expression of human serum albumin in the moss Physcomitrella patens. Plant Biotechnol J 2005;3:331–340. 53. Park Y and Cheong H. Expression and production of recombinant human interleukin-2 in potato plants. Protein Expr Purif 2002;25:160–165. 54. Kwon S, Yang Y, Hong C and Pyun K. Expression of active human interleukin-6 in transgenic tabacco. Mol Cells 1995;5:486–492. 55. Menassa R, Kennette W, Nguyen V, Rymerson R, Jevnikar A and Brandle J. Subcellular targeting of human interleukin-10 in plants. J Biotechnol 2004;108:179–183. 56. Gutierrez-Ortega A, Avila-Moreno F, Saucedo-Arias LJ, Sanchez-Torres C and Gomez-Lim MA. Expression of a single-chain human interleukin-12 gene in transgenic tobacco plants and functional studies. Biotechnol Bioeng 2004;85:734–740. 57. Panahi M, Alli Z, Cheng X, Belbaraka L, Belgoudi J, Sardana R, Phipps J and Altosaar I. Recombinant protein expression plasmids optimized for industrial E. coli fermentation and plant systems produce biologically active human insulin-like growth factor-1 in transgenic rice and tobacco plants. Transgenic Res 2004;13:245–259. 58. Hamamoto H, Sugiyama Y, Nakagawa N, Hashida E, Matsunaga Y, Takemoto S, Watanabe Y and Okada Y. A new tobacco mosaic virus vector and its use for the systemic production of angiotensin-I-converting enzyme inhibitor in transgenic tobacco and tomato. Biotechnology (NY) 1993;11:930–932. 59. Cramer CL, Weissenborn DL, Oishi KK, Grabau EA, Bennett S, Ponce E, Grabowski GA and Radin DN. Bioproduction of human enzymes in transgenic tobacco. Ann NY Acad Sci 1996;792:62–71.

139 60. Staub JM, Garcia B, Graves J, Hajdukiewicz PT, Hunter P, Nehra N, Paradkar V, Schlittler M, Carroll JA, Spatola L, Ward D, Ye G and Russell DA. High-yield production of a human therapeutic protein in tobacco chloroplasts. Nat Biotechnol 2000;18:333–338. 61. Reggi S, Marchetti S, Patti T, De Amicis F, Cariati R, Bembi B and Fogher C. Recombinant human acid beta-glucosidase stored in tobacco seed is stable, active and taken up by human fibroblasts. Plant Mol Biol 2005;57:101–113. 62. Kumagai MH, Turpen TH, Weinzettl N, della-Cioppa G, Turpen AM, Donson J, Hilf ME, Grantham GL, Dawson WO, Chow TP, Piatak M and Grill LK. Rapid, high-level expression of biologically active alpha-trichosanthin in transfected plants by an RNA viral vector. Proc Natl Acad Sci USA 1993;90:427–430. 63. Mor TS, Sternfeld M, Soreq H, Arntzen CJ and Mason HS. Expression of recombinant human acetylcholinesterase in transgenic tomato plants. Biotechnol Bioeng 2001;75: 259–266. 64. Gruber V, Berna P, Arnaud T, Bournat P, Cle´ment C, Mison D, Olagnier B, Philippe L, Theisen M, Baudino S, Beńicourt C, Cudrey C, Bloe¨s C, Duchateau N, Dufour S, Gueguen C, Jacquet S, Ollivo C, Poncetta C, Zorn N, Ludevid D, Van Dorsselaer A, Verger R, Doherty A, Me´rot B and Danzin C. Large-scale production of a therapeutic protein in transgenic tabacco plants: effect of subcellular targeting on quality of a recombinant dog gastric lipase. Mol Breeding 2001;7(4): 329–340. 65. Sorrentino A, Schillberg S, Fischer R, Rao R, Porta R and Mariniello L. Recombinant human tissue transglutaminase produced into tobacco suspension cell cultures is active and recognizes autoantibodies in the serum of coeliac patients. Int J Biochem Cell Biol 2005;37:842–851. 66. Parmenter DL, Boothe JG, van Rooijen GJ, Yeung EC and Moloney MM. Production of biologically active hirudin in plant seeds using oleosin partitioning. Plant Mol Biol 1995;29:1167–1180. 67. Boothe JG, Saponja JA and Parmenter DL. Molecular farming in plants: oilseed as vehicles for the production of pharmaceutical proteins. Drug Dev Res 1997;42: 172–178. 68. Hong SH, Kim KI, Chung HY, Kim YJ, Sunter G, Bisaro DM and Chung IS. Expression of recombinant endostatin in Agrobacterium-inoculated leaf disks of Nicotiana tabacum var. Xanthi Biotechnol Lett 2004;26:1433–1439. 69. Salmon V, Legrand D, Slomianny MC, el Yazidi I, Spik G, Gruber V, Bournat P, Olagnier B, Mison D, Theisen M and Merot B. Production of human lactoferrin in transgenic tobacco plants. Protein Expr Purif 1998;13:127–135. 70. Nandi S, Yalda D, Lu S, Nikolov Z, Misaki R, Fujiyama K and Huang N. Process development and economic evaluation of recombinant human lactoferrin expressed in rice grain. Transgenic Res 2005;14:237–249. 71. Li Y, Geng Y, Song H, Zheng G, Huan L and Qiu B. Expression of a human lactoferrin N-lobe in Nicotiana benthmiana with potato virus X-based agroinfection. Biotechnol Lett 2004;26:953–957. 72. Cabanes-Macheteau M, Fitchette-Laine AC, Loutelier-Bourhis C, Lange C, Vine ND, Ma JK, Lerouge P and Faye L. N-Glycosylation of a mouse IgG expressed in transgenic tobacco plants. Glycobiology 1999;9:365–372. 73. Bakker H, Bardor M, Molthoff JW, Gomord V, Elbers I, Stevens LH, Jordi W, Lommen A, Faye L, Lerouge P and Bosch D. Galactose-extended glycans of antibodies produced by transgenic plants. Proc Natl Acad Sci USA 2001;98:2899–2904.

140 74. Vaquero C, Sack M, Chandler J, Drossard J, Schuster F, Monecke M, Schillberg S and Fischer R. Transient expression of a tumor-specific single-chain fragment and a chimeric antibody in tobacco leaves. Proc Natl Acad Sci USA 1999;96:11128–11133. 75. Vaquero C, Sack M, Schuster F, Finnern R, Drossard J, Schumann D, Reimann A and Fischer R. A carcinoembryonic antigen-specific diabody produced in tobacco. Faseb J 2002;16:408–410. 76. Stoger E, Vaquero C, Torres E, Sack M, Nicholson L, Drossard J, Williams S, Keen D, Perrin Y, Christou P and Fischer R. Cereal crops as viable production and storage systems for pharmaceutical scFv antibodies. Plant Mol Biol 2000;42:583–590. 77. Ko K, Tekoah Y, Rudd PM, Harvey DJ, Dwek RA, Spitsin S, Hanlon CA, Rupprecht C, Dietzschold B, Golovkin M and Koprowski H. Function and glycosylation of plantderived antiviral monoclonal antibody. Proc Natl Acad Sci USA 2003;100:8013–8018. 78. Ma JK, Lehner T, Stabila P, Fux CI and Hiatt A. Assembly of monoclonal antibodies with IgG1 and IgA heavy chain domains in transgenic tobacco plants. Eur J Immunol 1994;24:131–138. 79. Larrick JW, Yu L, Naftzger C, Jaiswal S and Wycoff K. Production of secretory IgA antibodies in plants. Biomol Eng 2001;18:87–94. 80. Verch T, Yusibov V and Koprowski H. Expression and assembly of a full-length monoclonal antibody in plants using a plant virus vector. J Immunol Methods 1998;1:69–75. 81. Briggs K, Zeitlin L, Wang F, Chen L, Fitchen J, Glynn J, Lee VD, Zhang S and Whaley K. Anti-HSV antibodies produced in rice plants protect mice from vaginal HSV infection. Plant Biol 2000(Stuttg.), Abstract 15. 82. Zeitlin L, Olmsted SS, Moench TR, Co MS, Martinell BJ, Paradkar VM, Russell DR, Queen C, Cone RA and Whaley KJ. A humanized monoclonal antibody produced in transgenic plants for immunoprotection of the vagina against genital herpes. Nat Biotechnol 1998;16:1361–1364. 83. Yuan Q, Hu W, Pestka JJ, He SY and Hart LP. Expression of a functional antizearalenone single-chain Fv antibody in transgenic Arabidopsis plants. Appl Environ Microbiol 2000;66:3499–3505. 84. De Wilde C, De Neve M, De Rycke R, Bruyns A-M, De Jaeger G, Van Montagu M, Depicker A and Engler G. Intact antigen-binding MAK33 antibody and Fab fragment accumulate in intercellular spaces of Arabidopsis thaliana. Plant Sci 1996;114:233–241. 85. De Neve M, De Loose M, Jacobs A, Van Houdt H, Kaluza B, Weidle U, Van Montagu M and Depicker A. Assembly of an antibody and its derived antibody fragment in Nicotiana and Arabidopsis. Transgenic Res 1993;2:227–237. 86. Bruyns AM, De Jaeger G, De Neve M, De Wilde C, Van Montagu M and Depicker A. Bacterial and plant-produced scFv proteins have similar antigen-binding properties. FEBS Lett 1996;386:5–10. 87. Peeters K, De Wilde C and Depicker A. Highly efficient targeting and accumulation of a F(ab) fragment within the secretory pathway and apoplast of Arabidopsis thaliana. Eur J Biochem 2001;268:4251–4260. 88. McCormick AA, Kumagai MH, Hanley K, Turpen TH, Hakim I, Grill LK, Tuse D, Levy S and Levy R. Rapid production of specific vaccines for lymphoma by expression of the tumor-derived single-chain Fv epitopes in tobacco plants. Proc Natl Acad Sci USA 1999;96:703–708. 89. Mayfield SP, Franklin SE and Lerner RA. Expression and assembly of a fully active antibody in algae. Proc Natl Acad Sci USA 2003;100:438–442.

141 90. Ramirez N, Ayala M, Lorenzo D, Palenzuela D, Herrera L, Doreste V, Perez M, Gavilond JV and Oramas P. Expression of a single-chain Fv antibody fragment specific for the hepatitis B surface antigen in transgenic tobacco plants. Transgenic Res 2002;11:61–64. 91. Cox KM, Sterling JD, Regan JT, Gasdaska JR, Frantz KK, Peele CG, Black A, Passmore D, Moldovan-Loomis C, Srinivasan M, Cuison S, Cardarelli PM and Dickey LF. Glycan optimization of a human monoclonal antibody in the aquatic plant Lemna minor. Nat Biotechnol 2006;24:1591–1597. 92. Bouquin T, Thomsen M, Nielsen LK, Green TH, Mundy J and Hanefeld Dziegiel M. Human anti-rhesus D IgG1 antibody produced in transgenic plants. Transgenic Res 2002;11:115–122. 93. Sharp PM and Li WH. The codon Adaptation Index – a measure of directional synonymous codon usage bias, and its potential applications. Nucleic Acids Res 1987;15: 1281–1295. 94. Gustafsson C, Govindarajan S and Minshull J. Codon bias and heterologous protein expression. Trends Biotechnol 2004;22:346–353. 95. Kurland C and Gallant J. Errors of heterologous protein expression. Curr Opin Biotechnol 1996;7:489–493. 96. Perlak FJ, Fuchs RL, Dean DA, McPherson SL and Fischhoff DA. Modification of the coding sequence enhances plant expression of insect control protein genes. Proc Natl Acad Sci USA 1991;88:3324–3328. 97. Jensen LG, Olsen O, Kops O, Wolf N, Thomsen KK and von Wettstein D. Transgenic barley expressing a protein-engineered, thermostable (1,3-1,4)-beta-glucanase during germination. Proc Natl Acad Sci USA 1996;93:3487–3491. 98. Batard Y, Hehn A, Nedelkina S, Schalk M, Pallett K, Schaller H and Werck-Reichhart D. Increasing expression of P450 and P450-reductase proteins from monocots in heterologous systems. Arch Biochem Biophys 2000;379:161–169. 99. Hamada A, Yamaguchi K-I, Ohnishi N, Harada M, Nikumaru S and Honda H. Highlevel production of yeast (Schwanniomyces occidentalis) phytase in transgenic rice plants by a combination of signal sequence and codon modification of the phytase gene. Plant Biotechnol J 2005;3:43–55. 100. Hamilton AJ and Baulcombe DC. A species of small antisense RNA in posttranscriptional gene silencing in plants. Science 1999;286:950–952. 101. Aravin A and Tuschl T. Identification and characterization of small RNAs involved in RNA silencing. FEBS Lett 2005;579:5830–5840. 102. Voinnet O. Induction and suppression of RNA silencing: insights from viral infections. Nat Rev Genet 2005;6:206–220. 103. Voinnet O, Rivas S, Mestre P and Baulcombe D. An enhanced transient expression system in plants based on suppression of gene silencing by the p19 protein of tomato bushy stunt virus. Plant J 2003;33:949–956. 104. Scholthof HB, Scholthof KB and Jackson AO. Identification of tomato bushy stunt virus host-specific symptom determinants by expression of individual genes from a potato virus X vector. Plant Cell 1995;7:1157–1172. 105. Voinnet O, Pinto YM and Baulcombe DC. Suppression of gene silencing: a general strategy used by diverse DNA and RNA viruses of plants. Proc Natl Acad Sci USA 1999;96:14147–14152. 106. Goldstein DA and Thomas JA. Biopharmaceuticals derived from genetically modified plants. Q J Med 2004;97:705–716.

142 107. Hellwig S, Drossard J, Twyman RM and Fischer R. Plant cell cultures for the production of recombinant proteins. Nat Biotechnol 2004;22:1415–1422. 108. Masumura T, Morita S, Miki Y, Kurita A, Morita S, Shirono H, Koga J and Tanaka K. Production of biologically active human interferon-a in transgenic rice. Plant Biotechnol 2006;23:91–97. 109. Shirono H, Morita S, Miki Y, Kurita A, Morita S, Koga J, Tanaka K and Masumura T. Highly efficient production of human interferon-a by transgenic cultured rice cells. Plant Biotechnol 2006;23:283–289. 110. Raskin I, Ribnicky DM, Komarnytsky S, Ilic N, Poulev A, Borisjuk N, Brinker A, Moreno DA, Ripoll C, Yakoby N, O’Neal JM, Cornwell T, Pastor I and Fridlender B. Plants and human health in the twenty-first century. Trends Biotechnol 2002;20: 522–531. 111. Mason HS, Lam DM and Arntzen CJ. Expression of hepatitis B surface antigen in transgenic plants. Proc Natl Acad Sci USA 1992;89:11745–11749. 112. Frigerio L, Vine ND, Pedrazzini E, Hein MB, Wang F, Ma JK and Vitale A. Assembly, secretion, and vacuolar delivery of a hybrid immunoglobulin in plants. Plant Physiol 2000;123:1483–1494. 113. Schouten A, Roosien J, van Engelen FA, de Jong GA, Borst-Vrenssen AW, Zilverentant JF, Bosch D, Stiekema WJ, Gommers FJ, Schots A and Bakker J. The C-terminal KDEL sequence increases the expression level of a single-chain antibody designed to be targeted to both the cytosol and the secretory pathway in transgenic tobacco. Plant Mol Biol 1996;30:781–793. 114. Torres E, Vaquero C, Nicholson L, Sack M, Stoger E, Drossard J, Christou P, Fischer R and Perrin Y. Rice cell culture as an alternative production system for functional diagnostic and therapeutic antibodies. Transgenic Res 1999;8:441–449. 115. Fiedler U, Phillips J, Artsaenko O and Conrad U. Optimization of scFv antibody production in transgenic plants. Immunotechnology 1997;3:205–216. 116. Hood EE, Witcher DR, Maddock S, Meyer T, Baszczynski C, Bailey M, Flynn P, Register J, Marshall L, Bond D, Kulisek E, Kusnadi AR, Evangelista R, Nikolov ZL, Wooge C, Mehigh RJ, Hernan R, Kappel WK, Ritland D, Li CP and Howard JA. Commercial production of avidin from transgenic maize: characterization of transformant, production, processing, extraction and purification. Mol Breeding 1997; 3:291–306. 117. Artsaenko O, Kettig B, Fiedler U, Conrad U and Du¨ring K. Potato tubers as a biofactory for recombinant antibodies. Mol Breeding 1998;4:313–319. 118. Baum TJ, Hiatt A, Parrott WA, Pratt LH and Hussey RS. Expression in tobacco of a functional monoclonal antibody specific to stylet secretions of the root-knot nematode. Mol Plant-Microbe Interact 1996;9:382–387. 119. Sandhu JS, Krasnyanski SF, Domier LL, Korban SS, Osadjan MD and Buetow DE. Oral immunization of mice with transgenic tomato fruit expressing respiratory syncytial virus-F protein induces a systemic immune response. Transgenic Res 2000;9:127–135. 120. Borisjuk NV, Borisjuk LG, Logendra S, Petersen F, Gleba Y and Raskin I. Production of recombinant proteins in plant root exudates. Nat Biotechnol 1999;17:466–469. 121. Komarnytsky S, Borisjuk NV, Borisjuk LG, Alam MZ and Raskin I. Production of recombinant proteins in tobacco guttation fluid. Plant Physiol 2000;124:927–934. 122. Staub JM, Garcia B, Graves J, Hajdukiewicz PT, Hunter P, Nehra N, Paradkar V, Schlittler M, Carroll JA, Spatola L, Ward D, Ye G and Russell DA. High-yield production of a human therapeutic protein in tobacco chloroplasts. Nat Biotechnol 2000;18:333–338.

143 123. Stoger E, Sack M, Fischer R and Christou P. Plantibodies: applications, advantages and bottlenecks. Curr Opin Biotechnol 2002;13:161–166. 124. Gomord V, Wee E and Faye L. Protein retention and localization in the endoplasmic reticulum and the golgi apparatus. Biochimie 1999;81:607–618. 125. Goldstein DA and Thomas JA. Biopharmaceuticals derived from genetically modified plants. Qjm 2004;97:705–716. 126. Streatfield SJ. Approaches to achieve high-level heterologous protein production in plants. Plant Biotechnol J 2006;5:2–15. 127. Gaume A, Komarnytsky S, Borisjuk N and Raskin I. Rhizosecretion of recombinant proteins from plant hairy roots. Plant Cell Rep 2003;21:1188–1193. 128. Ma JK, Drake PM and Christou P. The production of recombinant pharmaceutical proteins in plants. Nat Rev Genet 2003;4:794–805. 129. Komarnytsky S, Borisjuk N, Yakoby N, Garvey A and Raskin I. Cosecretion of protease inhibitor stabilizes antibodies produced by plant roots. Plant Physiol 2006;141:1185–1193. 130. Drake PM, Chargelegue DM, Vine ND, van Dolleweerd CJ, Obregon P and Ma JK. Rhizosecretion of a monoclonal antibody protein complex from transgenic tobacco roots. Plant Mol Biol 2003;52:233–241. 131. Wandelt CI, Khan MR, Craig S, Schroeder HE, Spencer D and Higgins TJ. Vicilin with carboxy-terminal KDEL is retained in the endoplasmic reticulum and accumulates to high levels in the leaves of transgenic plants. Plant J 1992;2:181–192. 132. Tabe LM, Wardley-Richardson T, Ceriotti A, Aryan A, McNabb W, Moore A and Higgins TJ. A biotechnological approach to improving the nutritive value of alfalfa. J Anim Sci 1995;73:2752–2759. 133. Pueyo JJ, Chrispeels MJ and Herman EM. Degradation of transport-competent destabilized phaseolin with a signal for retention in the endoplasmic reticulum occurs in the vacuole. Planta 1995;196:586–596. 134. Conrad U and Fiedler U. Compartment-specific accumulation of recombinant immunoglobulins in plant cells: an essential tool for antibody production and immunomodulation of physiological functions and pathogen activity. Plant Mol Biol 1998; 38:101–109. 135. Pagny S, Denmat-Ouisse LA, Gomord V and Faye L. Fusion with HDEL protects cell wall invertase from early degradation when N-glycosylation is inhibited. Plant Cell Physiol 2003;44:173–182. 136. Triguero A, Cabrera G, Cremata JA, Yuen CT, Wheeler J and Ramirez NI. Plantderived mouse IgG monoclonal antibody fused to KDEL endoplasmic reticulum-retention signal is N-glycosylated homogeneously throughout the plant with mostly highmannose-type N-glycans. Plant Biotechnol J 2005;3(4):449–457. 137. Wandelt CI, Khan MR, Craig S, Schroeder HE, Spencer D and Higgins TJ. Vicilin with carboxy-terminal KDEL is retained in the endoplasmic reticulum and accumulates to high levels in the leaves of transgenic plants. Plant J 1992;2:181–192. 138. Gomord V, Denmat LA, Fitchette-Laine AC, Satiat-Jeunemaitre B, Hawes C and Faye L. The C-terminal HDEL sequence is sufficient for retention of secretory proteins in the endoplasmic reticulum (ER) but promotes vacuolar targeting of proteins that escape the ER. Plant J 1997;11:313–325. 139. Park M, Kim SJ, Vitale A and Hwang I. Identification of the protein storage vacuole and protein targeting to the vacuole in leaf cells of three plant species. Plant Physiol 2004;134:625–639. 140. Park M, Lee D, Lee GJ and Hwang I. AtRMR1 functions as a cargo receptor for protein trafficking to the protein storage vacuole. J Cell Biol 2005;170:757–767.

144 141. Vitale A and Pedrazzini E. Recombinant pharmaceuticals from plants: the plant endomembrane system as bioreactor. Mol Interv 2005;5:216–225. 142. Humphrey BD, Huang N and Klasing KC. Rice expressing lactoferrin and lysozyme has antibiotic-like properties when fed to chicks. J Nutr 2002;132:1214–1218. 143. Yang L, Tada Y, Yamamoto MP, Zhao H, Yoshikawa M and Takaiwa F. A transgenic rice seed accumulating an anti-hypertensive peptide reduces the blood pressure of spontaneously hypertensive rats. FEBS Lett 2006;580:3315–3320. 144. Yang D, Guo F, Liu B, Huang N and Watkins SC. Expression and localization of human lysozyme in the endosperm of transgenic rice. Planta 2003;216(4):597–603. 145. Arcalis E, Marcel S, Altmann F, Kolarich D, Drakakaki G, Fischer R, Christou P and Stoger E. Unexpected deposition patterns of recombinant proteins in post-endoplasmic reticulum compartments of wheat endosperm. Plant Physiol 2004;136:3457–3466. 146. Paris N and Neuhaus JM. BP-80 as a vacuolar sorting receptor. Plant Mol Biol 2002;50:903–914. 147. van Rooijen GJ and Moloney MM. Structural requirements of oleosin domains for subcellular targeting to the oil body. Plant Physiol 1995;109:1353–1361. 148. Seon JH, Szarka S and Molone MM. A unique strategy for recovering recombinant proteins from molecular farming: affinity capture on engineered oilbodies. J Plant Biotechnol 2002;4:95–101. 149. Nykiforuk CL, Boothe JG, Murray EW, Keon RG, Goren HJ, Markley NA and Moloney MM. Transgenic expression and recovery of biologically active recombinant human insulin from Arabidopsis thaliana seeds. Plant Biotechnol J 2006;4:77–86. 150. Daniell H and Dhingra A. Multigene engineering: dawn of an exciting new era in biotechnology. Curr Opin Biotechnol 2002;13:136–141. 151. De Cosa B, Moar W, Lee SB, Miller M and Daniell H. Overexpression of the Bt cry2Aa2 operon in chloroplasts leads to formation of insecticidal crystals. Nat Biotechnol 2001;19:71–74. 152. Faye L and Daniell H. Novel pathway for glycoprotein import into chloroplasts. Plant Biotechnol J 2006;4:275–279. 153. Maliga P. Plant biology: mobile plastid genes. Nature 2003;422:31–32. 154. Millan AF-S, Mingo-Castel A, Miller M and Daniell H. A chloroplast transgenic approach to hyper-express and purify Human Serum Albumin, a protein highly susceptible to proteolytic degradation. Plant Biotechnol J 2003;1:71–79. 155. Jobling SA, Jarman C, Teh MM, Holmberg N, Blake C and Verhoeyen ME. Immunomodulation of enzyme function in plants by single-domain antibody fragments. Nat Biotechnol 2003;21:77–80. 156. Daniell H, Streatfield SJ and Wycoff K. Medical molecular farming: production of antibodies, biopharmaceuticals and edible vaccines in plants. Trends Plant Sci 2001; 6:219–226. 157. Villarejo A, Buren S, Larsson S, Dejardin A, Monne M, Rudhe C, Karlsson J, Jansson S, Lerouge P, Rolland N, von Heijne G, Grebe M, Bako L and Samuelsson G. Evidence for a protein transported through the secretory pathway en route to the higher plant chloroplast. Nat Cell Biol 2005;7:1224–1231. 158. Michaud D, Vrain T, Gomord V and Faye L. Stability of recombinant proteins in plants. In: Methods in Biotechnology – Recombinant Proteins from Plants Production and Isolation of Clinically Useful Compounds, AJR P (ed), Totowa, NJ, Humana Press, 1998, pp. 177–188. 159. Michaud D. Gel electrophoresis of proteolytic enzymes. Anal Chim Acta 1998; 372:173–185.

145 160. Wong EY, Hironaka CM and Fischhoff DA. Arabidopsis thaliana small subunit leader and transit peptide enhance the expression of Bacillus thuringiensis proteins in transgenic plants. Plant Mol Biol 1992;20:81–93. 161. Van der Vyver C, Schneideret J, Driscoll S, Turner J, Kunert K and Foyer C. Oryzacystatin I expression in transformed tobacco produces a conditional growth phenotype and enhances chilling tolerance. Plant Biotechnol J 2003;1:101–112. 162. Michaud D, Anguenot R and Brunelle F. Method for increasing protein content in plant cells. (US Patent application) 2003. 163. Rivard D, Anguenot R, Brunelle F, Le VQ, Ve´zina LP, Tre´panier S and Michaud D. An in-built proteinase inhibitor system for the protection of recombinant proteins recovered from transgenic plants. Plant Biotechnol J 2006;4:359–368. 164. Aalberse RC, Koshte V and Clemens JG. Immunoglobulin E antibodies that crossreact with vegetable foods, pollen, and Hymenoptera venom. J Allergy Clin Immunol 1981;68:356–364. 165. Aalberse RC, Koshte V and Clemens JG. Cross-reactions between vegetable foods, pollen and bee venom due to IgE antibodies to a ubiquitous carbohydrate determinant. Int Arch Allergy Appl Immunol 1981;66:259–260. 166. Bardor M, Faveeuw C, Fitchette AC, Gilbert D, Galas L, Trottein F, Faye L and Lerouge P. Immunoreactivity in mammals of two typical plant glyco-epitopes, core alpha(1,3)-fucose and core xylose. Glycobiology 2003;13:427–434. 167. Jin C, Bencurova M, Borth N, Ferko B, Jensen-Jarolim E, Altmann F and Hantusch B. Immunoglobulin G specifically binding plant N-glycans with high affinity could be generated in rabbits but not in mice. Glycobiology 2006;16:349–357. 168. Nuttall J, Ma JK and Frigerio L. A functional antibody lacking N-linked glycans is efficiently folded, assembled and secreted by tobacco mesophyll protoplasts. Plant Biotechnol J 2005;3:497–504. 169. Strasser R, Stadlmann J, Svoboda B, Altmann F, Glossl J and Mach L. Molecular basis of N-acetylglucosaminyltransferase I deficiency in Arabidopsis thaliana plants lacking complex N-glycans. Biochem J 2005;387:385–391. 170. Downing WH, Galpin JD, Clemens S, Lauzon SM, Samuels AL, Pidkowich MS, Clarke LA and Kermode AR. Synthesis of enzymatically active human alpha-L-iduronidase in Arabidopsis cgl (complex glycan-deficient) seeds. Plant Biotechnol J 2006;4:169–182. 171. Koprivova A, Stemmer C, Altmann F, Hoffmann A, Kopriva S, Gorr G, Reski R and Decker EL. Targeted knockouts of Physcomitrella lacking plant-specific immunogenic N-glycans. Plant Biotechnol J 2004;2:517–523. 172. Sourrouille C. Inactivation de l’alpha(1,3)-fucosyltransfe´rase et de la beta(1,2)-xylosyltransferase, en vue de la production de proteínes recombinantes d’inte´reˆt the´rapeutique chez la luzerne (Universite´ de Rouen), 2005, p. 164. 173. Sriraman R, Bardor M, Sack M, Vaquero C, Faye L, Fischer R, Finnern R and Lerouge P. Recombinant anti-hCG antibodies retained in the endoplasmic reticulum of transformed plants lack core-xylose and core-alpha(1,3)-fucose residues. Plant Biotechnol J 2004;2(4):279–287. 174. Petruccelli S, Otegui MS, Lareu F, Tran Dinh O, Fitchette A-C, Circosta A, Rumbo M, Bardor M, Carcamo R, Gomord V and Beachy RN. A KDEL-tagged monoclonal antibody is efficiently retained in the endoplasmic reticulum in leaves, but is both partially secreted and sorted to protein storage vacuoles in seeds. Plant Biotechnol J 2006;4:511–527. 175. Palacpac NQ, Yoshida S, Sakai H, Kimura Y, Fujiyama K, Yoshida T and Seki T. Stable expression of human beta1,4-galactosyltransferase in plant cells

146

176.

177.

178.

179.

180. 181. 182. 183. 184.

185.

186. 187. 188. 189.

190.

191. 192. 193.

modifies N-linked glycosylation patterns. Proc Natl Acad Sci USA 1999;96: 4692–4697. Misaki R, Kimura Y, Palacpac NQ, Yoshida S, Fujiyama K and Seki T. Plant cultured cells expressing human beta1,4-galactosyltransferase secrete glycoproteins with galactose-extended N-linked glycans. Glycobiology 2003;13:199–205. Fujiyama K, Palacpac NQ, Sakai H, Kimura Y, Shinmyo A, Yoshida T and Seki T. In vivo conversion of a glycan to human compatible type by transformed tobacco cells. Biochem Biophys Res Commun 2001;289:553–557. Saint-Jore-Dupas C, Nebenfuhr A, Boulaflous A, Follet-Gueye ML, Plasson C, Hawes C, Driouich A, Faye L and Gomord V. Plant N-glycan processing enzymes employ different targeting mechanisms for their spatial arrangement along the secretory pathway. Plant Cell 2006;18:3182–3200. Bakker H, Rouwendal GJ, Karnoup AS, Florack DE, Stoopen GM, Helsper JP, van Ree R, van Die I and Bosch D. An antibody produced in tobacco expressing a hybrid beta-1,4-galactosyltransferase is essentially devoid of plant carbohydrate epitopes. Proc Natl Acad Sci USA 2006;103:7577–7582. Seveno M, Bardor M, Paccalet T, Gomord V, Lerouge P and Faye L. Glycoprotein sialylation in plants? Nat Biotechnol 2004;22:1351–1352, 1352–1353. Zeleny R, Kolarich D, Strasser R and Altmann F. Sialic acid concentrations in plants are in the range of inadvertent contamination. Planta 2006;224:222–227. Jarvis DL. Developing baculovirus-insect cell expression systems for humanized recombinant glycoprotein production. Virology 2003;310:1–7. Wee EG, Sherrier DJ, Prime TA and Dupree P. Targeting of active sialyltransferase to the plant Golgi apparatus. Plant Cell 1998;10:1759–1768. Misaki R, Fujiyama K and Seki T. Expression of human CMP-N-acetylneuraminic acid synthetase and CMP-sialic acid transporter in tobacco suspension-cultured cell. Biochem Biophys Res Commun 2006;339:1184–1189. Paccalet T, Bardor M, Rihouey C, Delmas F, Chevalier C, D’Aoust M-A, Faye L, Vezina LP, Gomord V and Lerouge P. Engineering of a sialic acid synthesis pathway in transgenic plants by expression of bacterial Neu5Ac-synthesizing enzymes. Plant Biotechnol J 2007;5:16–25. Gasdaska JR, Spencer D and Dickey LF. Advantages of therapeutic protein production in the aquatic plant Lemna. Bioprocess J 2003;3:50–56. Schaefer D, Zryd JP, Knight CD and Cove DJ. Stable transformation of the moss Physcomitrella patens. Mol Gen Genet 1991;226:418–424. Schaefer DG and Zryd JP. Efficient gene targeting in the moss Physcomitrella patens. Plant J 1997;11:1195–1206. Schaefer DG, Bisztray G and Zry¨d J-P. Genetic transformation of the moss Physcomitrella patens. In: Plant Protoplasts and Genetic Engineering V, Bajaj YPS (ed), Berlin, Springer-Verlag, 1994, pp. 349–364. Vietor R, Loutelier BC, Fitchette A, Margerie P, Gonneau M, Faye L and Lerouge P. Protein N-glycosylation is similar in the moss Physcomitrella patens and in higher plants. Planta 2003;218:269–275. Franklin SE and Mayfield SP. Prospects for molecular farming in the green alga Chlamydomonas. Curr Opin Plant Biol 2004;7:159–165. Leon-Banares R, Gonzalez-Ballester D, Galvan A and Fernandez E. Transgenic microalgae as green cell-factories. Trends Biotechnol 2004;22:45–52. Mayfield SP and Franklin SE. Expression of human antibodies in eukaryotic microalgae. Vaccine 2005;23:1828–1832.

147 194. Lienard D, Tran Dinh O, van Oort E, Van Overtvelt L, Bonneau C, Wambre E, Bardor M, Cosette P, Didier-Laurent A, Dorlhac de Borne F, Delon R, van Ree R, Moingeon P, Faye L and Gomord V. Suspension-cultured BY-2 tobacco cells produce and mature immunologically active house dust mite allergens. Plant Biotechnol J 2007;5:93–108. 195. Elbers IJ, Stoopen GM, Bakker H, Stevens LH, Bardor M, Molthoff JW, Jordi WJ, Bosch D and Lommen A. Influence of growth conditions and developmental stage on N-glycan heterogeneity of transgenic immunoglobulin G and endogenous proteins in tobacco leaves. Plant Physiol 2001;126:1314–1322.


149

Porous silicon protein microarray technology and ultra-/superhydrophobic states for improved bioanalytical readout Anton Ressine1,, Gyo¨rgy Marko-Varga2 and Thomas Laurell1 1 Department of Electrical Measurements, Lund Institute of Technology, Lund University, P.O. Box 118, SE-221 00 Lund, Sweden 2 Department of Analytical Chemistry, Lund University, P.O. Box 124, S-221 00 Lund, Sweden

Abstract. One attractive method for monitoring biomolecular interactions in a highly parallel fashion is the use of microarrays. Protein microarray technology is an emerging and promising tool for protein analysis, which ultimately may have a large impact in clinical diagnostics, drug discovery studies and basic protein research. This chapter is based upon several original papers presenting our effort in the development of new protein microarray chip technology. The work describes a novel 3D surface/platform for protein characterization based on porous silicon. The simple adjustment of pore morphology and geometry offers a convenient way to control wetting behavior of the microarray substrates. In this chapter, an interesting insight into the surface role in bioassays performance is made. The up-scaled fabrication of the novel porous chips is demonstrated and stability of the developed supports as well as the fluorescent bioassay reproducibility and data quality issues are addressed. We also describe the efforts made by our group to link protein microarrays to matrix-assisted laser desorption/ionization timeof-flight mass spectrometry (MALDI-TOF MS), suggesting porous silicon as a convenient platform for fast on-surface protein digestion protocols linked to MS-readout. The fabrication of ultra- and superhydrophobic states on porous silicon is also described and the utilization of these water-repellent properties for a new microscaled approach to superhydrophobic MALDI-TOF MS target anchor chip is covered. Keywords: porous silicon, protein microarray technology, water-repellency, superhydrophobic, ultrahydrophobic, biomarker discovery, antibody microarrays, lysate microarrays, high-speed biomarker identification, MALDI-TOF MS, superhydrophobic target anchor chips.

Preface What is in common among such different phenomena as: a morning dew condensing in droplets on tree leaves before the sun gets high; a portion of wine climbing on the wall of a crystal glass by capillarity, accumulating there and sliding down in the form of tears (Marangoni effect); a water strider floating over the surface of a pond searching for the prey; a splash of coffee dropped down from a cup on a table and drying to stain? There is something general behind all of them – the surface. These all take place at the phase-separation boundary and they are all governed or controlled by surface properties. As we scale down from the physical world where the effects of gravity and inertia dominate our experiences, we find ourselves in a completely different Corresponding author. Tel: +46 462223653. Fax: +46 462224527.

E-mail: [email protected] (T. Laurell); [email protected] (A. Ressine). BIOTECHNOLOGY ANNUAL REVIEW VOLUME 13 ISSN 1387-2656 DOI: 10.1016/S1387-2656(07)13007-6


150

Fig. 1. Rain droplets sitting on the surface of a leaf, a miniaturized microfluidic

system, illustrates a behavior governed mainly by interfacial forces. The surface properties thus play a significant role.

situation. Gravity and inertia are no longer major prevailing players anymore. The surface properties are now of key importance and they start to control the behavior of the microscale system to a great extent, as illustrated by rain droplets sitting on a leaf surface (Fig. 1). What are the strategies we can learn from nature to tune the phenomena and processes at interfaces for our own practical benefits? One, and perhaps the most straightforward way, could be to change the surface energy by chemical modification. Another option, not so obvious but extremely popular in nature, is to exploit roughness/structure. The water strider can easily slide over the water surface, without sinking down, utilizing the texture of its superhydrophobic legs and thereby keeping the contact area of its body with water to a minimum, unable to break the water tension. The gas exchange, which occurs with relatively slow rates within our lungs by diffusion, is nevertheless extremely efficient overall because it takes place over a large total surface area of about 80 m2 provided by large numbers of very small air spaces (the alveoli). These are several examples of how the nature utilize surface structure.

151 Unlike the nature, we are only beginning to harness surface structure for our practical use. In the on-going trend of miniaturization, the art of designing surfaces will definitely play an increasing role. That is esspecialy true for the field of microarray technology where all processes, starting from microarray printing and probe immobilization, and finishing by the affinity binding of targets to immobilized probes, occur at the solid/liquid or solid/ gas interface. In this chapter, we explore the possibilities to utilize the 3D morphologies fabricated on the surface of silicon and to tailor them to the development of protein microarray chip technology. Introduction Recently, protein microarray technology emerged as a novel effective tool for protein screening in a parallel fashion [1–8]. As a descendent to the DNA microarray technology, the protein microarray concept is based on a large number of capture agents that selectively bind to the analytes of interest (proteins) on solid surfaces [9,10]). As already demonstrated, the protein microarrays allow fast, easy and parallel detection of thousands of addressable elements in a single experiment, operating only with minute volumes of sample [11,12]. In the last few years protein microarray technology has shown a great potential in basic proteomics research, diagnostics and drug discovery [13–29]. It has been applied to the analysis of antibody–antigen, protein–protein, protein-small molecule interactions as well as enzyme– substrate interactions. Detection of biomarkers using protein arrays is currently one of the most active areas of research directed toward clinical diagnostic applications [1,16,26,28,30,31]. In spite of this there is currently a limited use of diagnostic protein microarrays in clinical practice, explained by a number of methodological and technical challenges [9,32]. Omitting the questions relating to the medical reliability of biomarkers, the present chapter we will focus mainly on the development of technological aspects associated with the quality of bioassay performance where the properties of the microarray surface/substrate will play an essential role [8,33–35]. Following one of the trends in the protein microarray technology area toward the utilization of 3-dimensional surfaces, our group introduced a macro-/ nanostructured silicon for protein microarray applications [36,37] and showed that the surface morphology can serve as an effective mean to tailor the substrate properties for the demands set by the specific applications. Namely, these applications are the following: antibody microarrays with fluorescent detection [37,38]; reverse phase/lysate bioassays [39]; protein microarrays with dual-readout (fluorescent detection of bound molecules and subsequent protein identification with MALDI-TOF MS) [40]; superhydrophobic MALDI-TOF MS target anchor chips [41] and accelerated protein digestion and analyte enrichment in nanovials [42].

152 As it is quite often within the research situation that one can unexpectedly face interesting phenomena, which were not foreseen at the beginning of the study when trying to solve certain challenges. Similarly, during the development of protein microarray technology based on porous silicon, superhydrophobic states on macro-/nanoporous silicon were reported for the first time [43]. It was further realized that superhydrophobicity of porous silicon opened a research direction with new routes in the field of miniaturized bioanalytical technology and methodology where fluid/surface interaction is of major importance [39]. Also as in the case of DNA-arrays, complex issues pertaining to spot morphology, homogeneity of signal across the spot, microarray smearing and non-specific adsorption are not adequately addressed in the case of protein microarrays [34,35,44,45]. In the present chapter, we made an attempt to cover some of them and to demonstrate that surface morphology/structure is in strong relation to the microarray readout quality, and that the directed methodological design of the substrate structure can be a powerful tool to control processes at a microarray surface. Porous silicon – a novel 3D substrate for protein microarray technology The first observation of porous silicon layers goes back 40 years and was done by Uhlir and Turner [46,47]. Since that time porous silicon gained steady attention as an interesting material for different areas of research and industry, including MEMS and microelectronics (sacrificial layers [48–50], silicon-on-insulator [51,52]); optoelectronics (electroluminescence [53–55], solar cells [56–58], optical filters [59,60]); power technology (fuel cells [61,62]); medical therapy (orthopedics and implantation [63,64], drug delivery [65,66]); sensors [67] and biosensors [68,69]. The strong interest in porous silicon can be explained by several factors: its extraordinary material properties such as a high surface area-to-volume ratio; a pore geometry and morphology that can easily be altered during the fabrication process; optical properties that can be tuned; biocompatibility and biodegradability and the relatively straightforward process for porous silicon fabrication, compatible with standard microelectronic and MEMS techniques. The use of porous silicon as a 3D substrate for the biosensor applications has been intensively reported in the literature [68,70–78] and is inspired mainly because of the tremendous increase in an immobilization capacity of this material. Silicon porosified to a porosity of 80–90% can lead to an increase of the surface to volume ratio of up to 500 m2/cm3 and even more [79]. Examples of bioanalytical applications have been shown by Drott and co-workers where porous silicon was used for the immobilization of enzyme (glucose oxidase), which led to increase in catalytic turn-over rate up to 350 times [80] as compared to a non-porous surfaces. Further, work by the Laurell group demonstrated the beneficial properties of PS as a chip-integrated surface enlarging material for microreactors in Lab-On-A-Chip applications

153 [81–88]. Additional attraction of porous silicon is that it can provide a direct transduction mechanisms as for example have been reported in [89,90]. This was an approach similar to chemiresistors and chemically sensitive field-effect transistors (chemFETs). A change in refractive index of the medium upon molecular interactions of species in solution with immobilized ligands or receptors within the porous silicon matrix can also be used as a simple and effective method for biomolecular sensing [71,73]. Another highly interesting and emerging application area for porous silicon is driven by the ability of PS to directly interface with living cells [91–94]. Fabrication of porous silicon One of the methods commonly used for porous silicon fabrication is silicon electrochemical etching. In this procedure, the monocrystalline silicon wafer is placed between two electrochemical cells filled with electrolyte solution containing hydrofluoric acid as illustrated in Fig. 2. The electrolyte provides liquid contacts with both sides of the silicon wafer. When a current is passed through the system, the anodic side of the silicon wafer gets porosified. The mechanism of pore initiation is still under debate; however, there are suggestions that defects or slight variations in surface potential due to defects or doping atoms are the starting point of the pores (Fig. 3). The main reaction during PS formation, assuming a hydrogen terminated Si surface, was suggested by Lehmann and Go¨sele [95] SiH2 þ 2F þ 2hþ ! SiF2 þ H2 ðdivalent dissolutionÞ

ðIÞ

SiF2 þ 4HF ! 2hþ þ SiF2 6 þ H2 ðin solutionÞ

ðIIÞ

More details on the fabrication process can be found in refs. [79,96].

Fig. 2. Porous silicon fabrication by electrochemical etching in a two-compartment cell. A 3 in. silicon wafer is placed between two compartments filled with electrolyte solution, providing liquid contacts to the wafer. As the current is passed through the substrate pores are formed on the anodic side of the wafer.

154

Fig. 3. Chemical pathway of silicon anodization and schematic drawing of the prin-

ciple for the formation of the nano-/macroporous silicon network. Formation process includes the following stages: (A) migration of charge carriers in the bulk silicon to the electrolyte interface, (B) random pore nucleation, (C) macropore formation and propagation and (D) nanoporous side branching. Reprinted with permission from Anal. Chem. 2003, 75(24), 6968–6974. Copyright 2003 American Chemical Society.

Tuning porous silicon morphology Porous silicon fabrication by electrochemical etching is dependent on a multitude of process parameters (current density, etching time, crystal orientation, silicon dopant type, doping level, illumination, electrolyte composition, temperature, surface roughness) [97,98,99]. Changing any of these parameters will affect morphology and/or geometry of the obtained porous silicon layers. In this respect, the multidimensional process parameter space offers many ways to change the characteristics of the PS and gives the possibility to tune the surface properties for the different applications [79,100,101]. In other words despite the one common term ‘‘porous silicon,’’ the layers produced in a different etching mode can differ significantly by their morphology and geometry, as illustrated by Fig. 4, as well as by physical and chemical properties.

155

Fig. 4. SEM images of different pore sizes and morphologies that can be obtained on

silicon wafer by electrochemical etching. Starting from micropores on sample A (pores are not distinguishable at the present length scale) and turning into macropore structure on samples B– H. The pore diameter, shape and the thickness of porous layers are also different. Scale bar on all images approximately 5 mm.

This makes it necessary to introduce a classification of the different pore types. According to the IUPAC standard, the three categories are distinguished by looking at average pore diameter and average distance between pores (pore geometry) [79]. Micropores – pore diameters and pore distances o10 nm. Mesopores – geometries in the 10–50 nm region. Macropores – geometries in the 450 nm region. Also the term nanoporous is very frequently used in literature to describe materials with characteristic sizes less than 100 nm. While the term geometry of the pore layer describes the parameters such as characteristic (average) pore diameter and distance between pores, it does not reflect information on a pore shape and orientation. For these purposes the term morphology of the pore layer is introduced. This term is used as the collective identifier for properties like the shape of the pores, level of pore branching, the range of pore layer fractality, orientation and interaction of pores. Transmission and scanning electron microscopy (TEM and SEM) are the methods used to directly image PS layers and achieve information on morphology and geometry of pore layers. Despite of the large number of research efforts aimed to tailor porous silicon for biosensor applications, there has been no attempts to utilize porous silicon as substrate in protein microarray technology. The present chapter demonstrates that the properties of porous silicon layers can be

156 successfully tailored to make porous silicon a promising material in protein microarray chip technology. Microarrays on porous silicon Our initial motivation to utilize porous silicon as substrate for protein microarray applications was driven by its vast surface area enlargement and thus presumed high loading capacity of antibodies per projected chip area. Macro-/nano-textured silicon was introduced as a novel material for protein microarray applications and shown that it was possible to tailor its properties for effective antibody immobilization [36,37]. We also discussed the criteria essential for porous silicon layers to allow optimal bioassay performance. Briefly, utilizing macro-/nanoporous silicon as substrate for printing protein microarrays can offer the following advantages. Low surface wetting (while still providing mild immobilization condi-

tions – a hydrophilic surface at the molecular level). Small spots (increased immobilization density over a spot, improved

reaction kinetics, high-density arraying). Homogeneous probe molecule coverage (uniform fluorescence intensity

over a spot, improved data quality). Low fluorescence background. Biocompatibility (immobilization and/or adsorption of biospecific bind-

ers with maintained affinity and selectivity, allowing protein digestion to be performed directly on the surface, allowing complex sample analysis such as blood). Compatibility with MS detection. While testing different porous layer morphologies for microarray printing, we have noticed that microfluidic properties are of key importance for good bioassay performance. For example, it is more beneficial to use substrates exhibiting low wetting (reflected by high values of water contact angle for drop resting on the surface) because of the following reasons: (1) smaller spot sizes intrinsically allow more dense printing without cross talk between adjacent spots; (2) the binding equilibrium is reached more rapidly with a reduced spot size; (3) reducing spot size/contact area during printing leads to an increase in immobilization density (shrinking the spot size by a factor 2 will yield 4-fold increase in probe molecule density over the smaller spot area); it turns into reduced consumption of immobilized reagents. From this respect, it is also important to note that the immobilization of probe molecules (antibodies) on hydrophobic surfaces (polytetrafluoroethylene/ Teflon, silanized glasses, etc.) is generally not the best strategy (even if it leads to reduced spot sizes) due to the protein denaturation at the surface, which is promoted by the surface hydrophobicity [35,102]. It has been shown, for

157 example, that IgG adsorbs on a hydrophobic Teflon surface by attaching its Fab parts, rather than the Fc fragment [103]. As a consequence, the Fab fragments undergo structural rearrangements, whereas the Fc does not, and subsequently compromise binding efficiency of the targeted protein. The differences in the degree of denaturation between Fab and Fc are ascribed to the higher degree of hydrophobicity of the Fab fragment. Also, as has been reported, many proteins adsorbed on polystyrene or silicone substrates suffer adsorption-induced conformational changes (ACC) and are partially or largely denatured [104]. Moreover, hydrophobic surfaces exhibit a higher degree of unspecific binding in comparison to hydrophilic support media [105,106]. Therefore, these surfaces might be less suited to systems in which highly complex samples should be analyzed, such as blood samples and protein lysates from cell lines or tissues. The presented 3D-textured macroporous silicon, while exhibiting lowwetting behavior (water contact angles on the surface can be more than 1101), still provides mild hydrophilic conditions for immobilization and increased immobilization capacity and is potentially applicable to complex sample analysis (cell and tissue lysates, blood) [37].

Antibody microarrays Our studies also revealed that morphologies implementing macroporous structures are most beneficial in regards to the fluorescence bioanalysis performance. There was a minimum of background fluorescence observed on the presented PS structures, conversely to the porous silicon composed only by nanoporous morphologies. The fluorescence readout of a model IgG-binding assay performed on macroporous silicon substrate is demonstrated in Fig. 5. An antibody directed against IgG was printed onto the surface (140 amol/ spot) by inkjet deposition of a single 100 pL droplet. The simple sandwich assay was performed according to the protocols schematically illustrated in Fig. 6. The chip was incubated in human blood plasma spiked with antigen, IgG (FITC-labeled), at a level of 10 pM. The spot density of the microarray is over 14,000 spot/cm2. The standard microarray protocol for antibody binding assays has the following steps (Fig. 6). Capture probes/antibodies are printed on the surface of microarray by non-contact spotting methods using, e.g., piezoelectric dispensing. Single nanoliter to picoliter droplets are delivered for each spot position. After spotting, the microarray is incubated with the so-called blocking solution. This is done to block adsorption sites on the surface and, thus, to prevent unspecific binding during the next incubation steps. The microarray subsequently carefully washed to remove loosely bound probe molecules. The microarray is finally incubated with the sample solution containing proteins of interest/targets. During this step binding of target

158

Fig. 5. IgG assay performed on macroporous silicon chip according to the protocol

illustrated in Fig. 9. Antibody directed against IgG was arrayed onto the surface (140 amol/spot) (inkjet deposition of a single 100 pL droplet is seen in the top right inset). The chip was incubated in human blood plasma spiked with antigen, IgG (FITC-labeled), at a level of 10 pM. The spot density was over 14,000 spot/cm2. SEM image of cross section of the porous chip (middle right, scale bar is 5 mm). Low inset in right shows a fluorescence intensity profile made along the light gray line. Reprinted from [173] with permission. Copyright 2007 Humana Press.

molecules to immobilized probes occur. A final washing step is applied afterwards and the chip is analyzed. As a first effort to evaluate the detection limit and dynamic range in a setting relevant to future clinical applications with the new porous silicon protein microarray surface, we analyzed fluorescence (FITC) labeled PSA added at concentrations ranging from 0.7 ng/mL to 7 mg/mL to serum [37]. Monoclonal IgG 2E9 specific for PSA [107] were arrayed on the porous silicon protein chip surface and the chips were processed according to the specified protocol. Briefly, antibody activated porous silicon chips were washed, blocked to avoid unspecific adsorption, washed, then incubated with the human blood serum spiked with PSA, and finally scanned in the fluorescence confocal set-up. Extracts of these findings are shown in Fig. 7(a–c), where: (a) represent the limit of detection (3 times noise level) at approximately 0.7 ng/mL (26 pM); (b) shows signal intensities at E70 ng/mL (2.6 nM) and (c) at E7 mg/mL (260 nM). The data obtained in the study demonstrated that both limit of detection (E0.7 ng/mL PSA) and dynamic range (103–104-fold) of the porous silicon high-density protein chip surface corresponded to those required to allow for discrimination of men with very early signs of either malignant or benign prostate disease conditions, as judged from population-based geometric mean PSA-levels in serum [108]. We also addressed the issue of reproducible manufacturing of the protein chip surfaces, as well as data quality and spot-to-spot variation analysis.

159

Fig. 6. Typical bioanalysis procedure for capture protein microarray. Antibodies are

immobilized on the microarray surface, microarray blocked to prevent unspecific protein adsorption to the surface, washed and incubated with the target analyte solution.

160

Fig. 7. (a) PSA assay in blood serum spiked with 26 pM (E700 pg/mL) FITC labeled

PSA; (b) 2.6 nM (E70 ng/mL), FITC labeled PSA in blood serum; (c) 260 nM (E7 mg/mL) FITC labeled PSA in blood serum. Cross-section scans along light gray line in the corresponding (left) microarray images. Reprinted from [37] with permission. Copyright 2005 Humana Press.

161 Reproducible manufacturing was accomplished as demonstrated by the low readout variation when standard IgG bioassays were performed at 100 pM antigen level on a series of protein chips scanned at widely different locations within a silicon wafer, as well as between different wafers from two different manufacturers. The relative standard deviation (RSD) of fluorescence spot intensity within an array on a chip was less than 20%. Mean spot intensity RSD was 19% for all 25 microarray chips in the study. Within manufacturer lot RSDs in chips from either manufacturer were o15% of mean spot intensity. Lysate microarrays In a direct detection immunoassay protocol, or another name – reverse phase microarray, a complex solution (e.g., cellular lysate) is printed onto a substrate surface [24,109–111]. The microarray is incubated with fluorescently labeled antibody solution. This type of protocol allows multiplex bioassays to be performed by spotting different samples in each position, enabling screening of many samples simultaneously. To evaluate the performance of macroporous chip in a lysate format, we performed analysis of cyclin E, which is a candidate for a prognostic biomarker in patients with breast cancer [112–114]. The level of cyclin E, part of a molecular network that controls the cell cycle, is increased in breast-cancer cell lines. In breast-cancer tissue, high levels of cyclin E are correlated with a poor outcome, whereas low levels are correlated with a good outcome. To demonstrate the proof of principle of the reverse protein array approach, we printed the cell lysates from two cell lines (one with the negative production of cyclin E and the other overproducing cyclin E, known from the independent data set obtained by Western blot technique) on to the porous chip. The lysate was printed in a series of dilutions and the surface was then blocked, washed and incubated with FITC-labeled anti-cyclin E antibody solution. The data (Fig. 8) show the absence of unspecific binding (lower trace); uniform spot intensity profiles and clearly demonstrates the possibility to work on a porous silicon surfaces in a lysate microarray format. Up-scaled fabrication of the macro-/nanoporous protein microarray surfaces and corresponding reproducibility performance A fundamental request for the success of a new microarray surface is the fact that the surface properties can be readily reproduced on chips in large numbers and long batch series. We have previously reported [37] the development of a process for up-scaling the fabrication of macro/-nanostructured porous silicon microarray chips to wafer scale level. Three-inch wafers were porosified by electrochemical dissolution in an HF solution. The porosified wafers were diced into 6 6 mm dice, producing about 70 chips from a single wafer.

162

Fig. 8. Lysate microarray protocol on porous silicon – proof of principle. Two cell lines (one overproducing cyclin E and the other is the negative control) were printed on the macroporous silicon chips in different dilution series. The chips incubated with FITC-labeled anti-cyclin E antibody solution. Negative cell line shows no unspecific biding, while successive determination of cyclin E in another cell line is demonstrated.

The chips were further fitted on a standard microscopic glass slide format, as shown in Fig. 9, optionally allowing employment to commercially available arraying stations and confocal scanners. To investigate the reproducibility of the up-scaled porous silicon fabrication, we performed the same bioassay on the chips made from different positions (A–E) on porosified 3 in. silicon wafers (for experimental details, see [37]). To represent the whole surface chips were taken from the positions denoted A–E as shown in Fig. 9. The study aimed at revealing if major variations in bioassay response could be detected in relation to the individual spatial position of the chip (A–E) on the wafer and between different wafers (1–5) (chips numbered as 1A, 1B, y 5E corresponding to the cut position). This in turn could possibly be traced back to variations in surface morphology and variations in the electrochemical anodization process. For that purpose, a series of model antibody-binding assays with the same protocol

163

Fig. 9. Wafer scale processing for the fabrication of macro-/nanostructured silicon

microarray chips from 3 in. silicon wafers. A,B,C,D indicates the positions on the wafer from which chips were taken for the reproducibility study (see Fig. 10). Reprinted from [37] with permission. Copyright 2005 Humana Press.

was made on the number of chips. The reproducibility for chips within the wafer as well as between different wafers was found quite satisfactory. Contributions to observed standard deviations may originate from the manual handling (washing and application of incubation solution, etc.), and nonperfect assaying conditions such as stagnant chip incubation (no agitation). The mean fluorescence intensity for all 25 chips in the reproducibility study was 467789 [AU], i.e., an RSD of 19%. When investigating performance from chips derived from silicon delivered by the same supplier, either Topsil or Addison, the corresponding RSDs were 13% and 10%, respectively. This was further explained by slightly different conductivities in the silicon substrates delivered by the two suppliers (Fig. 10). As an alternative method to characterize the quality of the porous silicon protein chip and the performed bioassay, the relative deviation to the mean spot intensity was estimated for each of the chips taken from two wafers (wafers 1 and 2), Fig. 11. The width of the relative deviation in mean spot intensity is indicative of the spot-to-spot reproducibility and is generally observed to be below 70.2. Other work on protein microarrays report similar data also claiming a good spot-to-spot reproducibility [115].

164

Fig. 10. Comparison of the fluorescent mean spot intensity after a reference IgG assay on the A,B,C,D,E chips (see Fig. 9) and wafer to wafer (1–5) variations. Error bars indicate the standard deviation in fluorescent spot intensity for 30 spots within the corresponding chip (1A–5E). Reprinted from [37] with permission. Copyright 2005 Humana Press.

Fig. 11. Relative deviation of the single-spot intensity from mean value of the chips A,B,C,D and E. The chips originated from two wafers made at different occasions. Microarrays with immobilized aRIgG-FITC were incubated in plasma spiked with 100 pM of RIgG. Thirty spots on each chip were analyzed. Reprinted from [37] with permission. Copyright 2005 Humana Press.

165 The comparison of the developed 3D substrates with existing and commercially available substrates for protein microarraying has been also studied in [38]. The issues concerning immobilization homogeneity and the quality of spot coverage are of key importance for optimal microarray performance and they are strongly related to the phenomena associated with the surface wetting and droplet evaporation. The wetting behavior and droplet evaporation during the microarray printing will affect the quality of the output data. To understand its influence on microarray performance, we will consider some of those phenomena in the next paragraph. Also we will show that by tuning the porous silicon morphology, it is possible to significantly improve microarray data quality.

Effect of porous texture on microarray data quality Effects at the surface associated with wetting An important but often underestimated aspect of microarray analysis is the quality of the spots printed onto the microarray substrate. Any error introduced by imperfect spotting inevitably will impair the data output. When measuring the amount of the analyte bound to the test spot nonhomogenous fluorescence spot intensity will hamper the meaningful biological signal, reducing data output quality [116]. The common artifacts here are ring-like spot intensity profiles, smearing fluorescent spots, spikes, ambiguities introduced by salt crystallization (see Fig. 12). These spot imperfections all reduce the microarray experiment reproducibility and drastically increase spot-to-spot and experiment-to-experiment

Fig. 12. Microarray fluorescence spot artifact shortcomings to quantitative data

analysis. (A) ring spot profile, (B) spot smearing, (C) spike, (D) ambiguity because of the salt crystallization and (E) both ring profile with smearing.

166 variations. This also raises the demands to introduce sophisticated data processing software to evaluate fluorescence spot intensity. One of the fundamental consequences of miniaturization is the prevalence of interfacial forces (or energies) and their influence on microfluidic systems, as illustrated by the raindrops sitting on a leaf surface, Fig. 1. While inkjetprinting microarrays by depositing analytes on a surface in picoliter droplets, with a characteristic diameter of 60 mm, the gravity force becomes negligible and the droplet behavior (during the deposition and drying) is completely governed by the interfacial properties of phases involved (solid – our substrate, liquid – buffer with dissolved proteins and vapor phase – ambient air atmosphere). Since the properties of the vapor and liquid phases in our case are to great extent fixed and cannot be changed (e.g., the liquid phase is dictated by the employed biological buffers with certain composition, pH, ionic strength, etc.; the vapor phase is commonly ambient air), we will mainly focus on the properties of the solid substrate surface and the possibility to tune this for efficient microarray performance. The spot quality will influence spot-to-spot variations within a protein array and the reproducibility of data across experiments, as is also the case for DNA microarrays [117]. Experimental studies of parameters influencing the spot quality are found in literature [117–122]. The problem arises from the inhomogeneous repartitioning of the immobilized probe molecules inside the spot (spot morphology) formed during the evaporation of the droplet (rings, spikes, etc.). These inhomogeneities lead to a shortcoming regarding the data processing and the interpretation of the signals [123,124]; all in turn reduce the reliability of microarray results. This is especially critical for the dedicated biomarker analysis microarrays where the precise quantitative determination of the analytes bound is necessary. Ideally, each spot should be identical with a uniform distribution of immobilized probes. Since the droplet vaporization plays a significant role on homogeneity of antibody coverage, we will herein briefly consider some of the effects associated with the droplet drying as well as the possibility to control it by means of the surface morphology. ‘‘Coffee-stain effect’’ and its effect on a spot intensity profile One of the effects hampering microarray spot quantitation – known as a ‘‘coffee-stain effect’’ or ‘‘coffee-ring effect’’ – is associated with the surface properties. The name of this effect originates from the phenomenon with a drop of coffee drying on a surface. One can find that after solvent evaporation, a ring-like coffee pattern will be formed with most solids deposited at the spot edge and very little deposition at the center of the spot (Fig. 13). The mechanism of such pattern formation was studied by Deegan and co-workers [119,124] and is based on evaporation-driven convection. Briefly the mechanism is the following. While, the droplet volume shrinks and, if the

167

Fig. 13. (A) A photograph of the dried coffee drop on a surface. When the drop

dries, it forms a ring-like pattern with elevated solid deposition at the rim of the spot and lower amount of the solid at the center. (B) Non-homogenous distribution of the solid deposit across the dried spot (the dashed white line on left image). This phenomenon is known as coffee-stain effect.

Fig. 14. Inkjet antibody deposition onto planar surface (100 pL volume drop was

spotted). (A) The fluorescent spot distribution of immobilized antibodies on a planar substrate (aminosilanized glass). The coffee-stain effect results in non-homogenous molecular immobilization. (B) Evaporation-driven convection that leads to ring analyte deposition on planar substrates with pinned drop contact line.

contact line of the droplet remains rigidly pinned to the surface, the droplet shape undergoes a change of its contact angle during the evaporation process. Evaporation is most effective at the rim of the droplet, where the probability that an evaporated solvent molecule recondenses to the droplet is less than that near the crest of the droplet. Since the contact line is pinned to the surface, the convection flow from the center of the droplet to the periphery brings more solutes to the edge of the spot (Fig. 14). This effect is quite general and is observed for many solute/solvent systems. It has also some technological relevance. Droplet-based deposition of materials is a very common process ranging from inkjet printers [125] to DNA [120,123] and protein microarrays [37]. In particular in the latter cases, the coffee-stain effect – resulting in non-homogenous molecular immobilization (Fig. 14A), the so-called ‘doughnut shape’ – is a well-known problem

168 in the biotechnology community. Figure 14B shows an evaporation-driven convection inside the evaporating droplet driving solvent to the edge and Fig. 14A is a resulting deposition pattern of FITC-labeled antibodies after dispensing a droplet of 100 pl volume onto the glass substrate. The following strategies have been proposed to eliminate the coffee-stain effect: heating the substrate during printing procedure and using special protein printing buffers containing a mixture of ionic and polymeric materials to suppress evaporation-driven convection [122,126–128]. Since the heating is not applicable to protein solutions for the reason of protein denaturation, the second is more appropriate way but still has limitations. The additives can reduce probe molecule immunogenicity, and an extra bufferexchange step is necessary. In [36,37], we rather show that texturizing the surface of the silicon substrate we can effectively suppress non-homogeneous coverage (Fig. 15).

Fig. 15. Introducing a texture on the substrate surface can effectively suppress/improve immobilization homogeneity. (A) Glass slide, (B) aminosilanized glass and (C) macro/nanostructured silicon chip. 100 pL droplets of RIgG-FITC solution in PBS were spotted. Fluorescence images illustrate improved spot homogeneity, eliminated coffee-stain shape and reduced spot size as a result of the introduced surface morphology. Reprinted from [37] with permission. Copyright 2005 Humana Press.

169 The suppression of the coffee-stain effect on low-wetting PS surfaces is explained by the fact that the deposited capture probe volume is hydrophobically confined to a minimized surface contact/spot area and as the evaporation rate from the perimeter (solid/gas/fluid interface) is low due to vapor saturation (CA 44901) as compared to the areas of the drying droplet with a free-fluid gas interface. The dissolved affinity probe is therefore not enriched at the solid/fluid/gas interface. Also, the advection inside the droplet develops a rotational flow that maintains the analytes well mixed throughout the drying process as illustrated in Fig. 18B. As the droplet drying comes to completion, a high and even concentration over the drying spot drives the binding equilibrium of the affinity probe to saturation across the whole spot surface. Effect of salty buffers and salt crystallization Conventional buffered solutions (PBS, etc.) used for protein storage and microarray spotting often contain a visible amount of salt. While the droplet is drying, the salt reaches its critical concentration and suddenly crystallizes at a nucleation point (usually some defect on the surface), which leads to irregular spot homogeneity as shown in Fig. 16A. The repartitioning of molecules follows the morphology of the salt spot and homogeneity is affected by buffer salt crystallization. To demonstrate that a PBS solution containing fluorescently labeled protein molecules (IgG) was spotted in 100 pl droplets on the two different surfaces: (A) -planar glass substrate and (B) -3D macro-/nanoporous silicon. The arrays were then blocked and washed to remove loosely bound molecules. The resulting fluorescence images show the distribution of immobilized molecules in Fig. 16.

Fig. 16. Spotting IgG labeled with FITC in PBS solution on (A) planar microarray

(glass) and (B) 3D porous silicon chip. Salt crystallization during printing procedure leads to the inhomogeneous probe immobilization on a planar substrate (A), while 3D porous structure suppresses salt crystallization (B).

170 This effect is also a drawback when chemical immobilization is performed via probe spotting (proteins, DNA, etc.) on a preactivated surface. The introduction of a macroporous texture on the surface of a silicon chip suppresses buffer crystallization and yields a more homogeneous spot coverage. The absence of larger crystal formations that interfere with the probe immobilization is hypothesized to be explained by a more homogenous and dense distribution of nucleation sites across the surface in the highly differentiated and fractalized porous silicon layer. Spot smearing effect Spot smearing is another effect that has adverse influence on the spot homogeneity and image quality as illustrated by Fig. 17. This is especially a drawback when physical adsorption is used as the immobilization method. After the probe molecules are spotted onto the surface, the blocking/washing step proceeds. During this step molecules loosely bound to the surface

Fig. 17. Spot smearing is another common artifact for planar microarray substrates (A). Texturizing the surface drastically improves spot coverage (B) (macro/nanoporous silicon substrate).

171 resolves in the blocking/washing solution and may readsorb in the area close to the initial spot. Figure 17 (A) shows a schematic illustration (left) of probe molecule readsorption and the resulting fluorescent image of the probe molecule distribution (right). This effect can be eliminated completely by texturizing the surface as seen on the image of the surface of the macroporous silicon substrate (Fig. 17B).

Marangoni effect To fulfill the description of the processes in the evaporating sessile droplet, the Marangoni effect or convection phenomena associated with the surface tension gradient over the droplet should also be mentioned [129,130]. Marangoni effects, e.g., manifested as ‘‘tears of wine,’’ were observed as early as in the 1800s [131]. In a wine glass, the evaporation of alcohol generates a surface tension gradient, which produces a traction on the wine surface causing the wine to climb up the side of glass where it forms a thin film. As the wine accumulates, a bulging rim of liquid forms along the top of the film, which eventually pinches into droplets that roll under their own weight, like tears, back into the wine. The Italian physicist Marangoni gave a detailed description of the movement of a liquid surface induced by a surface tension gradient, generated either by a composition or a temperature variation along the free surface. In a sessile evaporating drop Marangoni stress (surface tension gradient over the droplet) can be induced by the presence of temperature variation near fluid interface or concentration variation caused by solvent evaporation. These effects can lead to the circulating convective flows inside the droplet as was described by Hu et al. and Savino et al. [131,132] (Fig. 18B). The circulating convective flows can lead to intensive mixing of the analyte during evaporation helping to overcome mass transport limitations and can be the counter force to the coffee stain forming convection flow (Fig. 18A) [121,131].

Fig. 18. (A) Schematic of the internal advection in low contact angle droplets and enrichment of dissolved species at the droplet perimeter (B) the corresponding advective situation for a high contact angle drying droplet, resulting in a homogenous distribution of analyte in the droplet throughout the whole drying process. Reprinted from [173] with permission. Copyright 2007 Humana Press.

172 The complete understanding of all these phenomena and the surface effect upon them can provide better control over probe immobilization when the non-contact printing methods are used for microarray fabrication.

Superhydrophobic and ultrahydrophobic states on porous silicon In recent years, the field of superhydrophobic and ultrahydrophobic materials has been given increased attention as reflected by the growing number of research publications (Fig. 19). The majority of articles describe new types of water-repellent surfaces and the methods for their fabrication [133–144]. Only few papers concern the utilization of these interesting properties in different application areas [145–150]. One can expect in a close future the strong growth of research efforts toward different applications, as the field has matured and the number of cheap and robust methods for superhydrophobic surface fabrication has been described. To our opinion, the use of super- and ultrahydrophobic surfaces, surfaces with patterned or stimuli responsive wetting can bring new unprecedented possibilities in the fields of microfluidics, lab-on-a-chip systems and surface-based bioanalysis. In the following section, we will describe the fabrication of super-/ultrahydrophobic states on porous silicon and their application for protein bioanalysis. The attention to the self-cleaning and nanostructured surface of the Lotus flower leaf has triggered intense developments in the materials research and nanotechnology field [151–161], aiming at finding new ways of mimicking natures own strategies of controlling solid surface/fluid interface properties.

Fig. 19. Number of published articles containing either the concept superhydrophobic or the concept ultrahydrophobic per year according to the search in the Chemical Abstracts and Science Citation Index.

173 Superhydrophobic surfaces based on either randomly organized surfaces [151–155,158,159,161–164] or ordered nanostructured surfaces [157,165–169] have been reported. Ordered superhydrophobic surfaces are commonly obtained by means of nanolithographic and nanopatterning techniques, while randomly ordered surfaces are readily accomplished by bulk surface treatments that provide extremely differentiated/fractalized and, thus, high surface area substrates. The latter approach thus offers simple means to generate large areas with superhydrophobic properties. An example of this was presented by Erbil H.Y. et al. in the direct polymerization of polypropylene for bulk production of randomly ordered nanostructured super hydrophobic surfaces [152]. Also the use of sol–gel processing and self-assembly for fabrication of randomly ordered superhydrophobic surfaces have been reported [154,158,170] as well as electrochemical deposition [171], electrospinning [151], microwave plasma-enhanced CVD [172]. The superhydrophobic states on porous silicon were first reported by our group [43], suggesting this material as a promising substrate for the microfluidic and bioanalytical applications [41,173]. Lately, it was demonstrated that superhydrophobic states on porous silicon can also be achieved by a combination of electroless chemical etching with alkylsilane self-assembly [174] and also by two-step silicon etching technique [175]. In the section below, we demonstrated that the water repellency of the macro-/nanoporous structured silicon surfaces strongly depends on the etching parameters and resultant surface morphology. The adjustment of the porous morphology and geometry can be regarded as a convenient way to control wetting behavior of the microarray substrates and can bring an interesting insight to the role of the surface in various bioanalytical applications. The possible applications of ultra-/superhydrophobic porous silicon in the next generation nanobiotechnology applications (protein microarrays and matrix assisted MS protein identification) outlined in the last sections.

Superhydrophobicity. The effect of roughness on solid surface wetting The interesting property of certain surfaces to repel water or in other words superhydrophobicity has been discovered for quite a long time ago, for example, C.V. Boys noticed in 1902 that water deposited on a layer of lycopodium rolls itself up into perfect little balls [176] as quoted by Callies et al. [135]. Although, the actual explosion of the research in that field happened in the late 1990s, following two achievements: firstly, a systematic study of the water repellency of plants by two German botanists, Barthlott and Neinhuis, who emphasized the role of micro-textures on the surface of the leaves to promote such an effect [177]; and secondly, the making of fractal hydrophobic surfaces by T. Onda and co-workers, who reported contact angles as high as 1741 [178]. The term superhydrophobicity generally refers to the

174

Fig. 20. Sessile drop on a solid surface forms a contact angle with the surface as a

result of interfacial tensions interaction.

surfaces, which exhibit a contact angle higher than 1501 for water droplets and ultrahydrophobicity – higher than 1701. The sessile drop (the drop sitting on the solid surface) will maintain its spherical shape if it is small enough to neglect the flattening action of gravity (which almost is the case for droplet less than 1 mm in diameter) as illustrated in Fig. 20. The drop contacts the substrate surface on an area of diameter D with a borderline (the so-called contact line) where the three phases of the system (vapor, liquid and solid) intersect. Contact angle, y is the angle formed by a liquid at the three-phase boundaries. The shape of the drop is controlled by the three forces of interfacial tension shown in Fig. 20. y is a quantitative measure of the wetting of a solid by a liquid, but it can also be used as a quality control to measure in surface treatments and surface cleanliness. The value of the contact angle was first discussed by Young [179]. Each interface draws the contact line so as to minimize the corresponding surface area, so that balancing the surface tensions on the direction of potential motion (i.e., the horizontal) yields a relation presented by Young: cos y ¼

gSV0 gSL gLV

ð1Þ

where gIJ is the surface tension (i.e., energy per unit surface) of the interface IJ, and the letters S, L and V designate the phases solid, liquid and vapor. The designation 0 above the gSV0 shows that the solid surface must be in equilibrium with the saturated vapor pressure. Young’s equation represents an idealistic situation where it is assumed that the surface is planar (without any roughness) and chemically homogeneous. For the case of rough surfaces it has been realized that texture can affect wetting. Two mathematical models corresponding to the two different wetting regimes have been proposed to describe the influence of roughness on surface. According to the Wenzel model [180] the liquid drop sits on the

175

Fig. 21. Two wetting regimes – ‘‘sticky’’ (A) and ‘‘slippery’’ (B) – according to the

models proposed by Wenzel and Cassie (see explanation in the text).

surface filling up all grooves (Fig. 21A), and thus roughness enhance the contact area between liquid and the solid. This, the so-called ‘‘sticky’’ wetting regime describes the effective contact angle on the rough surface y*as following: y ¼ r cos y

ð2Þ

where y is the contact angle on the planar surface and r a ‘‘roughness factor,’’ which is the ratio between the actual surface area and the geometric projected area. According to the Wenzel model, the roughness will enhance water repellency for repellent surfaces (for which the planar contact angle is more than 901) and will enhance hydrophilicity for the surfaces with planar contact angles less than 901. In contrast to the Wenzel model of wetting, the model suggested by Cassie [181] refers to the so-called ‘‘slippery’’ regime of the wetting behavior. In this situation, the contact area of the liquid with the solid is reduced and the drop sits like on air-pads (Fig. 21B). The drop can in this case be treated as being in contact with a heterogeneous surface composed of actual solid and air fractions. The Young’s equation can be rewritten for the case of a heterogeneous surface composed of two fractions j1 and j2 each having its own planar contact angle y1 and y2 producing total effective contact angle y* equal to cos y ¼ j1 cos y1 þ j2 cos y2

ð3Þ

When j2 represents the area fraction of the trapped air with a direct contact with liquid, then js becomes the fraction area of the solid in contact with the drop and becomes js ¼ j1 ¼ 1j2 (the y2 then is 1801 and js – area fraction of the solid–liquid interface with planar contact angle y). Equation (3) can be rewritten as the following: cos y ¼ js ðcos y 1Þ

ð4Þ

According to the Cassie mode, the texture will enhance water repellency on the non-planar surfaces requiring that y* 4 y. The equation predicts an

176 enhancement of hydrophobicity, and a jump in the contact angle can often be observed once air trapping occurs. Despite the simple mathematical models, the situation on the real surface can be much more difficult to understand because these two wetting regimes can coexist, as, e.g., was proved by Lafuma and Que`re` [182,183] for substrates with moderate roughness (r E 2) and hydrophobicity (yE100). And moreover, intermediate states (i.e., hemi-Wenzel and hemi-Cassie) might also exist on real complex surfaces, for example, with dual-scale roughness, i.e., hierarchical structure (which is usually the case for naturally found leaves of water repellent plants) [184,185]. In this respect particularly interesting is the situation when hydrophilic material (with planar yo901) is made hydrophobic by texture, as for example was discovered in the superhydrophobic porous silicon experiments performed by our group and was also reported by Teare et al. and Woodward et al. [186,187], who described that alkanes deposited on fluorinated substrates could exhibit contact angles y* ¼ 1201–1401, while the angle on the planar surfaces was only about 401–601. The possible explanation may be the one suggested by Herminghaus that a certain class of self-affine profiles of surface roughness may render any substrate with a non-zero microscopic contact angle non-wetting, i.e., give it a macroscopic contact angle close to 1801 [188]. As also was stated by Que`re` in his progress report for this field [185], the role of the hierarchical roughness structures in amplifying hydrophobicity and even more complex ones such as fractals remains to be understood. Water-repellent porous silicon By tuning the anodization conditions, the PS surface can be made to display superhydrophobic behavior as a result of the obtained macro-/nanostructured porous matrix. Interestingly, altered anodization conditions can provide a surface with different wetting properties ranging from highly wetting (nanoporous silicon surface A) to water repellent surfaces (macroporous silicon surfaces C and D). The reason for the changed wetting behavior is governed by the obtained morphology (pore sizes and distances between adjacent pores, pore shape, orientation and level of branching). The exact models describing the relation between porous morphology (pore shape, level of pore branching, etc.) and apparent contact angle on that surface are still under debates, but in general it is true that improved water repellency from sample to sample can be explained by the increasing level of porous silicon fractalization [189] and the changing shape factor of pore tips [190,191], which reduce the fraction of the solid surfaces contacting the sessile drop. In this respect, it should be noted that the examples from nature of the waterrepellent surfaces (e.g., lotus leaf surface) posses hierarchical morphology. Consequently, the process for porous silicon fabrication provides very simple

177 means of generating surfaces with different wetting properties (water contact angles varying in the range of 20–1671). Figure 22 shows scanning electron micrographs (SEM) of the pore morphologies for each of the investigated surfaces. The inserts show a water droplet of approx. 200 mm diameter deposited on the surface (without any further surface chemical modification). The corresponding contact angels are denoted in Fig. 23. As a reference, measurements on the bare hydrogen-terminated silicon, processed with the same electrolyte but without anodization, gave a surface contact angle of 841. The fresh nanoporous silicon (surface A) exhibited water contact angle of 641, which was reduced to 201 already in an hour that was explained by surface oxidation. The value of CA lower than 841 (planar surface) shows that the droplet is in Wenzel mode. The creation of vertically oriented porous layers with mm-range characteristic pore sizes and sharp peak-like pore tips lead to increase in contact angle up to 1101 on PS surface B. The Wenzel mode also seems characteristic for this type of morphology, which can be reflected by the pinned liquid contact line [135] and respectively high tilting angle necessary to roll the drop off the surface [192]. The next investigated porous silicon layer (surface C) showed already superhydrophobic behavior (CA41501) and had a hierarchical porous structure (surface C, Fig. 22). The vertically oriented macropores (with characteristic pore diameter of 2–4 mm) were combined with the porosity in mesoscopic/sub-microscopic scale (50–100 nm) reminding of mesoporous silica beads. The highest contact angle 1671 (without extra surface modification by hydrophobic coupling agents) was obtained on the PS of surface D. The characteristic macropore diameter in that case was 4–7 mm, several times larger than the pore size in surface B, and the solid phase fraction in contact with a water droplet was thus further reduced. In this context it can be commented that examples from nature, e.g., the non-wetting leg of the water strider [193] (see Fig. 24B), displays a macro-/nanotextured surface that is remarkably similar with the morphologies seen in Fig. 21 (surface D). Notably, the contact angel for the water strider leg was also found to be 1671. The water contact angle increase for the surfaces B, C and D was explained by the increased level of the surface fractalization, which facilitates Cassie–Baxter wetting mode (drop sits on the air pads without penetrating deep inside the pore layer). Surface functionalization with hydrophobic termini may further improve the water repellency of the porous layers, Fig. 23. We subjected the porous samples to a chemical treatment using a hydrophobic surface coupling agent (1H,1H,2H,2H-perfluorooctyltrichlorosilane). Nanoporous silicon layers (surface A) after modification showed increase in CA values to 1151. It was also noted that the surface became ‘‘slippery,’’ i.e., the contact lined formed by droplet with the surface became non-pinned and the droplet moved easily during the droplet evaporation, indicating a transition to a Cassie–Baxter wetting mode. The fluorocarbon surface modification

178

Fig. 22. Tuning porous silicon morphology leads to changed wetting properties. Reprinted from [173] with permission. Copyright 2007 Humana Press.

179

Fig. 23. Water contact angle measurements on PS surfaces A– D (Fig. 22) as measured

on freshly prepared samples, after 160 days storage under ambient conditions and after fluorocarbon modification. Reprinted from [173] with permission. Copyright 2007 Humana Press.

Fig. 24. Superhydrophobic porous silicon. CA 1671. The morphology of that porous

layer (D,E) remarkably resembling texture of the leg of water strider (Gerridae) (A,B,C). Images A,B,C adapted from [193] with permission from Nature Publishing Group. Scale bar is 20 mm.

180 generated ultrahydrophobic states on all surfaces comprising macroporous texture (surfaces B, C and D) with water CA 41751. The obtained structures showed good mechanical and chemical stability and exhibited no decay in contact angle value over 7 months. MS compatible protein arrays/dual-readout chips One of the current challenges in biological and bioanalytical research is the overall complexity of protein interactions and concerns identifying what is really being bound to immobilized affinity probes, e.g., in immunoassays is a core issue. Practically, it is not possible to obtain antibodies with absolute no cross-reactivity to any other proteins. This is especially a fact when working in a high-density array format with thousands of probes immobilized onto the chip and when analyzing complex samples (blood, tissue and cell lysates). Also, most proteins exist in the cell as complexes with other proteins and these will lead to an inaccurate determination of how much protein is actually binding to a specific point in the array. One of the ways to solve those questions is the development of mass spectrometry compatible protein microarrays (dual-readout chips), which both can reveal a binding event in, e.g., an optical readout mode, where after the affinity bound species is identified in a mass spectrometry analysis mode. A step forward to the implementation of such an approach, reported by our group [40], covers the development and first proof-of-principle testing of a dual readout protein microarray based on macroporous silicon. The underlying idea of the dual readout chip is schematically illustrated in Fig. 25. In the first step, a fluorescent protein microarray immunoassay is used to identify spots displaying an affinity interaction with analyte molecules in a sample. The fluorescence readout serves as a primary screen, selecting which spots to address in the subsequent mass spectrometry (MALDI-TOF) step, thereby saving MS-runtime and reducing the time required for bioinformatics database search as non-binding assay spots are not evaluated. The dual readout principle, thereby, simultaneously provides both affinity and mass identity information. The dual readout approach also puts strict requirements for the microarray substrate, which should be amenable and efficient for both fluorescence and MALDI-TOF MS detection. We have shown that the properties of porous silicon can be tailored for both types of detection. The limit of detection of 500 zeptomol for Angiotensin II (Ang II) (peptide known for its potent pharmacological activities within inflammation processes) was obtained on a macroporous Si chip in order to demonstrate its compatibility with MS readout. The detection limit was obtained by direct deposition of Ang II on the chip surface and identified by MALDI TOF MS. The proof of principle of dual readout protocol has been demonstrated by applying it to the analysis of the Renin pathway activity. Ang I and Ang II

Fig. 25. The principle of a dual readout microchip array. At first, the fluorescent protein microarray immunoassay is performed

to identify spots displaying an affinity interaction with analyte molecules in a sample. The fluorescence readout serves as a primary screen, selecting which spots to address in the subsequent mass spectrometry (MALDI-TOF MS) step. The protocol for the dual readout chips is adapted for large molecules, where a step of on-chip protein digestion is performed prior to adding MALDI matrix and the subsequent MS interrogation. Reprinted with permission from [40]. Copyright 2003 American Chemical Society.

181

182 were specifically captured by the microarrayed antibodies from human blood plasma, fluorescently analyzed and subsequently identified by mass spectrometry [40]. If small molecules are analyzed (peptides) the fluorescence readout can be immediately followed by matrix deposition on the selected test sites and afterwards identified by MALDI MS. In the case of large molecules it is necessary to introduce the step of on-chip protein digestion prior to adding MALDI matrix. The challenge is the possible fragmentation of the probe antibodies, interfering with peptides of bound analyte molecules. This study also revealed the challenges encountered while implementing trypsin digestion protocols on a microarray platform (liquid handling) and paved the way to the transition to a microfluidic-based nanovial array format. Nanovial here stands for the ability to handle volumes and performing assays in nanoliter range.

Accelerated protein digestion and analyte enrichment in nanovials In the last years a large effort has been put into the development of new microfabricated analytical devices and their integration to create micro total analysis systems (mTAS). Such systems offer increased throughput, lower sample and reagent consumption, smaller size and lower operating costs than full size instrumentation [194,195]. This miniaturization trend has earlier been reported for well established analytical techniques, such as electrophoresis [196–198], electrochromatography [199–202], assays involving enzymes, [203–207] and immunoassays [208–210]. The requirements put by miniaturization on minute sample handling and rapid bioanalytical protocols have been driving factors for an effort toward the development of a nanovial array format. We fabricated a series of vial arrays with varying diameters and depths by isotropic etching of silicon wafers (Fig. 26) [42]. The nanovial array chips were further porosified by electrochemical etching to provide increased surface to volume ratio and optimized wetting behavior. Porous silicon layers with high water contact angles provided better liquid confinement and improved liquid handling. Piezoelectric dispensing served as a convenient technique allowing fast and automotive bioreagent delivery to the nanovial array (Fig. 26 (I)) and the porous vials serve as convenient compartments in which on-chip reactions can be performed (e.g., trypsin digestion of antibody bound proteins in microarray nanovials). On-chip trypsin digestion protocols in nanovial formats were also developed. The schematic illustration of nanovial protocol is given in Fig. 27. In general, digestion activity is known to be quite low for proteins in low micromolar levels and below. In the 5–50 mM range of substrate proteases such as trypsin exerts only 50% of the maximal activity [211]. Performing

183

Fig. 26. Porous silicon nanovial chip for accelerated protein digestion and analyte enrichment prior MALDI-TOF MS. (I) Liquid handling performed by inkjet printing technique. (II) (A) -porous nanovials provide convenient way of bioreagent liquid handling, (B-E) SEM images of the porous nanovial array along with several with close-up views. Reprinted from [42] with permission. Copyright 2006 WILEY-VCH.

Fig. 27. Schematic of the high-speed protein digestion protocol: (1) microdispenserbased deposition of enzyme; (2) dispensing of protein solution (under evaporation condition turn-over rate increased); (3) addition of matrix solution and (4) laser desorption/ionization and MS-readout. Reprinted from [42] with permission. Copyright 2006 WILEY-VCH.

reaction in the volume of dispensed droplet allows substantially increase the reaction rate. This accelerated enzyme digestion is due to the fact that water is continuously evaporated during the dispensing of substrate into the vials, leading to a increased substrate concentration that drives the formation of enzyme-substrate complexes according to the Michaelis–Menten kinetics

184 equation: V ¼ V max

½S ½S þ K M

ð5Þ

where Vmax is the maximal rate independent of substrate concentration and KM is equal to the substrate concentration at which the reaction rate is half of its maximal value. This mechanism for accelerated digestion on a plate is illustrated in Fig. 28. Upon the evaporation the concentration of the analyte increases [212]. This is especially important when dealing with extremely low or expensive amounts of starting material. An essential aspect in this regard is also the reaction time. The time scale of the seconds for digestion performed in nanovial format should be compared with hours for standard protocols performed in bulk solution. Since proteomic labs today often are highly automated with a high throughput of samples, it is desirable to reduce digestion time from hours down to minutes or even seconds. The influence of macro/nanostructured layers in a nanovial with respect to the enzyme kinetics as compared for a non-porous nanovial was verified in further nanovial digestion studies. The on-set of this work was based on

Fig. 28. Schematic over the accelerated enzyme kinetics due to the substrate enrichment, as the solvent evaporates from the nanovial. Vmax is the maximal turn-over rate independent of substrate concentration and KM is equal to the substrate concentration at which the reaction rate is half of its maximal value. t ¼ 0, enzyme kinetics rules due to bulk solution stochiometry. t ¼ t1, substrate enrichment increases the enzymatic turn-over due to solvent evaporation. t ¼ t2, maximum substrate enrichment properties have been reached, and surface area to volume ratio becomes dominant. Reprinted from [42] with permission. Copyright 2006 WILEY-VCH.

185

Fig. 29. Bovine serum albumin digested for 15 s at the concentrations of 20 fmol. The vials used were non-porous (left) and porous (right). The porous nanovials offered improved analytical performance for all the model proteins investigated, where the number of identified peptides was higher for porous nanovials as compared to the corresponding non-porous nanovials. Reprinted from [42] with permission. Copyright 2006 WILEY-VCH.

earlier work, where the introduction of a thin gold-wire in a vial [213], provided a considerably increased surface area, resulting in significantly more effective protein digestion and hence the porous nanovial was expected to perform better with respect to reaction kinetics due to its vastly increased surface area. Our study revealed improvement in on-target digests in porous vials as compared to the non-porous nanovials. This was also reflected by the increased number of identified peptides for the porous nanovial digestion, Fig. 29 (digestion time was 15 s). Superhydrophobic MALDI-TOF MS target anchor chips The hydrophobic and superhydrophobic states achieved by porosification of crystalline silicon wafers described in this chapter can bring advantages in the nanobiotechnology area as was also outlined by the application of these new surfaces in the protein microarrays and matrix assisted MS protein identification. Further modification of the porous layers with hydrophobic coupling agents led to the extreme water repellency, ultrahydrophobicity, of the porous silicon surfaces based on macro-/nanoporous morphologies (water CA 4175). A new concept where the ultrahydrophobic properties were beneficial in bioanalysis was proposed by our group in the fabrication of anchor point MALDI target chips for on-plate enrichment [41,173]. One of the ways to increase the sensitivity of MALDI MS is to load larger sample volumes on the target while keeping the contact area of the spot small, whereby the surface density of the bioanalyte is increased [214]. The concept of an anchor point chip for MALDI preparation is derived from the possibility to deposit

186

Fig. 30. (A) Schematic of enrichment by evaporation on a hydrophilic anchor point surrounded by superhydrophobic porous area. (B) Anchor point of the silicon MALDI-TOF MS target chip with the diameter of 500 mm surrounded by the ultrahydrophobic macroporous area. After deposition of 5 fmol of ADH digest in a 5 mL volume droplet diameter around 1 mm, the sample/matrix mixture was confined and crystallized only at the anchor point area. Inset shows porous silicon morphology. Top inset shows the sequence of sample deposition by pipette. Reprinted from [41] with permission. Copyright 2006 Elsevier Publisher.

Fig. 31. Schematic illustration of the direct proton beam writing on porous silicon.

(A) Focused proton beam write the desired pattern on silicon wafer (B) simulation of 2.5 MeV protons incident to one spot onto Si. The beam spreads 5% at the end of range, where it produces most of the damage. The large values of proton penetration depth allow fabrication of high-aspect-ratio structures (C) proton irradiation created areas with high resistivity and (D) during the electrochemical etching irradiated areas remain unporosified while the non-irradiated areas get porosified.

187 large volumes still having a good on-target spot confinement, i.e., a small drying area. For this purpose, we created circular anchor points with a diameter of 500 mm by proton beam irradiation of silicon, Fig. 30. More details about the direct proton beam writing can be found in [41]. The schematic illustration of this process is in Fig. 31. Briefly, proton irradiated areas do not porosify in a conventional porous silicon anodization process and thus a direct write patterning of porous and non-porous areas on a silicon chip is accomplished. The chip was further porosified to create a surrounding macroporous layer. As we have shown previously, the macroporous textured silicon layer can be modified with fluorocarbon coupling to display ultrahydrophobic behavior showing water contact angles up to 1761. The ultrahydrophobicity lead to the fact that the deposited volume contacts only the anchor point area and sample/matrix mixture shrinks down to the

Fig. 32. (A) The principle of enrichment by evaporation on anchor point of a MA-

LDI target chip. Five microliters of sample mixed with matrix was deposited on a hydrophilic anchor point. The anchor point on the silicon MALDI target chip, diameter ¼ 500 mm, was surrounded by a non-wetting ultrahydrophobic macroporous area. After deposition of 2 fmol of peptide mixture (Ang1, Ang2, Sub.P, Renine, ACTH) in a 5 ml volume droplet, diameter around 1 mm, the sample/matrix mixture was confined and crystallized only at the anchor point area. Inset photos (A) show a sequence of images with a drying sample. (B) The resulting MALDI spectra. (C) Control experiment: MALDI spectra of the same amount of sample deposited on a standard target. Reprinted from [173] with permission. Copyright 2007 Humana Press.

188

Fig. 33. Array of high aspect ration pillars obtained on o1004 p-type silicon. The grid of high aspect ration post surrounded by non-wetting regions can precisely direct the liquid deposition onto the top of the posts. Scale bars 100 mm. Reprinted from [41] with permission. Copyright 2006 Elsevier Publisher.

well defined and limited spot area on the chip during drop evaporation (Fig. 30). The resulting MS spectra are given in Fig. 32, where ADH digest can be successfully identified by five peptides. The anchor chip approach can be further optimized by reducing the anchor chip area. A first effort in this direction is illustrated in Fig. 33 where the anchor post diameter is reduced to less than 10 mm. Iterations on the optimal anchor point diameters and matching these to laser spot diameters and employed laser power remains, but the current work clearly demonstrates an interesting way forward along this route. Another possible implication can be foreseen for printing protein microarrays on grid of such posts surrounded by water repellent area. This can eliminate the problem of unspecific binding to the area between the test spots completely. Conclusions Electrochemical etching of silicon provides an easy way to fabricate randomly ordered macro-/nanostructured silicon substrates. These surfaces clearly display beneficial bioanalytical properties in applications were directed and tailored morphologies and surface properties play a major role for the analytical readout. The developed macro-/nanostructured silicon substrate presents a high surface area 3D matrix for efficient immobilization of e.g., antibodies. The high surface area provides both increased immobilization efficiency and beneficial microfluidic properties (low wetting, good spot confinement, small spot sizes and homogeneous spot intensity profiles).

189 Reproducible manufacturing and assaying further establishes this new surface as a strong alternative for use in protein chip microarray applications. The robustness and low background interference and low un-specific binding even in complex samples such as blood plasma and cell lysate possibly opens the route to the development of new multiplex assays for clinical diagnostics. Future efforts will also target the optimization of the protein chip surface and assay for direct MALDI TOF-MS readout, enabling both affinity interaction readout by fluorescence and molecular structure readout by means of mass spectrometry. Acknowledgments We would like to thank SWEGENE and the Wallenberg foundation for financial support. Carl Trygger Foundation, Royal Physiographic Society in Lund, Crafoord Foundation, Foundation for Strategic Research and the Swedish Research Council are also acknowledged for their support. References 1. 2. 3.

4. 5. 6.

7. 8. 9. 10. 11. 12.

13.

Dufva M and Christensen CBV. Diagnostic and analytical applications of protein microarrays. Exp Rev Proteomics 2005;2(1):41–48. MacBeath G. Protein microarrays and proteomics. Nat Genet 2002;32(Suppl.):526–532. O’Connor CD and Pickard K. Protein chips and microarrays. In: Microarrays & Microplates: Applications in Biomedical Sciences, Day INM (ed), BIOS Scientific Publ, 2003, pp. 61–88. Templin MF, Stoll D, Schwenk JM, Potz O, Kramer S and Joos TO. Protein microarrays: Promising tools for proteomic research. Proteomics 2003;3(11):2155–2166. Wilson DS and Nock S. Recent developments in protein microarray technology. Angew Chem Int Ed 2003;42(5):494–500. Zhou FX, Bonin J and Predki PF. Development of functional protein microarrays for drug discovery: progress and challenges. Comb Chem. High Th. Screen 2004;7(6): 539–546. Zhu H and Snyder M. Protein arrays and microarrays. Curr Opin Chem Biol 2001;5(1):40–45. Zhu H and Snyder M. Protein chip technology. Curr Opin Chem Biol 2003;7(1):55–63. Kambhampati D (ed). Protein microarray technology, Wiley, 2004, 276 pp. Schena M (ed). Protein microarrays. Jones and Bartlett Pub., 2005, 469 pp. MacBeath G and Schreiber SL. Printing proteins as microarrays for high-throughput function determination. Science 2000;289(5485):1760–1763. Zhu H, Bilgin M, Bangham R, Hall D, Casamayor A, Bertone P, Lan N, Jansen R, Bidlingmaier S, Houfek T, Mitchell T, Miller P, Dean RA, Gerstein M and Snyder M. Global analysis of protein activities using proteome chips. Science 2001;293(5537): 2101–2105. Chen GYJ, Uttamchandani M, Zhu Q, Wang G and Yao SQ. Developing a strategy for activity-based detection of enzymes in a protein microarray. ChemBiochem 2003;4(4):336–339.

190 14. Espina V, Mehta AI, Winters ME, Calvert V, Wulfkuhle J, Petricoin EF and Liotta LA. Protein microarrays: molecular profiling technologies for clinical specimens. Proteomics 2003;3(11):2091–2100. 15. Gulmann C, Espina V, Petricoin III. E, Longo DL, Santi M, Knutsen T, Raffeld M, Jaffe ES, Liotta LA and Feldman AL. Proteomic analysis of apoptotic pathways reveals prognostic factors in follicular lymphoma. Clin Cancer Res 2005;11(16):5847–5855. 16. Hanash S. Diagnostic applications of protein microarrays. Medical Biomethods Handbook 2005: 583–592. 17. Harwanegg C and Hiller R. Protein microarrays for the diagnosis of allergic diseases: state-of-the-art and future development. Laboratoriumsmedizin 2005;29(4):272–277. 18. Hudelist G, Singer CF, Kubista E and Czerwenka K. Use of high-throughput arrays for profiling differentially expressed proteins in normal and malignant tissues. Anticancer Drugs 2005;16(7):683–689. 19. Koopmann JO, McAndrew MB and Blackburn JM. Development of protein microarrays for drug discovery. In: Protein Microarrays, Schena M (ed), Chapter 22, Jones & Bartlett Pub., 2005, pp. 401–420. 20. Korf U and Wiemann S. Protein microarrays as a discovery tool for studying protein–protein interactions. Exp Rev Proteom 2005;2(1):13–26. 21. Kumble KD. Protein microarrays: new tools for pharmaceutical development. Anal Bioanal Chem 2003;377(5):812–819. 22. LaBaer J and Ramachandran N. Protein microarrays as tools for functional proteomics. Curr Opin Chem Biol 2005;9(1):14–19. 23. Nishizuka S, Chen ST, Gwadry FG, Alexander J, Major SM, Scherf U, Reinhold WC, Waltham M, Charboneau L, Young L, Bussey KJ, Kim SY, Lababidi S, Lee JK, Pittaluga S, Scudiero DA, Sausville EA, Munson PJ, Petricoin EF, Liotta LA, Hewitt SM, Raffeld M and Weinstein JN. Diagnostic markers that distinguish colon and ovarian adenocarcinomas: identification by genomic, proteomic, and tissue array profiling. Cancer Res 2003;63(17):5243–5250. 24. Paweletz CP, Charboneau L, Bichsel VE, Simone NL, Chen T, Gillespie JW, Emmert-Buck MR, Roth MJ, Petricoin EF and Liotta LA. Reverse phase protein microarrays which capture disease progression show activation of pro-survival pathways at the cancer invasion front. Oncogene 2001;20(16):1981–1989. 25. Poetz O, Ostendorp R, Brocks B, Schwenk JM, Stoll D, Joos TO and Templin MF. Protein microarrays for antibody profiling: specificity and affinity determination on a chip. Proteomics 2005;5(9):2402–2411. 26. Schneiderhan-Marra N, Kirn A, Doettinger A, Templin M, Sauer G, Deissler H and Joos TO. Protein microarrays – a promising tool for cancer diagnosis. Cancer Genom Proteom 2005;2(1):37–42. 27. Sreekumar A, Nyati MK, Varambally S, Barrette TR, Ghosh D, Lawrence TS and Chinnaiyan AM. Profiling of cancer cells using protein microarrays: discovery of novel radiation-regulated proteins. Cancer Res 2001;61(20):7585–7593. 28. Steller S, Angenendt P, Cahill DJ, Heuberger S, Lehrach H and Kreutzberger J. Bacterial protein microarrays for identification of new potential diagnostic markers for Neisseria meningitidis infections. Proteomics 2005;5(8):2048–2055. 29. Stoll D, Templin MF, Bachmann J and Joos TO. Protein microarrays: applications and future challenges. Curr Opin Drug Discov Dev 2005;8(2):239–252. 30. Cowherd SM, Espina VA, Petricoin III. EF and Liotta LA. Proteomic analysis of human breast cancer tissue with laser-capture microdissection and reverse-phase protein microarrays. Clin Breast Cancer 2004;5(5):385–392.

191 31. Davies DH, Liang X, Hernandez JE, Randall A, Hirst S, Mu Y, Romero KM, Nguyen TT, Kalantari-Dehaghi M, Crotty S, Baldi P, Villarreal LP and Felgner PL. Profiling the humoral immune response to infection by using proteome microarrays: highthroughput vaccine and diagnostic antigen discovery. Proc Natl Acad Sci USA 2005;102(3):547–552. 32. James P. Chips for proteomics: a new tool or just hype? BioTechniques Suppl 2002;4–10:12–13. 33. Kricka LJ and Master SR. Validation and quality control of protein microarray-based analytical methods. Methods Mol Med 2005;114(Microarrays in Clinical Diagnostics):233–255. 34. Kusnezow W and Hoheisel JD. Antibody microarrays: promises and problems. BioTechniques 2002;33:S14–S23. 35. Kusnezow W and Hoheisel JD. Solid supports for microarray immunoassays. J Mol Recog 2003;16(4):165–176. 36. Ressine A, Ekstrom S, Marko-Varga G and Laurell T. Macro-/nanoporous silicon as a support for high-performance protein microarrays. Anal Chem 2003;75(24):6968–6974. 37. Ressine A, Finnskog D, Malm J, Becker C, Lilja H, Varga GM and Laurell T. Macro/ nano-structured silicon as solid support for antibody arrays. Surface design, reproducibility, and assay characteristics enabling discovery of kallikrein gene products for prostate disease diagnostics. NanoBiotechnology 2005;1(1):93–104. 38. Steinhauer C, Ressine A, Marko-Varga G, Laurell T, Borrebaeck CAK and Wingren C. Biocompatibility of surfaces for antibody microarrays: design of macroporous silicon substrates. Anal Biochem 2005;341(2):204–213. 39. Ressine A, Corin I, Jaras K, Guant G, Simone C, Marko-Varga G and Laurell T. Porous silicon surfaces – a candidate substrate for reverse protein arrays in cancer biomarker detection. Electrophoresis 2007; in press. 40. Finnskog D, Ressine A, Laurell T and Marko-Varga G. Protein microchip bioassay with dual fluorescent – and maldi read-out. J Proteome Res 2004;3(5):988–994. 41. Ressine A, Auzelyte V, Kristiansson P, Marko-Varga G and Laurell T. Fabrication of sample target plate for MALDI MS using proton beam writing. Nucl Inst Meth B 2006;249:715–718. 42. Finnskog D, Jaras K, Ressine A, Malm J, Marko-Varga G, Lilja H and Laurell T. High-speed biomarker identification utilizing porous silicon nanovial arrays and MALDI-TOF mass spectrometry. Electrophoresis 2006;27(5-6):1093–1103. 43. Ressine A, Finnskog D, Marko-Varga G and Laurell T. Superhydrophobic and hydrophilic states on porous silicon. Proceedings of microTAS 2005 Conference, 2005b, vol. 1, pp. 256–258. 44. Morozov VN. Protein microarrays: principles and limitations. In: Protein Microarrays, Schena M (ed), Chapter 5, Jones & Bartlett Pub., 2005, pp. 71–105. 45. Scha¨ferling M and Kambhampati D. Protein microarray surface chemistry and coupling schemes. In: Protein Microarray Technology, Kambhampati D (ed), Chapter 2, John Wiley & Sons, 2004, pp. 11–38. 46. Turner DR. Electropolishing silicon in in HF acid solutions. J Electrochem Soc 1958(105):402–408. 47. Ulhir A. Electrolytic shaping of germanium and silicon. Bell SystTech J 1956(35):333. 48. Bell TE, Gennissen PTJ, Demunter D and Kuhl M. Porous silicon as a sacrificial material. J Micromech Microeng 1996;6(4):361–369. 49. Splinter A, Bartels O and Benecke W. Thick porous silicon formation using implanted mask technology. Sens Actuators B 2001;B76(1–3):354–360.

192 50. Steiner P, Richter A and Lang W. Using porous silicon as a sacrificial layer. J Micromech Microeng 1993;3(1):32–36. 51. Colinge JP. Silicon-on-insulator and porous silicon. In: Silicon: Evolution and Future of a Technology, Siffert P and Krimmel E (eds), Springer, 2005, pp. 139–167. 52. Takai H and Itoh T. Porous silicon layers and its oxide for the silicon-on-insulator structure. J Appl Phys 1986;60(1):222–225. 53. Bisi O, Ossicini S and Pavesi L. Porous silicon: a quantum sponge structure for silicon based optoelectronics. Surf Sci Rep 2000;38(1–3):1–126. 54. Cullis AG, Canham LT and Calcott PDJ. The structural and luminescence properties of porous silicon. J Appl Phys 1997;82(3):909–965. 55. Herino R. Porous silicon for microelectronics and optoelectronics. Mater Sci Technol 1997;13(11):965–970. 56. Menna P, Di Francia G and La Ferrara V. Porous silicon in solar cells: a review and a description of its application as an AR coating. Sol Energy Mater Sol Cells 1995;37(1):13–24. 57. Saadoun M, Ezzaouia H, Bessais B, Boujmil MF and Bennaceur R. Formation of porous silicon for large-area silicon solar cells: a new method. Sol Energy Mater Sol Cells 1999;59(4):377–385. 58. Solanki CS, Bilyalov RR, Poortmans J, Beaucarne G, Van Nieuwenhuysen K, Nijs J and Mertens R. Characterization of free-standing thin crystalline films on porous silicon for solar cells. Thin Solid Films 2004;451–452:649–654. 59. Berger MG, Dieker C, Thoenissen M, Vescan L, Lueth H, Muender H, Theiss W, Wernke M and Grosse P. Porosity superlattices: a new class of Si heterostructures. J Phys D 1994;27(6):1333–1336. 60. Vincent G. Optical properties of porous silicon superlattices. Appl Phys Lett 1994;64(18):2367–2369. 61. Aravamudhan S, Rahman ARA and Bhansali S. Porous silicon based orientation independent, self-priming micro direct ethanol fuel cell. Sens Actuat A 2005;A123–A124: 497–504. 62. Presting H, Konle J, Starkov V, Vyatkin A and Konig U. Porous silicon for micro-sized fuel cell reformer units. Mater Sci Eng B 2004;B108(1–2):162–165. 63. Aston R, and Canham LT. A porous and/or polycrystalline silicon orthopedic implant, PCT Int Appl 2001; 48 pp. Wo: (Secretary of State for Defence, UK). 64. Pramatarova L, Pecheva E, Dimova-Malinovska D, Pramatarova R, Bismayer U, Petrov T and Minkovski N. Porous silicon as a substrate for hydroxyapatite growth. Vacuum 2004;76(2–3):135–138. 65. Rajaraman S and Henderson HT. A unique fabrication approach for microneedles using coherent porous silicon technology. Sens Actuators B 2005;B105(2):443–448. 66. Salonen J, Paski J, Vaehae-Heikkilae K, Heikkilae T, Bjoerkqvist M and Lehto VP. Determination of drug load in porous silicon microparticles by calorimetry. Phys Stat Sol A 2005;202(8):1629–1633. 67. Sailor MJ. Sensor applications of porous silicon. EMIS Datareviews Ser 1997;18 (Properties of Porous Silicon):364–370. 68. Mathew FP and Alocilja EC. Porous silicon-based biosensor for pathogen detection. Biosens Bioelectron 2005;20(8):1656–1661. 69. Tinsley-Bown AM, Canham LT, Hollings M, Anderson MH, Reeves CL, Cox TI, Nicklin S, Squirrell DJ, Perkins E, Hutchinson A, Sailor MJ and Wun A. Tuning the pore size and surface chemistry of porous silicon for immunoassays. Phys Stat Sol A 2000;182(1):547–553.

193 70. Anon. Characterizing a porous silicon optical biosensor. Anal Chem 1999;71(21): 727A. 71. Lehmann V. Biosensors: barcoded molecules. Nat Mater 2002;1(1):12–13. 72. Martin-Palma RJ, Torres-Costa V, Arroyo-Hernandez M, Manso M, Perez-Rigueiro J and Martinez-Duart JM. Porous silicon multilayer stacks for optical biosensing applications. Microelectron J 2004;35(1):45–48. 73. Pacholski C, Sartor M, Sailor MJ, Cunin F and Miskelly GM. Biosensing using porous silicon double-layer interferometers: reflective interferometric fourier transform spectroscopy. J Am Chem Soc 2005;127(33):11636–11645. 74. Schoning MJ, Kurowski A, Thust M, Kordos P, Schultze JW and Luth H. Capacitive microsensors for biochemical sensing based on porous silicon technology. Sens Actuators B 2000;B64(1–3):59–64. 75. Starodub NF, Rebriev AV and Starodub VM. Biosensors and express biochemical diagnostics of some diseases. NATO Sci Ser I: Life Behavioural Sci 2002;346(Disease Markers in Exhaled Breath):391–394. 76. Starodub VM. Porous silicon as transducer for immune sensors: from theory to practice. NATO Sci Ser II: Math, Phy Chem 2002;57(Frontiers of Multifunctional Nanosystems):383–398. 77. Tinsley-Bown A, Smith RG, Hayward S, Anderson MH, Koker L, Green A, Torrens R, Wilkinson AS, Perkins EA, Squirrell DJ, Nicklin S, Hutchinson A, Simons AJ and Cox TI. Immunoassays in a porous silicon interferometric biosensor combined with sensitive signal processing. Phys Stat Sol A 2005;202(8): 1347–1356. 78. Zairi S, Martelet C, Jaffrezic-Renault N, M’Gaieth R, Maaref H and Lamartine R. Porous silicon. A transducer material for a high-sensitive (bio)chemical sensor: effect of a porosity, pores morphologies and a large surface area on a sensitivity. Thin Solid Films 2001;383(1,2):325–327. 79. Foll H, Christophersen M, Carstensen J and Hasse G. Formation and application of porous silicon. Mater Sci Eng R: Rep 2002;R39(4):93–141. 80. Drott J, Rosengren L, Lindstrom K and Laurell T. Pore morphology influence on catalytic turn-over for enzyme activated porous silicon matrices. Thin Solid Films 1998;330(2):161–166. 81. Bengtsson M, Drott J and Laurell T. Tailoring of porous silicon morphology in chip integrated bioreactors. Phys Stat Sol A 2000;182(1):533–539. 82. Collins A, Mikeladze E, Bengtsson M, Kokaia M, Laurell T and Csoregi E. Interference elimination in glutamate monitoring with chip integrated enzyme microreactors. Electroanalysis 2001;13(6):425–431. 83. Davidsson R, Genin F, Bengtsson M, Laurell T and Emneus J. Microfluidic biosensing systems. Part I. Development and optimisation of enzymatic chemiluminescent m-biosensors based on silicon microchips. Lab on a Chip 2004;4(5):481–487. 84. Drott J, Lindstroem K, Rosengren L and Laurell T. Porous silicon as the carrier matrix in microstructured enzyme reactors yielding high enzyme activities. J Micromech Microeng 1997;7(1):14–23. 85. Ekstrom S, Ericsson D, Onnerfjord P, Bengtsson M, Nilsson J, Marko-Varga G and Laurell T. Signal amplification using ‘‘spot-on-a-chip’’ technology for the identification of proteins via MALDI-TOF MS. Anal Chem 2001;73(2):214–219. 86. Lendl B, Schindler R, Frank J, Kellner R, Drott J and Laurell T. Fourier transform infrared detection in miniaturized total analysis systems for sucrose analysis. Anal Chem 1997;69(15):2877–2881.

194 87. Melander C, Momcilovic D, Nilsson C, Bengtsson M, Schagerloef H, Tjerneld F, Laurell T, Reimann CT and Gorton L. Microchip immobilized enzyme reactors for hydrolysis of methyl cellulose. Anal Chem 2005;77(10):3284–3291. 88. Yakovleva J, Davidsson R, Lobanova A, Bengtsson M, Eremin S, Laurell T and Emneus J. Microfluidic enzyme immunoassay using silicon microchip with immobilized antibodies and chemiluminescence detection. Anal Chem 2002;74(13):2994–3004. 89. Schoning MJ, Ronkel F, Crott M, Thust M, Schultze JW, Kordos P and Luth H. Miniaturization of potentiometric sensors using porous silicon microtechnology. Electrochim Acta 1997;42(20–22):3185–3193. 90. Thust M, Schoening MJ, Frohnhoff S, Arens-Fischer R, Kordos P and Lueth H. Porous silicon as a substrate material for potentiometric biosensors. Measure Sci Technol 1996;7(1):26–29. 91. Bayliss SC, Ashraf I and Sapelkin AV. Concerning signaling in in vitro neural arrays using porous silicon. NATO Sci Ser II: Mathe, Phys Chem 2004;152:467–471. 92. Buckberry L and Bayliss S. Semiconducting biomaterials: current status and future perspectives. Med Device Technol 2001;12(5):14–20. 93. de-Leon SB-T, Oren R, Spira ME, Korbakov N, Yitzchaik S and Sa’ar A. Porous silicon substrates for neurons culturing and bio-photonic sensing. Phys Stat Sol A 2005;202(8):1456–1461. 94. Mayne AH, Bayliss SC, Barr P, Tobin M and Buckberry LD. Biologically interfaced porous silicon devices. Phys Stat Sol A 2000;182(1):505–513. 95. Lehmann V and Go¨sele U. Porous silicon formation: a quantum wire effect. Appl Phys Lett 1991;58(8):856–858. 96. Zhang XG. Morphology and formation mechanisms of porous silicon. J Electrochem Soc 2004;151(1):C69–C80. 97. Carstensen J, Christophersen M, Loelkes S, Ossei-Wusu E, Bahr J, Langa S, Popkirov G and Foell H. Large area etching for porous semiconductors. Phys Stat Sol C 2005;2(9):3339–3343. 98. Ouyang H, Christophersen M and Fauchet PM. Enhanced control of porous silicon morphology from macropore to mesopore formation. Phys Stat Sol A 2005;202(8):1396–1401. 99. Vadjikar RM and Nandedkar RV. Pore size dependence on doping concentration in porous silicon. Curr Sci 1993;64(3):180–182. 100. Christophersen M, Carstensen J, Voigt K and Foll H. Organic and aqueous electrolytes used for etching macro- and mesoporous silicon. Phys Stat Sol A 2003;197(1):34–38. 101. Christophersen M, Langa S, Carstensen J, Tiginyanu IM and Foll H. A comparison of pores in silicon and pores in III–V compound materials. Phys Stat Sol A 2003;197(1):197–203. 102. Butler JE. Solid supports in enzyme-linked immunosorbent assay and other solid-phase immunoassays. Methods 2000;22(1):4–23. 103. Vermeer AWP, Giacomelli CE and Norde W. Adsorption of IgG onto hydrophobic teflon. Differences between the Fab and Fc domains. Biochim Biophys Acta, General Subjects 2001;1526(1):61–69. 104. Butler JE. Solid supports in enzyme-linked immunosorbent assay and other solid-phase immunoassays. Methods Mol Med 2004;94:333–372. 105. Piehler J, Brecht A, Geckeler KE and Gauglitz G. Surface modification for direct immunoprobes. Biosen Bioelectron 1996;11(6/7):579–590. 106. Piehler J, Brecht A, Valiokas R, Liedberg B and Gauglitz G. A high-density poly (ethylene glycol) polymer brush for immobilization on glass-type surfaces. Biosens Bioelectron 2000;15(9–10):473–481.

195 107. Lilja H, Christensson A, Dahlen U, Matikainen MT, Nilsson O, K. P and Lovgren T. Prostate-specific antigen in serum occurs predominantly in complex with alpha1-antichymotrypsin. Clin Chem 1991;37(9):1618–1625. 108. Oesterling JE, Jacobsen SJ, Klee GG, Pettersson K, Piironen T, Abrahamsson PA, Stenman UH, Dowell B, Lovgren T and Lilja H. Free, complexed and total serum prostate-specific antigen – the establishment of appropriate reference ranges for their concentrations and ratios. J Urol 1995;154(3):1090–1095. 109. Calvert VS, Tang Y, Boveia V, Wulfkuhle J, Schutz-Geschwender A, Olive DM, Liotta LA and Petricoin III. EF. Development of multiplexed protein profiling and detection using near infrared detection of reverse-phase protein microarrays. Clin Proteom 2004;1(1):81–89. 110. Espina V, Woodhouse EC, Wulfkuhle J, Asmussen HD, Petricoin EF and Liotta LA. Protein microarray detection strategies: focus on direct detection technologies. J Immunol Methods 2004;290(1–2):121–133. 111. Grubb RL, Calvert VS, Wulkuhle JD, Paweletz CP, Linehan WM, Phillips JL, Chuaqui R, Valasco A, Gillespie J, Emmert-buck M, Liotta LA and Petricoin EF. Signal pathway profiling of prostate cancer using reverse phase protein arrays. Proteomics 2003;3(11):2142–2146. 112. Borg A, Ferno M and Peterson C. Predicting the future of breast cancer. Nat Med 2003;9(1):16–18. 113. Keyomarsi K, Tucker Susan L, Buchholz Thomas A, Callister M, Ding Y, Hortobagyi Gabriel N, Bedrosian I, Knickerbocker C, Toyofuku W, Lowe M, Herliczek Thaddeus W and Bacus Sarah S. Cyclin E and survival in patients with breast cancer. New Engl J Med 2002;347(20):1566–1575. 114. Lindahl T, Landberg G, Ahlgren J, Nordgren H, Norberg T, Klaar S, Holmberg L and Bergh J. Overexpression of cyclin E protein is associated with specific mutation types in the p53 gene and poor survival in human breast cancer. Carcinogenesis 2004;25(3): 375–380. 115. Cheek BJ, Steel AB, Torres MP, Yu Y-Y and Yang H. Chemiluminescence detection for hybridization assays on the flow-thru chip, a three-dimensional microchannel biochip. Anal Chem 2001;73(24):5777–5783. 116. Tseng FG, Lin SC, Yao DJ, Huang H and Chieng CC. Technological aspects of protein microarrays and nanoarrays. In: Protein Microarrays, Schena M (ed), Chapter 16, Jones & Bartlett Pub., 2005, pp. 305–337. 117. Dugas V, Broutin J and Souteyrand E. Droplet evaporation study applied to DNA chip manufacturing. Langmuir 2005;21(20):9130–9136. 118. Blossey R and Lorke A. Wetting droplet instability and quantum ring formation. Phys Rev E 2002;65(2–1):021603/021601–021603/021603. 119. Deegan RD, Bakajin O, Dupont TF, Huber G, Nagel SR and Witten TA. Capillary flow as the cause of ring stains from dried liquid drops. Nature 1997;389(6653): 827–829. 120. Heim T, Preuss S, Gerstmayer B, Bosio A and Blossey R. Deposition from a drop: morphologies of unspecifically bound DNA. J Phys: Condens Matter 2005;17(9): S703–S716. 121. Hu H and Larson R. Analysis of the microfluid flow in an evaporating sessile droplet. Langmuir 2005;21:3963–3971. 122. McQuain MK, Seale K, Peek J, Levy S and Haselton FR. Effects of relative humidity and buffer additives on the contact printing of microarrays by quill pins. Anal Biochem 2003;320(2):281–291.

196 123. Blossey R and Bosio A. Contact line deposits on cDNA microarrays: A ‘‘Twin-Spot Effect’’. Langmuir 2002;18(7):2952–2954. 124. Deegan RD, Bakajin O, Dupont TF, Huber G, Nagel SR and Witten TA. Contact line deposits in an evaporating drop. Phys Rev E 2000;62:756–765. 125. Tekin E, de Gans B-J and Schubert US. Ink-jet printing of polymers – from single dots to thin film libraries. J Mater Chem 2004;14(17):2627–2632. 126. Ko SH, Chung J, Choi Y, Grigoropoulos CP, Bieri NR, Choi T-y, Dockendorf C and Poulikakos D. Laser based hybrid inkjet printing of nanoink for flexible electronics. Proceedings of SPIE-The International Society for Optical Engineering 5713 (Photon Processing in Microelectronics and Photonics IV) 2005, pp. 97–104. 127. Rickman DS, Herbert CJ and Aggerbeck LP. Optimizing spotting solutions for increased reproducibility of cDNA microarrays. Nucl Acids Res 2003;31(18):e109. 128. Sefiane K. Effect of nonionic surfactant on wetting behavior of an evaporating drop under a reduced pressure environment. J Colloid Interface Sci 2004;272(2): 411–419. 129. Neogi P. Tears-of-wine and related phenomena. J Colloid Int Sci 1985;105(1):94–101. 130. Vuilleumier R, Ego V, Neltner L and Cazabat AM. Tears of wine: the stationary state. Langmuir 1995;11(10):4117–4121. 131. Hu H and Larson R. Analysis of the effect of Marangoni stresses on the microflow in an evaporating sessile droplet. Langmuir 2005;21:3972–3980. 132. Savino R, Paterna D and Favaloro N. Buoyancy and Marangoni effects in an evaporating drop. J Thermophys Heat Trans 2002;16(4):562–574. 133. Artus GRJ, Jung S, Zimmermann J, Gautschi H-P, Marquardt K and Seeger S. Silicone nanofilaments and their application as superhydrophobic coatings. Adv Mater 2006;18(20):2758–2762. 134. Baldacchini T, Carey JE, Zhou M and Mazur E. Superhydrophobic surfaces prepared by microstructuring of silicon using a femtosecond laser. Langmuir 2006;22(11): 4917–4919. 135. Callies M and Quere D. On water repellency. Soft Matter 2005;1(1):55–61. 136. Coffinier Y, Janel S, Addad A, Blossey R, Gengembre L, Payen E and Boukherroub R. Preparation of superhydrophobic silicon oxide nanowire surfaces. Langmuir 2007; 23(4):1608–1611. 137. Feng X and Jiang L. Design and creation of superwetting/antiwetting surfaces. Adv Mater 2006;18(23):3063–3078. 138. Gao L and McCarthy TJ. Artificial lotus leaf Prepared using a 1945 patent and a commercial textile. Langmuir 2006;22(14):5998–6000. 139. Han J T, Kim S and Karim A. UVO-tunable superhydrophobic to superhydrophilic wetting transition on biomimetic nanostructured surfaces. Langmuir 2007, ACS ASAP. 140. Li Y, Huang XJ, Heo SH, Li CC, Choi YK, Cai WP and Cho SO. Superhydrophobic bionic surfaces with hierarchical microsphere/SWCNT composite arrays. Langmuir 2007;23(4):2169–2174. 141. Ma M and Hill RM. Superhydrophobic surfaces. Curr Opin Colloid Interface Sci 2006;11(4):193–202. 142. Xiu Y, Zhu L, Hess DW and Wong CP. Biomimetic creation of hierarchical surface structures by combining colloidal self-assembly and au sputter deposition. Langmuir 2006;22(23):9676–9681. 143. Zhai L, Berg MC, Cebeci FC, Kim Y, Milwid JM, Rubner MF and Cohen RE. Patterned superhydrophobic surfaces: toward a synthetic mimic of the Namib desert beetle. NanoLetters 2006;6(6):1213–1217.

197 144. Zhang J, Huang W and Han Y. A composite polymer film with both superhydrophobicity and superoleophilicity. Macromol Rapid Commun 2006;27(10):804–808. 145. Egatz-Gomez A, Melle S, Garcia AA, Lindsay SA, Marquez M, Dominguez-Garcia P, Rubio MA, Picraux ST, Taraci JL, Clement T, Yang D, Hayes MA and Gust D. Discrete magnetic microfluidics. Appl Phys Lett 2006;89(3):034106/034101–034106/ 034103. 146. Herbertson DL, Evans CR, Shirtcliffe NJ, McHale G and Newton MI. Electrowetting on superhydrophobic SU-8 patterned surfaces. Sens Actuators A 2006;A130–A131:189–193. 147. Hong X, Gao X and Jiang L. Application of superhydrophobic surface with high adhesive force in no lost transport of superparamagnetic microdroplet. J Am Chem Soc 2007;129(6):1478–1479. 148. Lu C, Juang YJ, Koh CG and Lee LJ. Superhydrophobic valve for microfluidics. In: Proceedings of the 64th Annual Technical Conference & Exhibition, Charlotte, NC (ed), May 7–11, Society of Plastics Engineers, 2006, pp. 2566–2570. 149. Lu C, Xie Y, Yang Y, Cheng MMC, Koh C-G, Bai Y, Lee LJ and Juang Y-J. New valve and bonding designs for microfluidic biochips containing proteins. Anal Chem 2007;79(3):994–1001. 150. Zhang J and Kwok DY. Contact line and contact angle dynamics in superhydrophobic channels. Langmuir 2006;22(11):4998–5004. 151. Acatay K, Simsek E, Ow-Yang C and Menceloglu YZ. Superhydrophobic films: tunable, superhydrophobically stable polymeric surfaces by electrospinning. Angew Chem Int Ed 2004;43(39):5210–5213. 152. Erbil HY, Demirel AL, Avci Y and Mert O. Transformation of a simple plastic into a superhydrophobic surface. Science 2003;299(5611):1377–1380. 153. Feng L, Song Y, Zhai J, Liu B, Xu J, Jiang L and Zhu D. Creation of a superhydrophobic surface from an amphiphilic polymer. Angew Chem Int Ed 2003;42(7): 800–802. 154. Han JT, Lee DH, Ryu CY and Cho K. Fabrication of superhydrophobic surface from a supramolecular organosilane with quadruple hydrogen bonding. JACS 2004;126(15): 4796–4797. 155. Jiang L, Zhao Y and Zhai J. Superhydrophobic surface: a lotus-leaf-like superhydrophobic surface: a porous microsphere/nanofiber composite film prepared by electrohydrodynamics. Angew Chem Int Ed 2004;43(33):4338–4341. 156. Jiang Y, Wang Z, Yu X, Shi F, Xu H, Zhang X, Smet M and Dehaen W. Self-assembled monolayers of dendron thiols for electrodeposition of gold nanostructures: toward fabrication of superhydrophobic/superhydrophilic surfaces and pH-responsive surfaces. Langmuir 2005;21(5):1986–1990. 157. Lau KKS, Bico J, Teo KBK, Chhowalla M, Amaratunga GAJ, Milne WI, McKinley GH and Gleason KK. Superhydrophobic carbon nanotube forests. NanoLetters 2003;3(12):1701–1705. 158. Shang HM, Wang Y, Limmer SJ, Chou TP, Takahashi K and Cao GZ. Optically transparent superhydrophobic silica-based films. Thin Solid Films 2005;472(1-2):37–43. 159. Shirtcliffe NJ, McHale G, Newton MI and Perry CC. Intrinsically superhydrophobic organosilica sol–gel foams. Langmuir 2003;19(14):5626–5631. 160. Sun T, Feng L, Gao X and Jiang L. Bioinspired surfaces with special wettability. Acc. Chem. Res. 2005, ACS ASAP. 161. Tadanaga K, Kitamuro K, Matsuda A and Minami T. Formation of superhydrophobic alumina coating films with high transparency on polymer substrates by the sol–gel method. J Sol–Gel Sci Technol 2003;26(1/2/3):705–708.

198 162. Chen W and al. e. Ultrahydrophobic and ultralyophobic surfaces: some comments and examples. Langmuir 1999;15:3395–3399. 163. Wu Y and al. e. Thin films with nanotextures for transparent and ultra water-repellent coatings produced from trimethylmethoxysilane by microwave plasma CVD. Chem Vap Deposition 2002;8:47–50. 164. Youngblood JP and T.J. M. Polymer surfaces prepared by simultaneous ablation of polypropylene and sputtering of poly(tetrafluoroethylene) using radio frequency plasma. Macromolecules 1999;32:6800–6806. 165. Lee CS, Lee SH, Park SS, Kim YK and Kim BG. Protein patterning on silicon-based surface using background hydrophobic thin film. Biosens Bioelectron 2003;18:437–444 166. Fuerstner R, Barthlott W, Neinhuis C and Walzel P. Wetting and self-cleaning properties of artificial superhydrophobic surfaces. Langmuir 2005;21(3):956–961. 167. Li Y, Cai W, Duan G, Cao B, Sun F and Lu F. Superhydrophobicity of 2D ZnO ordered pore arrays formed by solution-dipping template method. J Colloid Interface Sci 2005;287(2):634–639. 168. O¨ner D and McCarthy TJ. Ultrahydrophobic surfaces. Effects of topography length scales on wettability. Langmuir 2000;16:7777–7782. 169. Shiu J-Y, Kuo C-W, Chen P and Mou C-Y. Fabrication of tunable superhydrophobic surfaces by nanosphere lithography. Chem Mater 2004;16(4):561–564. 170. Ren S, Yang S, Zhao Y, Yu T and Xiao X. Preparation and characterization of an ultrahydrophobic surface based on a stearic acid self-assembled monolayer over polyethyleneimine thin films. Surf Sci 2003;546(2–3):64–74. 171. Li M, Zhai J, Liu H, Song Y, Jiang L and Zhu D. Electrochemical deposition of conductive superhydrophobic zinc oxide thin films. J Phys Chem B 2003;107(37): 9954–9957. 172. Wu Y, Sugimura H, Inoue Y and Takai O. Thin films with nanotextures for transparent and ultra water-repellent coatings produced from trimethylmethoxysilane by microwave plasma CVD. Chem Vap Deposition 2002;8(2):47–50. 173. Ressine A, Finnskog D, Marko-Varga G and Laurell T. Superhydrophobic properties of porous silicon improves surface based bioanalysis. NanoBiotechnology 2007; in press. 174. Cao M, Song X, Zhai J, Wang J and Wang Y. Fabrication of highly antireflective silicon surfaces with superhydrophobicity. J Phys Chem B 2006;110(26):13072–13075. 175. Wang M-F, Raghunathan N and Ziaie B. A nonlithographic top-down electrochemical approach for creating hierarchical (micro-nano) superhydrophobic silicon surfaces. Langmuir, 2007, ACS ASAP. 176. Boys CV. Soap Bubbles, London, Society for Promoting Christian Knowledge, 1902. 177. Barthlott W and Neinhuis C. Purity of the sacred lotus, or escape from contamination in biological surfaces. Planta 1997;202(1):1–8. 178. Onda T, Shibuichi S, Satoh N and Tsujii K. Super-water-repellent fractal surfaces. Langmuir 1996;12(9):2125–2127. 179. Young T. An essay on the cohesion of fluids. Philos Trans R Soc London 1805;95: 65–87. 180. Wenzel RN. Resistance of solid surfaces to wetting by water. Ind Eng Chem 1936(28):988–994. 181. Cassie ABD and Baxter S. Wettability of porous surfaces. Trans Faraday Soc 1944;40:546–551. 182. Lafuma A and Quere D. Superhydrophobic states. Nat Mater 2003;2(7):457–460. 183. Quere D, Lafuma A and Bico J. Slippy and sticky microtextured solids. Nanotechnology 2003;14(10):1109–1112.

199 184. Carbone G and Mangialardi L. Hydrophobic properties of a wavy rough substrate. Eur Phys J E: Soft Matter 2005;16(1):67–76. 185. Que´re´ D. Non-sticking drops. Rep Progr Phys 2005;68:2495–2532. 186. Teare DOH, Spanos CG, Ridley P, Kinmond EJ, Roucoules V, Badyal JPS, Brewer SA, Coulson S and Willis C. Pulsed plasma deposition of super-hydrophobic nanospheres. Chem Mater 2002;14(11):4566–4571. 187. Woodward I, Schofield WCE, Roucoules V and Badya JPS. Super-hydrophobic surfaces produced by plasma fluorination of polybutadiene films. Langmuir 2003;19(8):3432–3438. 188. Herminghaus S. Roughness-induced non-wetting. Europhys Lett 2000;52(2):165. 189. Schibuichi S, Yamamoto T, Onda T and Tsujii K. Super water repellent surfaces resulting from fractal surfaces. J Phys Chem 1996;100:19512–19517. 190. Bico J, Thiele U and Quere D. Wetting of textured surfaces. Coll Surf A 2002; 206(1–3):41–46. 191. Nosonovsky M and Bhushan B. Roughness optimization for biomimetic superhydrophobic surfaces. Microsystem Technol 2005;11(7):535–549. 192. Miwa M, Nakajima A, Fujishima A, Hashimoto K and Watanabe T. Effects of the surface roughness on sliding angles of water droplets on superhydrophobic surfaces. Langmuir 2000;16(13):5754–5760. 193. Gao X and Jiang L. Biophysics: water-repellent legs of water striders. Nature 2004;432(7013):36. 194. Bruin GJM. Recent developments in electrokinetically driven analysis on microfabricated devices. Electrophoresis 2000;21:3931–3951. 195. van der Berg A and Lammerink TSJ. Micro total analysis systems: microfluidic aspects, integration concept and applications. Top Curr Chem 1998;194:21–49. 196. Gao J, Xu JD, Locascio LE and Lee CS. Integrated microfluidic system enabling protein digestion, peptide separation, and protein identification. Anal Chem 2001;73:2648–2655. 197. Liu YJ, Foote RS, Jacobsson SC, Ramsey RS and Ramsey JM. Electrophoretic separation of proteins on a microchip with noncovalentpostcolumn labeling. Anal Chem 2000;72:4608–4613. 198. Wang J, Chatrathi MP, Mulchandani A and Chen W. Capillary electrophoresis microchips for separation and detection of organophosphate nerve agents. Anal Chem 2001;73:1804–1808. 199. Ceriotti L, de Rooij NF and Verpoorte E. An integrated fritless column for on-chip capillary electrochromatography with conventional stationary phases. Anal Chem 2002;74(74):639–647. 200. Ericson C, Holm J, Ericson T and Hjerteń S. Electroosmosis and pressure-driven chromatography in chip using continuous beds. Anal Chem 2000;72:81–87. 201. Ngola SM, Fintschenko Y, Choi WY and Sheppodd TJ. Conduct-as-cast polymer monoliths as separation media for capillary electrochromatography. Anal Chem 2001;73:849–856. 202. Throckmorton DJ, Schepodd TJ and Singh AK. Electrochromatography in microchips: reversed-phase separation of peptides and amino acids using photopatterned rigid polymer monoliths. Anal Chem 2002;74:784–789. 203. Drott J, Lindstrom K, Rosengren L and Laurell T. Porous silicon as the carrier matrix in microstructured enzyme reactors yielding high enzyme activities. J Micromech Microeng 1997(7):14–23. 204. Laurell T, Drott J, Rosengren L and Lindstrom K. Enhanced enzyme activity in silicon integrated enzyme reactors utilizing porous silicon as the coupling matrix. Sens Actuators 1996(31):161–166.

200 205. Mao H, Yang T and Cremer PT. Design and characterization of immobilized enzymes in microfluidic systems. Anal Chem 2002(74):379–385. 206. Wang J. On-chip enzymatic assays. Electrophoresis 2002;23:713–718. 207. Wang J, Chatrathi MP and Tian BM. Micromachined separation chips with a precolumn reactor and end-column electrochemical detector. Anal Chem 2000;72: 5774–5778. 208. Dodge A, Fluri K, Verpoorte E and de Rooij NF. Electrokinetically driven microfluidic chips with surface-modified chambers for heterogeneous immunoassays. Anal Chem 2001;73:3400–3409. 209. Ressine A, Ekstrom S, Marko-Varga G and Laurell T. Macro-/nanoporous silicon as a support for high-performance protein microarrays. Anal Chem 2003;24:6968–6974. 210. Sato K, Tokeshi M, Kimura H and Kitamori T. Determination of carcinoembryonic antigen in human sera by integrated bead-bed immunoassay in a microchip for cancer diagnosis. Anal Chem 2001;73:1213–1218. 211. James P. Chips for proteomics: a new tool or just hype? BioTechniques 2002;33:4–13. 212. Ericsson D, Ekstrom S, Nilsson J, Bergquist J, Marko-Varga G and Laurell T. Downsizing proteolytic digestion and analysis using dispenser-aided sample handling and nanovial matrix-assisted laser/desorption ionization-target arrays. Proteomics 2001;1(9):1072–1081. 213. Litborn E, Emmer A and Roeraade J. Chip-based nanovials for tryptic digest and capillary electrophoresis. Anal Chim Acta 1999;401(1–2):11–19. 214. Ekstrom S, Ericsson D, Oennerfjord P, Bengtsson M, Nilsson J, Marko-Varga G and Laurell T. Signal amplification using ‘‘Spot-on-a-Chip’’ technology for the identification of proteins via MALDI-TOF MS. Anal Chem 2001;73(2):214–219.

201

Production of plasmid DNA for pharmaceutical use Carsten VoX Fermentation Engineering, Faculty of Technology, Bielefeld University, Germany Abstract. The concept of curing diseases at the genetic level was already introduced in the 1970s, but only the evolution of molecular biology and tools for genetic manipulation brought the idea into labs and clinics during the last 16 years. Viral and non-viral vectors and delivery systems were developed to transfer therapeutic genes into the target cells. In the case of nonviral approaches plasmid DNA has become a very promising gene delivery vector because it can easily be genetically manipulated and produced by cultivation of plasmid harbouring Escherichia coli and subsequent downstream processing, thus making production easy in comparison to other gene delivery vectors. Another advantage in using plasmid DNA is the low risk of immunogenic reactions and oncogen activation that can arise while using viral vectors. This review describes the recent development in plasmid manufacturing ranging from bacterial cultivation in batch and fedbatch mode to produce plasmid-bearing E. coli over cell lysis and subsequent purification to storage, application, and process and quality control. Keywords: gene therapy, plasmid, manufacturing, cultivation, cell lysis, Rnase, chromatography, affinity separation, capillary gel electrophoresis, stability.

Introduction The purpose of gene medicine is the introduction of therapeutic genes into mammals in order to cure hereditary or acquired diseases. First efforts were aimed at monogenetic illnesses like adenosine deaminase (ADA) deficiency, cystic fibrosis, or gaucher disease. Later on acquired diseases like cancer, HIV [1–4] or hepatitis infections [5, 6] or cardiovascular diseases became the preferred targets for several gene therapy efforts (overview: [7]). These strategies aim at correcting an insufficient or nonexistent gene function or knocking out a detrimental gene expression. In case of genetic vaccination an antigen-coding therapeutic sequence is introduced into the organism in order to create an immunogenic reaction [8]. In 2006 there were 1206 gene therapy trials [9]. In most cases (70.2%) a virus was used as delivery vector for the therapeutic genes while 17% used plasmid DNA as vector system. However, plasmid DNA has several advantages over viral vectors. Therapeutics based on plasmid DNA can be produced by one process, thus making regulatory affairs more simple. The risk of unwanted immunogenic reactions [10] or oncogen activation [11] is negligible in comparison to viral vectors. In case of vaccination different antigen-coding vectors can be used in one formulation which is important for vaccinating a larger population being the case in veterinary medicine. The recent uprise of avian flu even in parts of Europe underlines the need for a fast and simple vaccine development process. E-mail: [email protected] (C. VoX). BIOTECHNOLOGY ANNUAL REVIEW VOLUME 13 ISSN 1387-2656 DOI: 10.1016/S1387-2656(07)13008-8


202 Therefore, plasmid DNA-based vaccines might be an alternative for preventive vaccination against future epidemics. Since the amount of plasmid DNA per dose is at the mg–mg scale in human gene medicine, scalable and sustainable production processes need to be developed to meet the demands of current clinical trials and future therapeutics. Plasmid production The production process comprises several steps. At first a suitable therapeutic vector has to be transformed and amplified in E. coli cells. Next to regulatory genes for the expression of the therapeutic protein in the treated organism, this vector needs an origin of replication for E. coli and marker genes facilitating the selection of transformed cells. Plasmid vectors with high copy numbers in E. coli are the preferred molecules for cloning therapeutic sequences. These vectors are transferred into suitable E. coli host cells. In optimised cultivation processes large amounts of plasmid containing biomass is produced from which the plasmid DNA is released by cell disruption and subsequent purification. Next to chromatographic steps, several other unit operations become more interesting for this purpose while production scales increase. Analytical techniques for process analysis and quality control are also used for stability testing during storage and application. Biomass harbouring therapeutic plasmids Current clinical trials as well as future therapeutics require the production of plasmid DNA in gram and kilogram quantities. Regulatory guidelines set by the European Medicines Agency (EMEA) [12] require to omit animal-derived material throughout the production process. With respect to bacterial cultivation, this implies to resign from using extracts derived from animal sources. To avoid BSE (bovine spongiforme enzephalopathy) or TSE (transmitable spongiforme encephalopathy) risk material, the use of fully defined media for plasmid production should be favoured. A scalable cultivation process that is able to generate large amounts of biomass and fulfils the regulatory requirements is an important step in biopharmaceutical plasmid production. The production of large amounts of biomass can be achieved by high-celldensity cultivation employing fedbatch techniques. Many of these processes have been described for a variety of products like recombinant proteins [13], antibodies [14], or polyhydroxybutyric acid [15]. The feed flow is controlled by monitoring different operating variables in the bioreactor-like pH [16] or dissolved oxygen [17, 18], indirect determination of the specific growth rate [19] or by online monitoring of a limiting substrate [20]. Some of these strategies were also adapted for plasmid production. Next to increasing the biomass concentration several attempts were made to raise the average

203 plasmid copy number during cultivation. Reinikainen et al. [21] examined the influence of pH and temperature on plasmid copy number in cultivations on a semi-defined medium. However, no statement was made regarding plasmid homogeneity. Another strategy to increase the relative plasmid content was described by Lahijani et al. [22]. They used a pBR322-derived plasmid that carried a temperature sensitive point mutation. Setting the cultivation temperature to 421C in the exponential growth phase resulted in a plasmid concentration of 37 mg L 1 in batch mode and 220 mg L 1 in fedbatch mode. The isolated plasmid DNA, however, was a heterogenous product consisting of several different multimeric forms of the plasmid, and the segregative stability in E. coli was maintained by the addition of an antibiotic. A dissolved oxygen-controlled fedbatch cultivation on a defined glycerol medium was described by Schmidt et al. [18]. A product concentration of 100 mg L 1 and a dry biomass concentration of 48 g L 1 were achieved, resulting in a selectivity of 2.1 mg g 1. The isolated plasmid DNA met the requirements regarding product homogeneity with more than 90% of the plasmid DNA in the preferred supercoiled form. High cell density can also be achieved in unfedbatch operation. Glycerol is a suitable carbon source for this purpose because it can be added in high concentrations and no osmotic effects on E. coli or the formation of growth inhibiting byproducts like acetate are observable. The use of a fully defined medium with glycerol as carbon source resulted in a plasmid concentration of 45 mg L 1, while the selectivity of 2.7 mg g 1 was comparable to cultivations on semi-defined media [23]. Next to the carbon source the nitrogen supply also played a crucial role in the production process. Ammonia is readily taken up by E. coli and best results with respect to plasmid production were achieved in the presence of 37 mM ammonia. The product quality was also tested throughout the cultivation. A high homogeneity was maintained during the whole process with more than 90% supercoiled plasmid DNA. Bacterial cell disruption The first crucial step in the purification of plasmid DNA is the disruption of the bacterial cells. The large size of DNA molecules in comparison to proteins make damage by shear forces likely. For this reason mechanical lysis is considered to be unsuitable for this purpose. Other techniques like heat treatment or chemical lysis are applied in conjunction with unit operations like centrifugation or filtration where heavy shear stress may lead to damage and therefore product loss and contamination. Levy et al. [24] investigated the influence of shear forces on plasmid DNA of different sizes ranging from 13 kb to 89 kb. Their work showed that plasmids with a size bigger than or equal to 20 kb are sensitive to shear. Commonly applied therapeutic vectors have a size below this level. However, E. coli chromosomal DNA has about 48 kb and is thus susceptible to damage which will eventually result in the

204 contamination of the product stream with small fragments of DNA that are difficult to separate from the plasmid. Most of the lysis steps described in pharmaceutical plasmid production are based on the alkaline lysis originally published by Birnboim and Doly [25]. In this case a suspension of plasmid containing E. coli is mixed with a lysis buffer comprising 1% sodiumdodecylsulphate and 0.2 M NaOH. The surfactant destabilises the cell wall’s integrity and releases its cellular components. Under the alkaline conditions at a pH between 12.0 and 12.5 the host organism’s chromosomal DNA is irreversibly denaturated while the denaturation of the plasmid DNA is reversed by neutralisation with acetic potassium acetate solution. Addition of potassium acetate also results in precipitation of protein potassiumdodecylsulphate complexes, chromosomal DNA and other aggregated cellular debris as a white viscoelastic and flocculated material. Challenges for biochemical engineering arise in the scale-up of this method. Insufficient mixing of the bacterial cell suspension with alkaline lysis buffer results either in undisrupted cells in parts of the reaction volume where the pH is below 12.0 or in irreversible denaturation of plasmid DNA when the pH is above 12.5 [25]. However, agitation cannot simply be increased because the release of the high molecular weight components of the cell severely raises the viscosity of the medium [26] and under these conditions chromosomal DNA is very susceptible to damage by shear forces. In this way a reactor scale alkaline lysis is limited in its reaction volume. Another problem arises after neutralisation when protein complexes, chromosomal DNA, and cell debris are precipitated. The physical properties of this precipitate make clarification difficult because shear forces during centrifugation or filtration may also result in the damage of the chromosomal DNA. First attempts to use filter media in combination with filter aids [27] resulted in tedious and time-consuming processes. A more sophisticated method makes use of the low density of the flocculated material [28] that is drifting slowly to the surface of the reaction volume. However, this process takes at least 1 h at 101C and additional filtration for complete clarification of the drained liquid is necessary. Cell disruption in this work is also still carried out in a stirred tank reactor. To circumvent the limitations of the lysis step several authors have described a continuous flow of cell suspension and lysis solution after mixing in a small volume. Wan et al. [29] described the use of a static mixer for lysis and neutralisation. However, especially during the neutralisation step a pressure drop can occur which might also result in shear damage of chromosomal DNA. A different form of mixing is described by Wright et al. [30] where they use a Y connector to premix cell suspension and lysis buffer. The reacting solution is pumped into a stirred tank reactor where it is stirred for an additional 30 min and neutralised afterwards by the addition of acetic potassium acetate solution. The authors did not mention if stirring in the reactor resulted in the generation of small fragments of

205 Table 1. Relative plasmid concentration in comparison with a manually lysed sample with respect to the linear flow rate. Linear flow rate (cm min 1)

Relative plasmid concentration (%)

Manual lysis 280 550 830 900 1080 1400

100 86 84 78 99 63 38

chromosomal DNA in this process. Since the produced plasmid DNA was used for transient transfection of mammalian cells, no complete quality control was made that would be considered necessary for gene therapy vectors. In both cases the precipitated material has to be removed by filtration or centrifugation. The most sophisticated method was described by VoX et al. [31]. In this work cell suspension and alkaline lysis buffer where pumped through a T connector where they were efficiently mixed. Cell disruption took place in the following flow volume. A short resident time between 20 s and 40 s was sufficient for complete cell disruption (Table 1). Neutralisation was achieved by pumping the lysed solution into a column filled with acetic potassium acetate solution (Fig. 1a). For efficient mixing and separation of the precipitated material air was sparged at the bottom of the column. As can be seen in Fig. 1b this process resulted in a cleared lysate that could be processed further without any additional clarification steps. Optical density measurements showed an OD600 of 0.002. For comparison a manually lysed sample showed an OD600 of 0.03 after centrifugation and filtration. Another promising technique is based on the heat lysis originally described by Holmes and Quigley [32]. Lee and Sagar [33] described a continuous thermal lysis process which was capable of releasing sufficient amounts of plasmid DNA. However, this method still required an additional clarification step to remove denaturated precipitates. O’Mahony et al. [34] described the use of the filter aid Cellpure to bind the biomass in a filter cake where plasmid DNA can be released by the addition of hot lysis buffer. This method comprises cell harvesting and disruption. The obtained plasmid DNA showed a very high homogeneity in comparison to plasmids also prepared with thermal lysis [35]. Purification of plasmid DNA Bacterial cleared lysates usually contain several contaminants that have to be separated from the final product. As indicated in Table 2 [36] plasmid DNA

206

Fig. 1. Continuous lysis of bacterial biomass (A) and cleared lysate after froth flo-

tation (B).

207 Table 2. Constituents of Escherichia coli lysates. Content of bacterial cell lysates

(%)

Proteins RNA Host chromosomal DNA Lipopolysaccharides Plasmid DNA Others

55 21 3 3 3 15

Table 3. Important criteria for quality assurance and quality control of plasmid DNA medicines (selection). Test

Analytical method

DNA concentration General purity Homogeneity (ccc content) Purity (visible) Purity (chromosomal DNA)

UV-absorption (260 nm) UV-scan (220–320 nm) CGE (capillary gel electrophoresis) Visual inspection Agarose gel (visual); southern blot; quantitative PCR Agarose gel (visual); fluorescence assay; quantitative PCR BCA test LAL test Bioburden test; sterility test Restriction fragment lengths conform to reference in AGE (1–3 enzymes) Sequencing (double strand)

Purity (RNA) Purity (protein) Purity (LPS) Purity (microorganisms) Identity (vector structure) Identity (sequence)

only constitutes about 3% of the material applied to the purification process. Major impurities are proteins and RNA, the latter being difficult to separate because of its close structural relation to plasmid DNA. In addition, the low capacities of common stationary phases used in subsequent chromatographic separation make this part of downstream process challenging [37]. The desired level of purity can be determined by suitable detection methods for in-process and quality control as given in Table 3. Specifications regarding purity usually are considered to be guidelines and not fixed values because they are due to changes as plasmid manufacturing and gene therapy trials evolve. Therefore no fixed values are mentioned here. Regulatory guidelines can be obtained from Food and Drug Administration (FDA) [38] and EMEA [39]. The large amounts of RNA present in a bacterial cleared lysate can simply be removed by the addition of RNase A prior to chromatographic

208 purification [40, 41]. However, RNase A is usually prepared from bovine pancreas, making its use in pharmaceutical plasmid purification critical. In order to fulfil the regulatory guidelines requiring to omit any animal-derived material [12] throughout the production process, a more complex means of separation has to be used. Several methods have been described in the literature. The precipitation of RNA [42] and plasmid DNA [43–46] has also been described. Calcium chloride was used to precipitate high molecular weight RNA [42]. Lower molecular weight RNA still had to be removed by subsequent ultrafiltration. Next to product loss that occurs during preciptation this technique still requires an additional separation step to completely remove the RNA. More selective precipitation is achieved by using affinity systems. By coupling triple helix-forming oligonucleotide sequences to a thermoresponsive polymer plasmid DNA can be removed from the product stream and redisolved in suitable buffers [44]. However, these processes are only applicable at the laboratory scale or afflicted with high product loss due to unsatisfying selectivity. Extraction is a highly scalable unit operations with large capacities of the receptor phase. It usually requires only inexpensive chemicals and equipment in comparison to chromatographic media. For biopolymers the use of aqueous two-phase systems and reverse micellar phase systems has been described. The extraction of plasmid DNA by aqueous two-phase systems was shown by Ribero et al. [47]. However, high concentrations of polyethylenglycole (PEG) and potassium phosphate are necessary for efficient partitioning of DNA into the PEG phase. Under these conditions the extraction system is very susceptible to precipitation of RNA as well as DNA at the interface. For subsequent downstream processing by, e.g., anion exchange chromatography, the plasmid DNA has to be partitioned into a salt phase by a backward extraction process. Next to product loss in both extractions, the high salt concentration in the salt phase may interfere with further purification and makes additional separation steps necessary. The second extraction can be circumvented by applying a thermoseparating aqueous two-phase system [48]. In this case plasmid DNA and RNA are distributed between two aqueous phases formed by the addition of an ethylenoxid-propylenoxid-copolymer (EO50PO50) and Dextran T-500. Plasmid DNA is efficiently transferred to the EO50PO50 phase while about 80% of the contaminating RNA remained in the bottom phase. Thermoprecipitation of the copolymer left the plasmid DNA along with some RNA in an aqueous phase which will make further purification more simple. Since 20% of the RNA still contaminated the product, additional purification steps are still necessary. The concept of reverse micellar extraction originally introduced for protein purification by Hatton [49] can also be applied for DNA isolation, especially since salt concentrations in these systems are considerably lower than in aqueous two-phase systems, thus making precipitation of nucleic acids at the interface less likely. In general, distribution between a

209 reverse micellar phase and an aqueous phase is controlled mainly by pH and ionic strength. Goto et al. [50] showed the distribution of a 300 bp fragment of salmon testicle DNA in the presence of different cationic surfactants under the influence of ionic strength. In our recent studies [51] we investigated the potential of this extraction method for plasmid DNA purification. First results showed that the distribution of nucleic acids allowed the separation of plasmid DNA from RNA (Fig. 2). The capacity of the reverse micelles was another critical aspect. A loading capacity of up to 2 mg mL 1 (Table 4) showed that the capacity of these phases is even superior to common porous beads used in anion exchange chromatography of DNA. A backward extraction with a sodium chloride concentration below 0.5 M allows direct application to subsequent purification process like anion exchange chromatography.

Fig. 2. Partitioning of nucleic acids in reverse micellar systems comprising 50 mM TOMAC in isooctane under the influence of different ionic strength. Lanes 1, 4, and 7: Nucleic acid sample before extraction comprising plasmid DNA and RNA. Lanes 2 and 3: Aqueous phase after extraction in the presence of 200 mM NaCl. Lanes 5 and 6: Aqueous phase after extraction in the presence of 210 mM NaCl. Lanes 8 and 9: Aqueous phase after extraction in the presence of 220 mM NaCl.

Table 4. Capacity of reverse micellar phases comprising isooctan and TOMAC [51]. CDNA before extraction (mg L 1)

CDNA in RM phase (mg L 1)

CDNA in aqueous phase (mg L 1)

Recovery (%)

50 600 1000 1200 1400 1600 1800 2000

50.4 614.4 991.3 1233.7 1296,1 1534.7 1786.9 2068.9

0.3 0.1 0.4 0.1 0.2 0.2 0.1 0.1

101 102 99 103 93 96 99 103

210 In chromatographic separation anion exchange chromatography has dominated the field of plasmid purification in the recent years [43, 52]. Although under preparative conditions this method proved to be insufficient to completely separate RNA from plasmid DNA, it is still used in several multistep protocols for further purification of plasmid DNA in order to decrease the concentration of pyrogenic lipopolysaccharides and to concentrate the final product [53]. The use of hydrophobic interaction chromatography (HIC) has also been described for plasmid DNA purification [53, 54]. Although these processes were able to deplete a sufficient amount of RNA from the product stream this method is severely limited by the low capacity of the chromatographic media for nucleic acids. A more interesting application of HIC is demonstrated by the use of the thioaromatic ligand mercaptopyridine [53]. In the presence of 2 M ammonium sulphate this ligand selectively binds supercoiled plasmid DNA while the oc form remains unbound. This type of hydrophobic interaction chromatography is therefore suitable for polishing plasmid preparation with respect to a high content of covalently closed circular (ccc) plasmid DNA. Ultrafiltration (UF) separates molecules by size. Application of this method in plasmid purification [55] only results in the depletion of low molecular weight RNA while high molecular weight RNA still remains in the retentate together with plasmid DNA. Complete separation of plasmid DNA from RNA could also be achieved by a group separation step with Sepharose 6 FF in the presence of 2 M ammonium sulphate buffer [53], but this separation is severely limited by the low capacity of the gel filtration medium and the time-consuming operating conditions. Gustavsson et al. [56] reported the application of an anion exchange material with functional groups only present in the inner pore volume and a hydrophilic outer shell. These lid beads were able to bind RNA in the pore volume while plasmid DNA was not retained and bound by a second column filled with a conventional anion exchanger. This method is not limited by the low capacities of gel filtration media, but the separation method is once again based on the rather slow diffusion of RNA into the pore volume of the chromatographic material. Another disadvantage is the need to dilute the bacterial lysate three times in order to facilitate binding of the nucleic acids to the anion exchanger. This goes along with longer process time or the application of a diafiltration process for buffer exchange. Affinity separation being the most selective purification method can also be applied in plasmid production. Even one of the most popular forms for the isolation of recombinant proteins namely immobilised metal chelate affinity chromatography (IMAC) can be used to deplete RNA from plasmid preparations [57]. In this case the free electron pair of the nitrogen in purin bases builds a coordinative bond to the complexed metal ion on the chromatographic matrix as shown in Fig. 3. Since RNA has prolonged singlestranded areas, it can bind to the IMAC material and thus be separated from

211

Fig. 3. Adenine (right) representing one of the purine bases in nucleic acids contains the same heterocylcic imidazole ring that is also found in histidine (left). Under suitable conditions single-stranded nucleic acids can coordinate to IMAC material via the free electron pair of this structure.

double-stranded plasmid DNA. However, this strategy suffers from the same practical constraints that are common to IMAC in protein purification, namely the expensive chromatographic material and the leaking of metal ions into the product stream. Nevertheless, advances in affinity tag chemistry are likely to circumvent these constraints. Triple helix formation is also suitable for plasmid DNA purification and has been used for precipitation [44] and chromatographic separation [58], but the method suffers from the slow kinetic of the triplex helix formation and the poor chemical stability of the affinity ligands. Recent publications have demonstrated the potential of protein–DNA interaction for selective purification of plasmid DNA. The immobilisation of a zink-finger protein via a glutathione-S-transferase (GST) affinity linker was used to bind plasmid DNA with a specific recognition sequence on this chromatographic support [59]. However, the authors were not able to elute the DNA without the GST affinity linker and did not mention capacities of the chromatographic resin. In a following article [60] three different matrices (Sepharose, Fractogel, Streamline) were used for GST mediated plasmid DNA isolation and compared with respect to their protein and plasmid-binding capacity. Fractogel and Streamline showed similar results regarding their plasmid-binding capability with 0.16 mg mL 1 bound, while Sepharose had a slightly higher capacity with 0.19 mg mL 1. However, the binding experiments were conducted on a 50 mL scale and the margin of error might be relatively high. Since all three tested chromatographic carriers are porous beads, they are expected to bind plasmid DNA only on the outer surface of the particle [37]. Therefore it could be expected that results would be similar. Nevertheless, anion exchangers based on porous beads have significantly higher capacities for plasmid DNA in the range 0.5–3 mg mL 1 depending on the nature of the chromatographic carrier. Lundeberg et al. [61] purified and immobilised a lac repressor fusion protein and used it for selective purification of short DNA sequences containing the lac operator. Kumar et al. [62] optimised the purification of the lac repressor by displacement chromatography and showed its

212 retained biological activity after covalent immobilisation to a Sepharose matrix. Hasche and VoX [63] used a his-tagged lac repressor protein for selective binding of plasmid DNA. The authors were able to noncovalently immobilise the protein to different IMAC materials (Sepharose, membrane adsorber, paramagnetic Ni-beads) and tested their nucleic acid binding capabilities. As indicated in Fig. 4 plasmid DNA was successfully bound to a membrane adsorber, but binding capacity of the material was way beyond that obtained with anion exchange materials. They were also able to show that the repressor was able to bind plasmid DNA containing the lac operator and showed no binding activity when loaded with RNA. Nevertheless, noncovalent immobilisation is a convenient method to screen for several combinations of chromatographic carriers and affinity proteins and the use

Fig. 4. Binding of plasmid DNA with a recombinant lac repressor after non-covalent

immobilisation to a metall chelate membrane adsorber using a his-tag as affinity linker (negative agarose gel) [63]. Lane M: Marker. Lane 1: Plasmid sample loaded on membrane adsorber. Lane 2: Flow-through. Lane 3: Plasmid DNA fraction after elution with 10 mM isopropylthiogalactoside (IPTG).

213 of the his-tag as affinity linker allows oriented immobilisation of the DNAbinding protein. In conclusion, purification methods based on protein–DNA interaction have to be tested with regard to chemical and biochemical stability of the affinity ligands, their selectivity for double-stranded nucleic acids, and the binding capacities of stationary phases coupled with these ligands. Commonly used chromatographic material was originally designed for protein purification. As already indicated before, these porous beads have a very low capacity for plasmid DNA. Another disadvantage of porous beads is the slow diffusion of molecules into the pore volume limiting the dynamic binding capacities of these materials and thus increasing process time. These disadvantages were circumvented by the development of membrane adsorbers and monolithic columns. In the first case, microfiltration membranes are chemically modified with chromatographic active groups. The resulting membrane can simply be operated in a chromatographic mode. Monolithic materials are prepared by polymerisation of the stationary phase in the presence of porogenic substances [64] creating a sponge like structure inside the material which can be chemically modified. These columns have no void volume and the mobile phase can pass the chromatographic material by convective flow. Due to the mass transport inside membrane adsorbers and monolithic columns both show a high and flow rate independent dynamic binding capacity. Both systems have been successfully applied in plasmid purification processes [64–66]. A very simple and convenient method was demonstrated by the use of a recombinant RNase originally occuring in Bacillus amyloliquefaciens and overexpressed in E. coli [67]. This RNase Ba (also called barnase) is a 12 kDa protein. Its ribonucleolytic activity inside the cell is suppressed by coexpression of a specific inhibitor protein called barstar. Simple cation exchange chromatography allowed the isolation of RNase Ba in sufficient purity for RNA degradation (Fig. 5). The purified RNase showed no DNase activity and could effectively compete with RNase A with respect to its enzymatic properties (Fig. 6). However, it has to be pointed out that the success of such an enzyme is strongly depending on its commercial availability and costs in comparison to other means of purification used for the depletion of RNA in biopharmaceutical plasmid manufacturing.

Stability of plasmid DNA during storage and application The stability of biopharmaceutical drugs is a crucial aspect not only in downstream processing but also during storage and application. In case of plasmid DNA, a formulation has to be applied which allows protection during storage, shipment, and application, even under critical environmental conditions.

214

Fig. 5. Different fractions of RNase Ba applicable for enzymatic digestion of RNA in plasmid preparations. The homogeneous fraction was prepared by repeated cation exchange chromatography while the heterogeneous fraction was simply prepared by acid precipitation of insoluble proteins. Next to RNase Ba its inhibitor barstar and the noncovalent complex formed by both proteins are visible.

Stability assessment by capillary gel electrophoresis Capillary gel electrophoresis (CGE) has proven to be a valuable tool to determine the distribution of topological isoforms in plasmid samples [68]. Next to the preferred monomeric supercoiled or the so-called covalently closed circular (ccc) form for biopharmaceutical application several other plasmid topologies can exist as shown in Fig. 7. A relaxed circular plasmid can be formed under the influence of topoisomerase enzyme. This form has an intact double helix and is difficult to distinguish from the open circular (oc) form. The latter is generated by a single strand break due to nucleolytic activity or exposure to energetic radiation like UV light. Prolonged exposure to UV light or digestion with endonucleases results in the formation of the linear form. However, this form is seldom found in plasmid preparations. Multimeric forms like dimers and higher multimers are more often found in plasmid preparations. Their formation in the host cell is dependent on homologous recombination events. Like their monomeric counterparts, they can also exist in the ccc, oc, or linear form. To obtain a highly homogenous preparation the formation of such multimers should be avoided by the choice

215

Fig. 6. Agarose gel electrophoresis of a bacterial cleared lysate prepared according

to standard protocols without RNase (lanes 1 and 2), with bovine RNase A at a concentration of 100 mg mL 1 (lanes 3 and 4), and with recombinant RNase Ba at a concentration of 100 mg mL 1. Lane M: 1 kb DNA-marker. Complete removal of RNA below the detection limit of the ethidium bromide-stained agarose gel was achieved. In addition no degradation of the ccc plasmid was visible.

Fig. 7. Different plasmid DNA isoforms that can be found in monomeric or mul-

timeric forms. The supercoiled ccc form as well as the oc form are commonly found in plasmid preparations while the linear form is usually found as a result of digestion with restriction enzymes.

of the proper host cells. These forms can be separated by CGE using a dissolved hydoxypropylmethylcellulose as a sieving matrix. The DNA can be visualised by prestaining the samples with the fluorescent dye YOYO and online detection with laser-induced fluorescence (LIF).

216 Long-term storage analysis The long-term storage stability of plasmid DNA used in non-viral gene therapy is decisive for an efficient DNA transfer and the expression of the therapeutic gene. Walther et al. [69] correlated the stability of a LacZ encoding plasmid pCMVb over a period of 13 months. The plasmid was stored at two different temperatures, 41C and 801C, and plasmid homogeneity was analysed by CGE. Initial control studies with the plasmid DNA showed that 90% of the sample were in the ccc form, 8% were a ccc dimer, and 2% were oc forms. The change in plasmid form distribution was documented by CGE. Under this condition the homogeneity obviously remains

Fig. 8. Analysis of plasmid DNA by CGE after prolonged storage over a period of

13 months at 41C (A) and

801C (B).

217 unchanged. However, a different result is obtained with the sample stored at 41C. The amount of ccc forms is reduced and oc forms become prominent. After 13 months of storage another signal representing the linear plasmid form appears, indicating a stronger degradation process. The plasmid homogeneity during storage is summarised in Fig. 8. Storage at 801C conserves the high amount of ccc monomer form and the low fractions of ccc dimer and oc forms. At 41C the degradation of plasmid DNA is already observed after 6 months of storage. The data obtained by this stability analysis correlated with in vivo transfer efficiencies by jet-injection show that suitable storage conditions not only stabilise the plasmid isoforms but also ensure reproducible results in in vivo gene transfer applications. Stability during application Next to suitable storage conditions the method of application can influence the efficiency of the gene expression in vivo. To determine appropriate conditions for DNA delivery by jet-injection experiments CGE analysis of plasmid homogeneity was also applied to samples used in such a hydrodynamic injection method [70]. As the results in Table 5 indicate, jet-injection below pressures of 2.5 bar does not significantly degrade the ccc form in the sample, but pressures above this level resulted in a significant increase in the amount of oc forms. Nevertheless, jet-injection at pressures below 2.5 bar showed insufficient gene expression inside the tumour tissue probably due to ineffective transfer of the DNA into the target material. In consequence, appropriate conditions for efficient DNA delivery into different target tissues have to be developed taking plasmid size and stability as well as efficient penetration into account. Table 5. Plasmid topology distribution of untreated pCMVb sample and after jetinjection of pCMVb at different pressures [70]. Sample

Sample description

ccc-form (%)

oc-form (%)

Sample 2 Sample 4

DNA before filling into sample space DNA from sample space after injection at 2.0 bar DNA from sample space after injection at 2.5 bar DNA from sample space after injection at 3.0 bar Ejected DNA after injection at 2.0 bar Ejected DNA after injection at 2.5 bar Ejected DNA after injection at 3.0 bar

97.6 93.5

2.4 6.5

85.4

14.6

81.6

18.4

93.3 84.6 77.2

6.7 15.4 22.8

Sample 5 Sample 6 Sample 7 Sample 8 Sample 9

218 Conclusions The increasing demand for pharmaceutical plasmid DNA has resulted in several different efforts to facilitate scalable and sustainable processes for the manufacturing of this therapeutic. High-cell-density batch and fedbatch cultivations on defined media allow the production of plasmid-bearing biomass at the kilogram scale. The application of continuous alkaline lysis and froth flotation can readily be applied for fast cell disruption without the risk of contamination of the product stream by sheared fragments of chromosomal DNA. In the field of downstream processing several other unit operations have been applied instead or in combination with chromatographic separations. Extraction processes have a high potential because they are very scalable and require inexpensive equipment. High capacities and sufficient selectivity are advantages that justify further research in this field. The use of a recombinant RNase can certainly simplify further purification. However, the success of such a process is depending on the commercial availability of the enzyme. In the field of chromatographic separation two developments have to be considered. The large size of plasmid DNA outruled the use of conventional porous beads that were originally developed for protein purification. The limitation of these materials was circumvented by the development of membrane adsorbers and monolithic columns. In both cases transport is facilitated by convective flow resulting in high flow rate independent dynamic capacities. The use of affinity ligands allows very selective purification of plasmid DNA or depletion of contaminants. Their use is currently limited by the low stability of these ligands as well as their high prize. However, sophisticated development platforms should allow faster screening of more stable and very selective affinity systems. In combination with membrane adsorbers or monolithic columns these systems may be able to compete with current multistep protocols for biopharmaceutical plasmid purification. The use of capillary gel electrophoresis in quality control and for the analysis of product stability has also been demonstrated. References 1. 2.

3.

Mascola J and Nabel G. Vaccines for the prevention of HIV-1 disease. Curr Opin Immunol 2001;13:489–495. Shiver JW, Fu T-M, Chen L, Casimiro DR, Davies M-E, Evans RK, Zhang Z-Q, Simon AJ, Trigona WL, Dubey SA, et al. Replication incompetent adenoviral vaccine vector elicits effective anti-immunodeficiency-virus immunity. Nature 2002;415:331–335. Barouch DH, Craiu A, Kuroda MJ, Schmitz JE, Zheng XX, Santra S, Frost JD, Krivulka GR, Lifton MA, Crabbs CL, et al. Augmentation of immune responses to HIV-1 and simian immunodeficiency virus DNA vaccines by IL-2/Ig plasmid administration in rhesus monkeys. Proc Natl Acad Sci USA 2002;97:4192–4197.

219 4.

5.

6.

7. 8. 9. 10.

11. 12.

13.

14.

15.

16. 17.

18.

19.

20.

Barouch DH, Santra S, Schmitz JE, Kuroda MJ, Fu T-M, Wagner W, Bilska M, Craiu A, Zheng XX, Krivulka GR, et al. Control of virimia and prevention of clinical AIDS in rhesus monkeys by cytokine-augmented DNA vaccination. Science 2002;290:486–492. Michel M-L, Davis HL, Schleef M, Mancini M, Tiollais P and Whalen RG. DNAmediated immunization to the hepatitis B surface antigen in mice: aspects of the humoral response mimic hepatitis B viral infection in humans. Proc Natl Acad Sci USA 1995;92: 5307–5311. Major ME, Vitvitski L, Mink MA, Schleef M, Whalen RG, Tre´po C and Inchauspe´ G. DNA based immunisation using chimeric vectors for the induction of immune responses against the hepatitis C virus nucleocapsid. J Virol 1995;69:5798–5805. Mountain A. Gene therapy: the first decade. Trends Biotechnol 2000;18:119–128. Wolff JA, Malone RW, Williams P, Chong W, Acsadi G, Jani A and Felgner PL. Direct gene transfer into mouse muscle in vivo. Science 1990;247:1465–1468. Wiley. Gene therapy clinical trials worldwide, 2006, http://www.wiley.co.uk/genmed/ clinical/ Raper SE, Yudkoff M, Chirmule N, Gao G-P, Nunes F, Haskal ZJ, Furth EE, Propert KJ, Robinson MB, Magosin S, et al. A pilot study of in vivo liver directed gene transfer with adenoviral vector in partial ornitine transcarbamylase deficiency. Human Gene Ther 2002;13:163–175. Check E. Gene therapy: a tragic setback. Nature 2002;420:116–118. EMEA. Note for guidance on minimising the risk of transmitting animal spongiform encephalopathy agents via human and veterinary medicinal products, CPMP/410/01 rev 1, London, 2001. Schroeckh V, Hartmann M, Birch-Hirschfeld E and Riesenberg D. Improvement of recombinant gene-expression in Escherichia coli for glucose-controlled continuous and fed-batch cultures. Appl Microbiol Biotechnol 1992;36:487–492. Horn U, Strittmatter W, Krebber A, Knu¨pfer U, Kujau M, Wenderoth R, Mu¨ller K, Matzku S, Plu¨ckthun A and Riesenberg D. High volumetric yields of functional dimeric miniantibodies in Escherichia coli, using an optimized expression vector and high-celldensity fermentation under non-limited growth conditions. Appl Microbiol Biotechnol 1996;46:524–532. Wang F and Lee SY. High cell density culture of metabolically engineered Escherichia coli for the production of poly(3-hydroxybutyrate) in a defined medium. Biotechniol Bioeng 1998;58:325–328. Lee SY and Chang HN. High cell density cultivation of Escherichia coli using sucrose as a carbon source. Biotechnol Lett 1994;15:971–974. Nakano K, Rischke M, Sato S and Ma¨rkl H. Influence of acetic acid on the growth of Escherichia coli K12 during high-cell-density cultivation in a dialysis reactor. Appl Microbiol Biotechnol 1997;48:597–601. Schmidt T, Schleef M, Friehs K and Flaschel E. Hochzelldichtefermentation zur Gewinnung von Plasmid-DNA fu¨r Gentherapie und genetische Impfung. BIOforum 1999;22:174–177. Macaloney G, Hall JW, Rollins MJ, Draper I, Anderson KB, Preston J, Thompson BG and McNeil B. The utility and performance of near-infrared spectroscopy in simultaneous monitoring multiple components in a high cell density recombinant Escherichia coli production process. Bioprocess Eng 1997;17:157–167. Paalme T, Tiisma K, Kahru A, Vanatalu K and Vilu R. Glucose-limited fed-batch cultivation of Escherichia coli with computer-controlled fixed growth rate. Biotechnol Bioeng 1990;35:312–319.

220 21. Reinikainen P, Korpela K, Nissinen V, Olkku J, So¨derlund H and Markkanen P. Escherichia coli plasmid production in fermenter. Biotechnol Bioeng 1989;33:386–393. 22. Lahijani R, Hulley G, Soriano G, Horn NA and Marquet M. High-yield production of pBR322-derived plasmids intended for human gene therapy by employing a temperaturecontrollable point mutation. Hum Gene Ther 1996;7:1971–1980. 23. Voss C, Schmidt T, Schleef M, Friehs K and Flaschel E. Effect of ammonium chloride on plasmid DNA production in high cell density batch culture for biopharmaceutical use. J Chem Technol Biotechnol 2004;79:57–62. 24. Levy MS, O’Kennedy RD, Ayazi-Shamlou P and Dunnill P. Biochemical engineering approaches to the challenges of producing pure plasmid DNA. Trends Biotechnol 2000;18:296–305. 25. Birnboim HC and Doly J. A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res 1979;7:1513–1523. 26. Ciccolini LAS, Shamlou PA, Titchener-Hooker NJ, Ward JM and Dunnill P. Time course of SDS-alkaline lysis of recombinant bacterial cells for plasmid release. Biotechnol Bioeng 1998;60:768–770. 27. Theodossiou I, Collins IJ, Ward JM, Thomas ORT and Dunnill P. The processing of a plasmid-based gene from E. coli. Primary recovery by filtration. Bioprocess Eng 1997;16:175–183. 28. Theodossiou I, Thomas ORT and Dunnill P. Methods of enhancing the recovery of plasmid genes from neutralised cell lysate. Bioprocess Eng 1999;20:147–156. 29. Wan NC, McNeilly DS and Christopher CW. Method for lysing cells, US Patent 5 837 529, 1998. 30. Wright JL, Jordan M and Wurm FM. Extraction of plasmid DNA using reactor scale alkaline lysis and selective precipitation for scalable transient transfection. Cytotechnology 2001;35:165–173. 31. VoX C, Schmidt T and Schleef M. From bulk to delivery: plasmid manufacturing and storage. In: DNA Pharmaceuticals, Schleef M (ed), Weinheim, Germany, Wiley-VCH, 2005. 32. Holmes D and Quigley M. A rapid boiling method for the preparation of bacterial plasmids. Anal Biochem 1981;114:193–197. 33. Lee AL and Sagar S. A method for large scale plasmid purification, WO 96/36706, 1999. 34. O’Mahony K, Freitag R, Hilbrig F, Mu¨ller P and Schumacher I. Proposal for a better integration of bacterial lysis into the production of plasmid DNA at large scale. J Biotechnol 2005;119:118–132. 35. Zhu K, Jin H, Ma Y, Ren Z, Xiao C, He Z, Zhang F, Zhu Q and Wang B. A continuous thermal lysis procedure for the large-scale preparation of plasmid DNA. J Biotechnol 2005;118:257–264. 36. Stadler J, Lemmens R and Nyhammar T. Plasmid DNA purification. J Gene Med 2004;6:S54–S66. 37. Ljunglo¨f A, Bergvall P, Bhikhabhai R and Hjorth R. Direct visualisation of plasmid DNA in individual chromatography adsorbent particles by confocal scanning laser microscopy. J Chromatogr A 1999;844:129–135. 38. FDA. Guidance for Industry: Guidance for Human Somatic Cell Therapy and Gene Therapy, Rockville, MD, Center for Biologics Evaluation and Research, FDA, 1998. 39. EMEA. Safety studies for gene therapy products, CPMP/SWP/112/98 draft, London, 1998. 40. Schorr J, Moritz P and Schleef M. Production of plasmid DNA in industrial quantities accorging to cGMP guidelines. In: DNA Vaccines: Methods and Protocols, Lowrie DB and Whalen RG (eds), Totowa, NJ, Humana Press, 1999, pp. 11–21.

221 41. Bussey L, Adamson R and Atchley A. Methods for purifying nucleic acids, WO 98/ 05673, 1998. 42. Eon-Duval A, Gumbs K and Ellett C. Precipitation of RNA impurities with high salt in a plasmid DNA purification process: use of experimental design to determine reaction conditions. Biotechnol Bioeng 2003;83:544–553. 43. Horn N, Budahazi G and Marquet M. Purification of plasmid DNA during column chromatography, US 5 707 812, 1998. 44. Costioli MD, Fisch I, Garret-Flaudy F, Hilbrig F and Freitag R. DNA purification by triple-helix affinity precipitation. Biotechnol Bioeng 2003;81:535–545. 45. Murphy JC, Wibbenmeyer JA, Fax GE and Willson RC. Purification of plasmid DNA using selective precipitation by compaction agents. Nature Biotchnol 1999;17: 822–823. 46. Lander RJ, Winters MA, Meacle FJ, Buckland BC and Lee AL. Fractional precipitation of plasmid DNA from lysate by CTAB. Biotechnol Bioeng 2002;79:776–784. 47. Ribeiro SC, Monteiro GA, Cabral JMS and Prazeres DMF. Isolation of plasmid DNA from cell lysates by aqueous two-phase systems. Biotechnol Bioeng 2002;78:376–384. 48. Kepka C, Lemmens R, Vasi J, Nyhammar T and Gustavsson P-E. Integrated process for purification of plasmid DNA using aqueous two-phase systems combined with membrane filtration and lid bead chromatography. J Chromatogr A 2004;1057:115–124. 49. Hatton TA. Reversed micellar extraction of proteins. In: Surfactant-Based Separation Processes, Scamehorn JF and Harwell JH (eds), New York, Marcel Dekker Inc., 1989. 50. Goto M, Ono T, Horiuchi A and Furusaki S. Extraction of DNA by reversed micelles. J Chem Eng Japan 1999;32:123–125. 51. VoX C. Innovative Prozessstrategien zur Herstellung von Plasmid-DNA als Wirkstoff. BIOforum 2004;9:38–39. 52. Colpan M, Schorr J and Moritz P. Process for producing endotoxin-free or endotoxinpoor nucleic acids and/or oligonucleotides for gene therapy, WO 95/21177, 1995. 53. Lemmens R, Olsson U, Nyhammar T and Stadler J. Supercoiled plasmid DNA: selective purification by thiophilic/aromatic adsorption. J Chromatogr B 2003;784:291–300. 54. Diogo MM, Queiroz JA and Prazeres DMF. Studies on the retention of plasmid DNA and Escherichia coli nucleic acids by hydrophobic interaction chromatography. Bioseparation 2001;10:211–220. 55. Eon-Duval A, MacDuff RH, Fisher CA, Harris MJ and Brook C. Removal of RNA impurities by tangential flow filtration in an RNase-free plasmid DNA purification process. Anal Biochem 2003;316:66–73. 56. Gustavsson P-E, Lemmens R, Nyhammar T, Busson P and Larsson P-O. Purification of plasmid DNA with a new type of anion-exchange beads having a non-charged surface. J Chromatogr A 2004;1038:131–140. 57. Murphy JC, Jewell DL, White KI, Fox GE and Willson RC. Nucleic acid separations utilizing immobilized metal affinity chromatography. Biotechnol Prog 2003;19:982–986. 58. Schluep T and Cooney CL. Purification of plasmids by triplex interaction. Nucleic Acids Res 1998;26:4524–4528. 59. Woodgate J, Palfrey D, Nagel DA, Hine AV and Slater NKH. Protein-mediated isolation of plasmid-DNA by a zinc finger-glutathione S-transferase affinity linker. Biotechnol Bioeng 2002;79:450–456. 60. Ghose S, Forde GM and Slater NKH. Affinity adsorption of plasmid DNA. Biotechnol Prog 2004;20:841–850. 61. Lundeberg J, Wahlberg J and Uhlen M. Affinity purification of specific DNA fragments using a lac repressor fusion protein. Genet Anal Tech Appl 1990;7:47–52.

222 62. Kumar A, Galaev IY and Mattiasson B. Purification of Lac repressor protein using polymer displacement and immobilization of the protein. Bioseparation 1999;8:307–316. 63. Hasche A and VoX C. Immobilisation of a repressor protein for binding of plasmid DNA. J Chromatogr A 2005;1080:76–82. 64. Strancar A, Podgornik A, Barut M and Necina R. Short monlithic columns as stationary phases for biochromatography. Adv Biochem Eng Biotechnol 2002;76:49–85. 65. Giovannini R, Freitag R and Tennikova TB. High-performance membrane chromatography of supercoiled plasmid DNA. Anal Chem 1998;70:3348–3354. 66. Urthaler J, Schlegl R, Podgornik A, Strancar A, Jungbauer A and Necina R. Application of monoliths for plasmid DNA purification Development and transfer to production. J Chromatogr A 2005;1065:93–106. 67. Voss C, Lindau D and Flaschel E. Production of recombinant RNase Ba and its application in downstream processing of plasmid DNA for pharmaceutical use. Biotechnol Prog 2006;22:737–744. 68. Schmidt T, Friehs K, Schleef M, Voss C and Flaschel E. Quantitative analysis of plasmid forms by agarose and capillary gel electrophoresis. Anal Biochem 1999;274:235–240. 69. Walther W, Stein U, Voss C, Schmidt T, Schleef M and Schlag PM. Stability analysis for long-term storage of naked DNA: impact on nonviral in vivo gene transfer. Anal Biochem 2003;318:230–235. 70. Walther W, Stein U, Fichtner I, VoX C, Schmidt T, Schleef M, Nellessen T and Schlag PM. Intratumoral low volume jet-injection for efficient non-viral gene transfer. Mol Biotechnol 2002;21:105–115.

223

Potentials of phenolic molecules of natural origin and their derivatives as anti-HIV agents Mahmud Tareq Hassan Khan1,2, and Arjumand Ather3 1 Pharmacology Research Laboratory, Faculty of Pharmaceutical Sciences, University of Science and Technology Chittagong, Chittagong, Bangladesh 2 School of Molecular and Structural Biology and Department of Pharmacology, Institute of Medical Biology, University of Tromsø, Tromsø 9037, Norway 3 The Norwegian Structural Biology Center (NorStruct), Department of Chemistry, University of Tromsø, Tromsø 9037, Norway

Abstract. Identification of phenolic compounds and their derivatives interfering the several steps of the viral life cycle of the human immunodeficiency virus type 1 (HIV-1) is focused for the development of novel molecules for the treatment of AIDS. Several phenolic compounds isolated and characterized from natural sources have been studied in detail and found to exhibit inhibitory effects against different steps of the HIV-1 life cycle, including virus–cell fusion and virus absorption, reverse transcription, integration (IN) and proteolytic cleavage. In the review, we are summarizing some strong evidences demonstrating several phenolic molecules and their derivatives from natural sources display promising anti-HIV-1 activities. The anti-HIV compounds have been organized in this review according to their mechanism of action in the life cycle of HIV. We also mentioned some findings using in silico approaches, like virtual screening, docking, neural network, etc., and even the chemogenomics and/or functional genomics approaches could be useful for the quick identifying promising new lead anti-HIV molecules without having any other unwanted pharmacological effects. Plants having large amount of phenolic compounds, can be considered as strong sources of molecules for the treatment of HIV-1. Despite the continuous advances made in antiretroviral combination therapy, AIDS has become the leading cause of death in Africa and the fourth worldwide. Today, many research groups are exploring the bio- and chemo-diversity of the plant kingdom to find new and better anti-HIV drugs with novel mechanisms of action [1]. Keywords: HIV, anti-HIV, virtual screening, docking, neural network, natural products’ databases, chemogenomics, phenolics, flavonoids, coumarin, chalcone, catechin, gallate, reverse transcriptase, integrase, protease, HIV promoter, Protein Databank.

Abbreviations used 3D 3TC ABC AIDS AZT BGCDG CFI

three dimensional lamivudine Abacavir acquired immunodeficiency syndrome zidovudine 20 ,200 -bisepigallocatechin digallate chalcone-flavanone isomerase

Corresponding author: Tel (Off): +47-776-46755; Tel (Cell): +47-97794171.

Fax: +47-776-45310. E-mail: [email protected]; [email protected] (M.T.H. Khan). BIOTECHNOLOGY ANNUAL REVIEW VOLUME 13 ISSN 1387-2656 DOI: 10.1016/S1387-2656(07)13009-X


224 CG CS d4T DCQAs DCTAs ddC DDDP ddI dTTP ECG EGCDG EGCG Fis FL FO FS GCG GSE HAART HIV-1 IN LPS Luc MLR NN PAL PARG PBMC PDB PHA Pis PLS PR RDDP RNase H ROS RT RTis-N RTis-NN SAR SCRIP’s SOM TF TPA

catechin gallate chalcone synthase stavudine dicaffeoylquinic acids dicaffeoyltartaric acids zalcitabine DNA-dependent DNA polymerase didanosine di-methyl thymidine triphosphate epicatechin gallate epigallocatechin 3,5-digallate epigallocatechin gallate fusion inhibitors flavanone lyase flavanone oxidase flavonol synthase gallocatechin gallate grape seed extract highly active antiretroviral therapy human immunodeficiency virus type 1 integrase lipopolysaccharide luciferase multiple linear regression neural network L-phenylalanine-ammonia lyase poly(ADP-ribose) glycohydrolase peripheral blood mononuclear cells Protein Databank phytohemagglutinin protease inhibitors partial least squares protease RNA-dependent DNA polymerase ribonuclease H reactive oxygen species reverse transcriptase nucleoside reverse transcriptase inhibitors non-nucleoside reverse transcriptase inhibitors structure–activity relationship single-chain ribosome inactivating proteins self-organizing map theaflavin 12-O-tetra-decanoyl phorbol-13-acetate

225 UNAIDS VS

United Nations Program on AIDS virtual screening

Introduction The acquired immunodeficiency syndrome (AIDS) is caused by the human immunodeficiency virus (HIV) and is an immunosuppressive disease resulting in life-threatening opportunistic infections and malignancies. First reported in 1981 in the United States, AIDS has become a major worldwide epidemic. The United Nations Program on AIDS (UNAIDS) estimated that at the end of 2002 nearly 42 million have died of AIDS. During 2002, about 3 million people became infected. AIDS is presently the leading cause of death in Africa and the fourth leading cause of death worldwide [1]. The AIDS and its therapy The causative agent HIV is a member of the lentivirus family of animal retroviruses. Retroviruses carry their genome as RNA, packaged in a protein capsid and surrounded by a lipid envelope. One of the proteins encoded by the retroviral genome is the enzyme reverse transcriptase (RT), which is responsible for the synthesis of a complementary DNA molecule, using viral RNA as a template. Of the two known HIV types, HIV-1 is most pathogenic and causes over 99% of HIV infections, while HIV-2 is much less pathogenic and is endemic in West Africa [1]. The life cycle of the HIV-1 is one of the major targets for the development of pharmaceutical molecule. Accordingly, significant efforts have been made in the recent past to identify molecules inhibiting the different biological steps of the HIV-1 life cycle. In this respect, this research field takes great advantage from the fact that the molecular biology of the HIV-1 life cycle is well known and has been the object of several excellent research papers and reviews [2–9]. The most important steps in HIV-1 infection are virus–cell attachment, gp120-CD4 binding, gp120-coreceptor binding, viral fusion, viral assembly and disassembly, reverse transcription, nuclear import of the pre-integration complex, proviral integration, viral transcription, processing of viral transcripts and nuclear export, assembly of new virions. In addition to HIV-1 proteins, several cellular factors are involved in HIV replication [4,6]. During the last two decades, anti-HIVtherapy has made great progress with the appearance of 11 inhibitors of reverse transcriptase (8 NRTIs and 3 NNRTIs) and 8 inhibitors of protease and with the birth of a fourth new class of anti-HIV drugs, i.e., fusion inhibitors [10]. Nevertheless, treating HIV with antiviral agents can be quite expensive. HIV-1 treatment using RTIs and

226 PIs cost about 20,000 US$ per patient per year. Reinforcing the anti-HIV therapy by a convenient diet may limit this cost [11]. Since 1996, highly active antiretroviral therapy (HAART) was designed to rapidly and efficiently control HIV replication. As recently described by Cos et al. [1], antiviral research has been focused on compounds that interfere with the different steps of the viral life cycle. For example, most of the current anti-HIV drugs are targeted against proteins encoded by the virus itself, i.e., RT and protease (PR). However, since several cellular factors are assumed to be involved in the replication of HIV or in HIV pathogenesis, it would seem logical to develop a complementary anti-HIV strategy, targeting both viral and cellular factors (Table 1 and Fig. 1). As far as anti-HIV-1 drugs, the current therapeutic approach is based on the combined use of different molecules, such as zidovudine (AZT), enfuvirtide (the first fusion inhibitor), tenofovir (an inhibitor of RT), atazanavir (a protease inhibitor), tipranavir (another protease inhibitor) [12–15].

Table 1. A partial list of antiviral molecules (for structures, see Fig. 1) approved for the treatment of HIV infections. Antiviral molecules (codes used)

Commercial names

Classes (according to their mechanism of action)

Zidovudine (AZT) Didanosine (ddI) Zalcitabine (ddC) Stavudine (d4T) Lamivudine (3TC) Abacavir (ABC) Tenofovir Nevirapine Delavirdine Efavirenz Saquinavir Ritonavir Indinavir Nelfinavir Amprenavir Lopinavir Enfuvirtide

Retrovir Videx Hivid Zerit Epivir Ziagen Viread Viramune Rescriptor Sustiva Invirase, Fortovase Norvir Crixivan Viracept Agenerase Kaletrad Fuzeon

RTi-Na RTi-N RTi-N RTi-N RTi-N RTi-N RTi-N RTi-NNb RTi-NN RTi-NN PRic Pri Pri Pri Pri Pri Fie

Source: Developed from Cos et al. (2004). a RTi-N are nucleoside reverse transcriptase inhibitor. b RTi-NN are non-nucleoside reverse transcriptase inhibitor. c PRi are protease inhibitor. d Used in combination with ritonavir. e Fi is fusion inhibitor.

227

Fig. 1. The structural features of different anti-HIV molecules.

As far as possible, interest on medicinal plants and compounds isolated from them, it is relevant to note that by simply looking at the recent literature, large number of scientific reports and reviews have been published in which plant extracts have been claimed to exhibit anti-HIV-1 activity [16–26]. In 1992, Schinazi reviewed the sources, structures and biological activities of the natural products having the ability to treat HIV infection on the following chemical classes phenolics, terpenoids, alkaloids, peptides and carbohydrates [22,23]. In 1997 a mini-review has been published by Ng et al. [20], which aimed at summarizing research findings concerning natural products which are endowed with the ability to inhibit HIV. An emphasis was placed on HIV RT inhibitors (RTis) because the bulk of the literature was focused on these compounds. Ng and co-workers [20] found that a spectacular diversity of chemical structures encompassing proteins, terpenoids, coumarins, xanthones, alkaloids, flavonoids, polyphenols and polysaccharides, which are elaborated by plant species as phylogenetically remote as the algae, gymnosperms and angiosperms, were capable of rendering the retroviral enzyme less active. Some natural polyphenols generally belonging to the super-family of lignans (including neolignans and norlignans) were found to be active as antiviral during wide screenings of traditional medicinal plants [11]. In this review, we are reporting about the members of the polyphenol super-family and their activities against the HIV. The compounds have been distributed in this review according to their mechanism of action(s) in the life cycle of HIV. We also discussed about some findings dealing with the

228 in silico, like virtual screening (VS), docking, neural network (NN), etc., approaches utilized for the promising anti-HIV lead molecules. Plant materials and extracts against HIV-1: Few examples Recently, Kanyara and Njagi [27] reported screening studies of anti-HIV activity of extracts obtained from Kenyan medicinal plants through an in vitro HIV-1 RT assay. In the assay they have employed [ 3H]-methyl thymidine triphosphate (dTTP) as the enzyme substrate and polyadenylic acid.oligodeoxythymidylic acid [poly(rA).p(dT)] as the template-primer dimer. The assay has been optimized and standardized with respect to the various experimental parameters in a micro-titer plate methodology [27]. The assay was then applied to test for potential antiviral activities of several Kenyan medicinal plant extracts and the IC50 of the HIV-1 RT were determined [27]. Table 2 shows the species used in this study the family, parts used and types of extracts used and their IC50 values, where phosphonoformic acid has been used as reference inhibitor. el-Mekkawy et al. [28] reported the inhibitory effects of Egyptian folk medicines on HIV-1 RT. They have examined the activities of the extracts of 41 medicinal plants and found that the extracts of fruits of Phyllanthus emblica, Quercus pedunculata, Rumex cyprius, Terminalia bellerica, Terminalia chebula and Terminalia horrida retain significant inhibitory activity on HIV-1 RT [28]. Interestingly, this study led, through a bioassay guided-fractionation Table 2. Anti-HIV activities of the Kenyan medicinal plant extracts and their IC50 values. Name of the species

Family

Parts used

Types of the extract

IC50 (mg/ml)

Maytenus buchanani Maytenus senegalensis Prunus Africana Acacia mellifera

Celastraceae

Stembark

6.0

Celastraceae

Roots

Rosaceae Leguminosae

Stem bark Stem bark

Rhus natalensis

Anacardiaceae

Leaves

Vernonia jugalis

Compositae

Leaves

Melia azedarach

Meliaceae

Leaves

Hot water extract Hot water extract Hot water Hot water extract Warm water extract Warm water extract Warm water extract

Phosphonoformic acid Source: Modified from Kanyara and Njagi (2005) [27].

4.0 20.0 3.5 5.5 18.0 7.0 0.15

229 of the methanol extract of the fruit of P. emblica, to the isolation of putranjivain A as a potent inhibitory compound, together with 1,6-diO-galloyl-b-D-glucose, 1-O-galloyl-b-D-glucose, kaempferol-3-O-b-D-glucoside, quercetin-3-O-b-D-glucoside and digallic acid [28]. In a similar study, Park et al. [29], in order to identify substances with antiHIV activity in traditional medicines, screened 101 extracts of Korean medicinal plants for their inhibitory effects on HIV-1 PR. Among the extracts tested, strong inhibitory effects were observed in the acetone extracts of the pericarp and leaves of Camellia japonica, in the aqueous extract of the leaves of Sageretia theezans and in the methanol extract of the aerial part of Sophora flavescens. Camelliatannin H, isolated from the pericarp of C. japonica, showed a potent inhibitory activity on HIV-1 PR. To identify substances with anti-HIV activity, Park et al. [30] recently screened and reported 12 extracts from Rosaceae family for their inhibitory effects against HIV-1 PR. Among the extracts tested, the strongest inhibitory effects have been observed in the root of Rosa rugosa and the leaves of Prunus sargentii. Rosamultin has been isolated from the root of R. rugosa and inhibited HIV-1 PR by 53% at a conc. of 100 mM. In another recent study, Bessong and co-workers [31] screened 17 aqueous and methanol extracts from ethnobotanically selected nine South African medicinal plants against HIV-1 RT. Additionally, they have evaluated antiHIV activity of isolated compounds. The strongest inhibition for the RDDP activities of RT reported to be observed with the methanol extract of the stem-bark of Peltophorum africanum (Fabaceae) (IC50 3.5 mg/ml), while the methanol extract of the roots of Combretum molle (Combretaceae) reported to be the most inhibitory for the RNase H activities (IC50 ¼ 9.7 mg/ml) [31]. The known compounds bergenin and catechin, and a red colored gallotannin composed of m-depside chains of gallic and protocatechuic acids esterified to a 1-O-isobutyroly-b-D-glucopyranose core, have been isolated from the methanol extract of the roots and stem-bark of Peltophorum africanum. The gallotannin inhibited the RDDP and RNase H functions of RT with IC50 values of 6.0 and 5.0 mM, respectively, and abolished the 30 -end processing activity of IN at 100 mM. Catechin exhibited no effect on the RT but showed moderate inhibition against the HIV-1 IN. Bergenin did not show any activity against both the enzymes. The aqueous and methanol extracts were non-toxic in a HeLaP4 cell line at a conc. of 400 mg/ml [31]. In a very recently reported study, Paskaleva et al. [32] Sargassum fusiforme extract blocked HIV-1 infection and replication by over 90% in T cells, human macrophages and microglia, and it also inhibited the pseudotyped HIV-1 (VSV/NL4-3) infection in human astrocytes in more than 70% [32]. Their results with VSV/NL4-3 infection, suggested that the inhibition of both the entry and post-entry events of the virus life cycle. Absence of cytotoxicity and high viability of treated cells also suggest that S. fusiforme is a potential source of novel naturally occurring antiretroviral compounds that inhibit HIV-1 infection and replication at more than one site of the virus life cycle [32].

230 In silico approaches for the discovery of anti-HIV molecules from natural products The in silico approaches, like VS and docking small molecules into a known protein structure of a clinically important drug target is a powerful tool for drug design. Recently, Sangma and co-workers [33] used combined approaches of VS, docking of small molecule database and NN of natural products to design new anti-HIV molecules. For these studies they have utilized self-organizing map (SOM), which has been developed and applied to screen anti-HIV-1 molecules for two targets, HIV-1 RT and HIV-1 PR, from active compounds available in the Thai Medicinal Plants Database. Based on nevirapine and calanolide A as reference molecules against the HIV-1 RT binding site and XK-263 for the HIV-1 PR binding site, 2,684 compounds in the database have been docked into the target enzymes [33]. SOMs were generated with respect to three types of pharmacophoric groups. The SOM of the reference molecules were superimposed on the feature maps of all screened compounds. Only the structures having similar features to the reference compounds were accepted. By using the SOMs, the number of candidates for HIV-1 RT have been reduced to six and nine compounds consistent with nevirapine and calanolide A, respectively, as references. For the HIV-1 PR target, about 135 screened compounds exhibited promising agreement with the SOM of XK-263 feature [33]. Now finally Sangma and co-workers [33] are planning to screen these compounds for further testing for their HIV-1 inhibitory affinities. The results obtained from the studies of Sangma et al. [33] indicated that this combined method is clearly helpful to perform the successive screening and to reduce the analyzing step from AutoDock and scoring procedure [33]. Alves and co-workers [34] employed the molecular orbital semi-empirical method PM3 approaches to calculate a set of molecular properties (variables or descriptors) of 21 flavonoids with anti-HIV activity. The correlation between anti-HIV activity and the structural properties was obtained by using the multiple linear regression (MLR) and partial least squares (PLS) methods. The model obtained showed not only statistical significance but also predictive ability. The significant molecular descriptors related to the compounds with anti-HIV activity were: electronegativity (w) and the charges on atoms C3 and C7 (Q3 and Q7, respectively). These variables led to a physical explanation of electronic molecular property contributions to HIV inhibitory potency[34]. Figure 2 shows the basic skeleton of the flavonoids used in these studies. Docking and virtual screening Large numbers of anti-HIV potential molecules have been identified using different docking and VS techniques. There are some very good reviews

231

Fig. 2. The basic skeleton of the flavonoid taken considerations in the study of Alves

et al. [34].

published, which described much detailed about the techniques utilized for the in silico identification of anti-HIV compounds [35,36]. Several research papers published, which mentioned about the VS techniques for easy identification of anti-HIV candidates from large libraries of small molecules [35–43]. Even recently, Sangma et al. [35] described the utilities of combined VT and NN technologies to discover anti-HIV molecules from databases of natural product origins.

Caffeic acid derivatives from food supplements against HIV The HAART strategy still suffers from issues of patient compliance, cost, deleterious side effects and emerging drug resistance. Therefore, expansion of novel anti-HIV drugs and targets will be critical in the coming years. In this context, discovering anti-HIV agents from natural sources and particularly from plants, food, etc., may highlight the principle of a nutritional antioxidant antiretroviral diet [11]. Very recently, Bailly and Cotelle [11] reviewed the putative anti-HIV activity of simple caffeic acid derivatives. In their review they have also mentioned that, well-known caffeic acid derivatives, such as chicoric, rosmarinic and lithospermic acids, may be designed as future leads multi-target anti-HIV compounds and the plants and vegetables containing them as potent nutritional therapeutic supplementation source. They are not expected to replace the actual antiretroviral therapy, but more likely, to complete and perhaps lighten it by adapted diet. Figure 3 shows structures of caffeic acid and some commonly found caffeic acid derivatives in food materials and vegetables. Caffeic acid and related compounds (for structures, see Fig. 3) are widely distributed in fruits, vegetables, wine, olive oil, teas, coffee beans, honeybee propolis, etc [11]. However, treatment of HIV with effective antiviral drugs is quite expensive. HIV-1 treatment using RTIs and PIs cost about 20,000 USD per patient

232

Fig. 3. Structural features of some common caffeic acid and its derivatives (adapted

from Bailly and Cotelle, 2005) [11].

per year. Bailly and Cotelle [11] recommended that reinforcing the anti-HIV therapy by a convenient diet may limit this cost. The HIV-1 life cycle and the mechanism of action of anti-HIV compounds from medicinal plants Despite the fact that the molecular target(s) of the biological action of several anti-HIV substances, including alkaloids (O-demethyl-buchenavianine, papaverine), polysaccharides (acemannan), lignans (intheriotherins, schisantherin), phenolics (gossypol, lignins, catechol dimers such as peltatols, naphthoquinones such as conocurvone) and saponins (celasdin B, Gleditsia and Gymnocladus saponins), has not been fully elucidated, the molecular targets of several isolated compounds from medicinal plants have been identified [17]. As already pointed out, the HIV-1 life cycle has been deeply investigated and several reviews are available on this particular issue [2–6]. The first step is of course virus–cell fusion and virus adsorption; other critical molecular events

233 are RT and integration. After integration, transcription of HIV-1 is of great importance, as well as TAR–Tat and RRE–Rev interactions. Finally, translation, proteolytic cleavage, glycosylation and assembly/release are biological steps that can be considered as molecular targets of anti-HIV activity. Table 3 reports a partial list of phenolic compounds, divided for their effects on specific HIV-1 life cycle steps. Vlietinck et al. in 1998 reviewed that several compounds of plant origin have been identified that inhibit different stages in the replication cycle of HIV, including (1) Virus adsorption (chromone alkaloids such as, schumannificine; isoquinoline alkaloids such as, michellamines; sulphated polysaccharides and polyphenolics, flavonoids, coumarins such as, glycocoumarin, licopyranocoumarin; phenolics such as, caffeic acid derivatives, galloyl acid derivatives, catechinic acid derivatives; tannins and triterpenes such as, glycyrrhizin and analogues, soya-saponin and analogues); Table 3. Steps of the HIV-1 life cycle affected by molecules isolated from plant extracts. (1) Virus adsorption

(2) Virus-cell fusion (3) Reverse transcription

(4) Integration (6) Proteolytic cleavage (protease inhibition) (7) Glycosylation

(8) Assembly/release

Polyphenolics Flavonoids Phenolics (caffeic acid derivatives, galloyl acid derivatives, catechinic acid derivatives) Tannins Lectins (mannose- and N-acetylglucosamine-specific) Triterpenes (betulinic acid and its analogues) Flavonoids (robustaflavone, hinokiflavone, kaempferol acetylrhamnosides) [44] Biflavonoids [45] Tannins Flavonone-xanthone glucoside (swertifrancheside) [46] Phenolics (curcumin, O-caffeoyl derivatives) [47,48] Flavonoids (quercetin 3-O-(200 -galloyl)-a-Larabinopyranoside) [48] Xanthones (mangostin and analogues) Tannins (camelliatannin H) [29] Epigallocatechin-(4b–48, 2b–4O-7)-epicatechin [49] Alkaloids including indolizidines (castanospermine and analogues), piperidines (1-deoxynojirimicin and analogues) and pyrrolizidines (australine and analogues) Naphthodianthrones (hypericin and pseudohypericin), photosensitisers (terthiophenes and furoisocoumarins), phospholipids

Source: Unless otherwise indicated, information is from Vlietinck et al. (1998) [17].

234 (2) Virus–cell fusion: lectins (mannose- and N-acetylglucosamine-specific) and triterpenes (betulinic acid and its analogues); (3) Reverse transcription: alkaloids (benzophenanthridines, protoberberines, isoquinolines, quinolines), coumarins (calanolides and its analogues), flavonoids, phloroglucinols, lactones (protolichesterinic acid), tannins, iridoids (fulvoplumierin) and triterpenes; (4) Integration: coumarins (3-substituted-4-hydroxycoumarins), depsidones, O-caffeoyl derivatives, lignans (arctigenin and analogues) and phenolics (curcumin); (5) Translation: single chain ribosome inactivating proteins (SCRIPs); (6) Proteolytic cleavage (protease inhibition): saponins (ursolic and maslinic acids), xanthones (mangostin and its analogues) and coumarins; (7) Glycosylation: alkaloids including indolizidines (castanospermine and analogues), piperidines (1-deoxynojirimicin and analogues) and pyrrolizidines (australine and analogues); and (8) Assembly/release: naphthodianthrones (hypericin and pseudohypericin), photosensitisers (terthiophenes and furoisocoumarins) and phospholipids. The target of action of several anti-HIV substances including alkaloids (Odemethyl-buchenavianine, papaverine), polysaccharides (acemannan), lignans (intheriotherins, schisantherin), phenolics (gossypol, lignins, catechol dimers such as peltatols, naphthoquinones such as conocurvone) and saponins (celasdin B, Gleditsia and Gymnocladus saponins), has not been elucidated or does not fit in the proposed scheme. Despite the huge amounts of data that are in the anti-HIV activities of plant extracts and derived compounds, only a very few of these plant-derived anti-HIV products have been used in a limited number of patients suffering from AIDS, for instances, glycyrrhizin, papaverine, trichosanthin, castanospermine, N-butyl-1-deoxynojirimicin, acemannan, etc [17]. Phenolics – a promising class of molecules against HIV The phenolic compounds are secondary metabolic products believed to be produced as a result of the plant’s interaction with the environment [50–54]. The phenolics are derived from phenylalanine and absorb light in the low UV range. Among the natural phenolics, flavonoids are members of a huge sub-class of compounds, most of which are well-known inhibitors of several enzymes that are essential for HIV replication, such as RT [55], viral protease [56] and IN [48]. Figure 4 shows the structures of some of the flavones type phenolic compounds commonly found in plant extracts, with their common styles of substitutions and numbering and Fig. 5 shows biosynthetic pathways.

235

Fig. 4. Structure of some common aglycone flavanone, flavone and flavonol compounds from natural sources.

Very recently, Likhitwitayawuid and co-workers [57] reported a new flavone (30 ,50 -dimethoxy-[200 ,300 ,7,8]-furanoflavone) and three known compounds isolated from the leaves of the Millettia erythrocalyx. They also analyzed the anti-HIV activity against a wild-type HIV-1 (LAI) isolate of oxyresveratrol, which exhibited moderate inhibition against HIV (EC50 ¼ 28.2 mM), showing no toxicity in PBM, CEM and Vero cells at 100 mM. Accordingly, the heartwood of A. lakoocha, which contains a large amount of oxyresveratrol, could be considered as a source of starting material for the development of new natural product-based anti-HIV agents [57]. Reddy et al. [58] isolated and reported a novel bis-andrographolide ether and six known compounds andrographolide, 14-deoxy-11,12-didehydroandrographolide, andrograpanin, 14-deoxyandrographolide, (+/)-5-hydroxy-7, 8-dimethoxyflavanone, and 5-hydroxy-7,8-dimethoxyflavone from the aerial parts of Andrographis paniculata and their anti-HIV potentials.

Some biological activities of phenolics Flavonoids may act as inducers [52] and as phytoalexins [53,54], lowmolecular-weight antimicrobial compounds that are both synthesized and accumulated in plant cells as a defense mechanism after exposure to microorganisms [53,59]. Psoralens (linear furocoumarins) are toxic to insects, especially in the presence of UV light [60], and have been identified as phytoalexins in celery [50]. In addition, phenolics appear to have desirable

236

Fig. 5. Common biosynthetic pathways yielding the phenolic class of compounds.

medicinal properties. Some of these have been reported to be antitumor agents and to exhibit antiviral and antimicrobial activities [61], hypotensive effects [62] and antioxidant properties [63]. Psoralens are used in conjunction with UV light to treat psoriasis and other human skin disorders [64]. Recent evidences suggest that phenolics may play an important role in the regulation of plant metabolism. For example, flavonoids have been shown to be naturally occurring auxin transport regulators [65]. In short, the plant phenolics play a major role in both plant and animal health. Although much basic research still remains to be done, it is possible that many of these compounds, either as isolates or in conjunction with other molecules, may be employed in both agricultural and pharmaceutical roles [66]. A further example is the finan compounds, Kadsulingnanns L–N, were isolated from the seeds of Kadsura coccinea, used traditional medicine in China. Interestingly, Kadsulingnanns M (for structures see Fig. 6) exhibited an anti-HIV activity (IC50 ¼ 119 mM, EC50 ¼ 6.03 mM) [67].

237

Fig. 6. Structure of Kadsulingnanns M (from Liu and Li, 1995) [67].

Fig. 7. The chemical structure of quercetin 3-O-b-D-glucuronide showing anti-HIV activity (from Kashiwada et al., 2005) [68].

Phenolics acting on viral targets Important advances were made in the field of plant-derived anti-HIV agents, leading to the understanding of their mechanism of action. In this chapter, some phenolics with an interesting anti-HIV activity will be discussed according to their viral target(s). Recently, Kashiwada et al. [68] isolated quercetin 3-O-b-D-glucuronide isolated from the leaves of Nelumbo nucifera (Nymphaceae), and demonstrated its anti-HIV activity showing EC50 of 2 mg/ml. The chemical structure of this compound is shown in Fig. 7. In a recent review, Zhang et al. [69] reported the advances in the past 10 years of studies on the phenolic compounds, like coumarins, for their biological activities, giving particular emphasis on their anti-HIV activities. Coumarins have been also reported to possess several biological activities, including antitumor, anti-hypertension, anti-arrhythmia, anti-osteoporosis, assuaging pain, preventing asthma and antisepsis [69]. Zhang et al. [69] have recommended these compounds for further investigations on the improvement of the techniques for extraction and separation, searching the effective

238 precursor compounds and synthesizing and screening coumarin derivatives with high activity and low toxicity. Entry of virus Answer the fact that, flavonoids can inhibit several critical steps of the HIV life cycle[1], virus entry is a very important target. For instance Liu et al. (2005) reported very recently, that several tea polyphenols containing galloyl moiety can inhibit HIV-1 replication with multiple mechanisms of action. They also observed that the theaflavin derivatives had more potent anti-HIV-1 activity than catechin derivatives. These tea polyphenols inhibit HIV-1 entry into target cells by blocking HIV-1 envelope glycoprotein-mediated membrane fusion. The fusion inhibitory activity of the tea polyphenols was correlated with their ability to block the formation of the gp41 six-helix bundle, a fusion-active core conformation. Computeraided molecular docking analyses indicate that these tea polyphenols, theaflavin-3,30 -digallate (TF3), may bind to the highly conserved hydrophobic pocket on the surface of the central trimeric coiled coil formed by the Nterminal heptad repeats of gp41. These results indicate that tea, especially black tea, may be used as a source of anti-HIV agents and theaflavin derivatives may be applied as lead compounds for developing HIV-1 entry inhibitors targeting gp41. The molecular structures of catechin and theaflavin derivatives from green tea and black tea, respectively, are shown in Fig. 8. The inhibitions of p24 production and virus cell fusion data by tea polyphenols are shown in Table 4. Flavanones with an -OH group at position C-30 , such as taxifolin (see Fig. 9), inhibit viral protease, RT, CD4/gp120 interaction by binding to the V3 loop of gp120, and bind to non-specific proteins. Flavanones lacking an OH group at position C-30 , such as aromadendrin (for structures, see Fig. 9), are more specific in their antiviral activity and inhibit CD4/gp120 interaction, but do not inhibit viral protease or RT [71]. Another example of this nonspecific anti-HIV-1 activity was shown for (–)epigallocatechin 3-O-gallate (see structures in Fig. 9), the major tea catechin, which exhibited a destructive effect of virus particles and post-adsorption entry and inhibited viral protease and RT [72]. Yamaguchi et al. [72] reported that epigallocatechin (EGC) impinges on each step of the HIV life cycle. Thus, destruction of the viral particles, viral attachment to cells, post-adsorption entry into cells, RT, viral production from chronically infected cells and the level of expression of viral mRNA, were analyzed using T-lymphoid (H9) and monocytoid (THP-1) cell systems, and antiviral protease activity was measured using a cell-free assay. Inhibitory effects of EGCg on specific binding of the virions to the cellular surfaces and changes in the steady state viral regulation (mRNA expression) due to EGCg were not observed. However, EGCg had a destructive effect on the viral

239

Fig. 8. The molecular structures of catechin and theaflavin derivatives from green tea

and black tea, respectively (from Liu et al., 2005) [70].

particles, and post-adsorption entry and RT in acutely infected monocytoid cells were significantly inhibited at concentrations of EGCg greater than 1 mM, and protease kinetics were suppressed at a concentration higher than 10 mM in the cell-free study. Viral production by THP-1 cells chronically infected with HIV-1 was also inhibited in a dose-dependent manner and the inhibitory effect was enhanced by liposome modification of EGCg. As expected, increased viral mRNA production was observed in lipopolysaccharide

240 Table 4. Inhibition of HIV-1 infections by tea polyphenols. Tannin derivatives

Gallocatechin gallate (GCG) Epigallocatechin gallate (EGCG) Epigallocatechin 3,5-digallate (EGCDG) 20 ,2-Bisepigallocatechin digallate (BGCDG) Theaflavin (TF1) Theaflavin-3-gallate (TF2A) Epitheaflavin-30 -gallate (TF2B) Theaflavin digallate (TF3)

Inhibition of p24 production (IC50, mM7SD)

Inhibition of virus-cell fusion (IC50, mM7SD)

4.6171.28 9.8971.05 2.6370.55 1.5170.46

2.4570.36 3.4471.07 2.4170.48 0.6470.10

5.3370.37 2.5270.23 1.0470.21 1.1570.36

13.0572.09 3.1070.38 1.2870.20 1.9670.08

Source: Modified from Liu et al., 2005 [70].

Fig. 9. Structures of flavanone showing antiviral activity and inhibiting CD4/gp120. (A) For Taxifolin R is OH and for Aromadendrin R is H. (B) Structure of the (-)epigallocatechin 3-O-gallate [71].

(LPS)-activated chronically HIV-1-infected cells. This production was significantly inhibited by EGCg treatment of THP-1 cells. In contrast, production of HIV-1 viral mRNA in unstimulated or LPS-stimulated T-lymphoid cells (H9) was not inhibited by EGCg. Anti-HIV viral activity of EGCg may, thus, result from an interaction with several steps in the HIV-1 life cycle [72]. Recent studies documented that the b-chemokine receptors, CCR2b, CCR3 and CCR5, and the a-chemokine receptors, CXCR1, CXCR2 and CXCR4 serve as entry co-receptors for HIV-1. Although flavonoids and polyphenolic compounds elicit anti-HIV effects such as inhibition of HIV-1 expression and virus replication, the molecular mechanisms underlying these effects remain to be clearly elucidated [73]. Nair and co-workers [73] investigated and reported the effect of flavonoid constituents of a proprietary grape seed extract (GSE) on the expression of

241 HIV-1 co-entry receptors by immuno-competent mononuclear leukocytes. Their results exhibited that GSE significantly down regulated the expression of the HIV-1 entry co-receptors, CCR2b, CCR3 and CCR5 in normal PBMC in a dose-dependent manner. Additionally, GSE-treated cultures exhibited drastically lesser number of CCR3 positive cells as quantitated by flowcytometry analysis that supports RT-PCR gene expression data [73]. Investigations of the mechanisms underlying the anti-HIV-1 effects of GSE may help to identify promising natural products useful in the prevention and/or amelioration of HIV-1 infection [73]. Integration The HIV replication requires integration of viral cDNA into the host genome, a process mediated by the viral enzyme integrase (IN). Figure 10 showing the crystal structure of HIV-1 IN from the Brookhaven Protein Databank (PDB) [74,75]. David et al. [76] isolated and reported a new series of inhibitors of HIV IN thalassiolins A–C (for structures, see Fig. 11), which were isolated from the Caribbean sea grass Thalassia testudinum. The thalassiolins were distinguished from other flavones previously studied by the substitution of a sulfated b-D-glucose at the 7-position, a substituent that imparts increased potency against IN utilizing biochemical assays. The most active of these molecules, thalassiolin A (where R ¼ OH), displays in vitro inhibition of the IN catalyzed strand transfer reaction (IC50 ¼ 0.4 mM) and an antiviral IC50 of 30 mM [76].

Fig. 10. Crystal structure of the catalytic domain of HIV-1 Integrase (IN) (PDB code 1itg) from the Brookhaven Protein Databank [74,75]. The ribbon diagram of the 3D structure of IN was created using Accelrys Discovery Studio VisualizerTM (www.accelrys.com).

242

Fig. 11. Structures of thalassiolins, where as in case of A, B and C, R is –OH, –OCH3

and –H, respectively (modified from David et al., 2002) [76].

Finally David et al. [76] performed molecular modeling (docking with AutoDock ver. 3.0) studies indicate a favorable binding mode is probable at the catalytic core domain of HIV-1 integrase. Computational docking studies can help to generate hypotheses about protein–inhibitor interactions. The estimated binding free energies of A–C compares well with the expected affinities based on the IC50 values. Their docking results suggest that the thalassiolins can be accommodated near the catalytic center of the enzyme IN. The substitution of the 30 -OH (A) with a methoxy group (B) or a hydrogen atom (C) may have a more dramatic effect on the binding energy than observed in the docking simulations, in which the protein is required to be rigid [76]. Tewtrakul et al. [77] isolated and reported two new flavanone glucosides, (2R)- and (2S)-5-O-b-D-glucopyranosyl-7,49-dihydroxy-39,59-dimethoxyflavanone[pervianoside I, peruvianoside II] and a new flavonol glycoside, quercetin 3-O-{b-D-glucopyranosyl-(1-2)-[a-L-rhamnopyranosyl-(1-6)]-bD-galactopyranoside} (peruvianoside III), isolated from the leaves of Thevetia peruviana SCHUM., together with nine known flavonol glycosides and two known iridoid glucosides. Their inhibitory effects against HIV-1 reverse transcriptase and HIV-1 integrase (IN) were investigated. The structures of these compounds are shown in Fig. 12. Table 5 shows the experimentally found IC50 values (mM) of these compounds against HIV-1 IN and HIV-1 RT-associated DNA polymerase (RNA-dependent DNA polymerase (RDDP) and DNA-dependent DNA polymerase (DDDP)). Mazumder et al. [78] have synthesized and tested analogs of curcumin to explore the SARs and mechanism of action against the HIV-1 IN. They observed that two curcumin analogs, dicaffeoylmethane (6) and rosmarinic acid (9), inhibited both activities of integrase with IC50 values below 10 mM. In addition, they demonstrated that two curcumin analogs exhibited equivalent potencies against both IN mutant and wild-type IN, suggesting that the curcumin-binding site and the substrate-binding site may not overlap (see Table 6) [78]. Furthermore, kinetic studies of these analogues suggest that they bind to the enzyme at a slow rate [78].

243

Fig. 12. The structures of the compounds reported by Tewtrakul et al., 2002 [77], found to be active against HIV integrase (IN).

Table 5. HIV-1 RT (RDDP and DDDP) and HIV-1 IN inhibitory activity of compounds isolated from T. peruviana. Compounds

6 7 8 9 10 11 12 14 15 (Quercetin) 16 (Kaempferol) Adriamycin (positive control) Suramin (positive control)

IC50 (mM) RDDP

DDDP

IN

4100 33 4100 20 52 41 75 38 43 4100 27 NA

4100 69 4100 42 4100 4100 4100 4100 4100 4100 6 NA

59 7 30 5 31 45 4100 43 15 40 NA 2.4

Notes: NA, not applicable. Source: Modified from Tewtrakul et al. (2002) [77].

244 Table 6. The SAR (for structures, see Fig. 13) of the curcumin and its analogues against the HIV-1 IN. Compounds

2 4 5 (Curcumin) 6 7 9 (Rosmarinic acid)

Substitutions

IC50 (mM)

R1

R2

R3

R4

–H –H –OCH3 –OH –OCH3

–OH –OH –OH –OH –OH

–H –OCH3 –OCH3 –OH –OH

–OH –OH –OH –OH –OH

3-processing Strand transfer 120 140 150 6.0 18 9.0

80 120 140 3.1 9.0 4.0

Source: Modified from Mazumder et al. (1997) [78].

Fig. 13. Structures of the Curcumin and its analogues exhibiting potentials against

the HIV-1 IN (modified from Mazumder et al., 1997) [78].

Figure 13 shows the molecular structures of the curcumin and its analogues having potent inhibitory activities against the mutant and wild-type HIV-1 IN and Table 6 shows the SAR of these molecules for the inhibition of the 30 -processing and strand transfer activities of HIV-1 IN. In enzymatic assays, the dicaffeoylquinic acids (DCQAs), such as 3,5dicaffeoylquinic acid, and dicaffeoyltartaric acids (DCTAs), such as l-chicoric acid (for structures, see Fig. 14), demonstrated a 10- to 100-fold higher preference for inhibition of HIV integrase than of HIV RT [79]. SAR studies on these synthesized compounds demonstrated that l-chicoric acid and Dchicoric acid exhibited similar anti-HIV-1 integrase activity, and removal of one or both of its carboxylic groups did not result in a significantly lower IN inhibitory activity [80].

245

Fig. 14. Structures of caffeol carboxylic acid derivatives (modified from McDougall et al., 1998 and Lin et al., 1999) [79,80].

Reverse transcription Reverse transcriptase (RT), an enzyme of human HIV, has been demonstrated to be important for the viral replication. The crucial role of RT in the early stages of the HIV life cycle has made it one of the most reliable targets for potential anti-AIDS chemotherapy [81]. Figure 15 shows the 3D structure of the HIV-1 RT (PDB code 1tkt) complexed with an inhibitor GW426318 (shown in CPK model), from the Brookhaven PDB [75, 82]. Large number of structurally diverse natural coumarins was found to display potent anti-HIV activity and constant advancement is expected in the discovery of new leads and in the development of these agents as potential anti-AIDS drug candidates. Some current reports show that naturally occurring coumarin and their semi-synthetic or synthetic analogues serve as potent non-nucleoside RT-inhibitors, another as inhibitors of HIV-integrase or HIV-protease [83]. Kitamura and co-workers [55] in 1998 discovered flavonoid baicalin as (for structure see Fig. 16) a potent inhibitor of HIV-1 through its important enzyme RT, which is important for viral life cycle in transcription. They found that baicalin markedly inhibited the replication of HIV-1 in a conc.dependent manner in normal peripheral blood mononuclear cells (PBMC) stimulated with phytohemagglutinin (PHA) in vitro. The effect was found to be more pronounced when the cells were pretreated with baicalin. Furthermore, baicalin inhibited the HIV-1 replication in PHA-stimulated PBMC from asymptomatic HIV-1-seropositive carriers. The IC50 for the HIV-1 replication was approximately 0.5 mg/ml. At the conc. of 2 mg/ml of baicalin,

246

Fig. 15. Crystal structure of the HIV-1 RT (PDB code 1tkt, in ribbon) complexed with an inhibitor GW426318 (shown in CPK model), from the Brookhaven Protein Databank [75,82]. The 3D structure of RT was created using Accelrys Discovery Studio VisualizerTM (www.accelrys.com).

Fig. 16. Structure of 5,6,7-trihydroxyflavone-7-O-b-D-glucopyranosideuronic acid

monohydrate (from Kitamura et al., 1998) [55].

copy numbers of HIV-1 proviral DNA were approximately 50 times less than in untreated controls. In a cell-free infection system, baicalin inhibited the activity of HIV-1 RT, but not the activity of human DNA polymerases a and w (DNA polymerase b was also slightly inhibited). Kitamura et al. [55] suggested that the anti-HIV-1 effect of baicalin may at least partly be due to the inhibition of HIV-1 RT. Ahn et al. [84] reported four phlorotannin derivatives, eckol (1), 8, 80 -bieckol (2), 8,4000 -dieckol (3), and phlorofucofuroeckol A (4) by the bioassay-directed isolation from a marine brown alga Ecklonia cava. Among these, 2 and 3 exhibited an inhibitory effect on HIV-1 RT and protease. Specifically, they inhibited RT more potently than protease. The inhibitory activity of compound 2 (IC50, 0.51 mM) against HIV-1 RT was comparable to that of nevirapine (IC50, 0.28 mM), used as a reference

247 compound. From the enzyme kinetic assay, it was found that compound 2 inhibited the RNA-dependent DNA synthesis activity of HIV-1 RT noncompetitively against dUTP/dTTP with a Ki of 0.78 mM. With respect to the homopolymeric template/primer, (rA)n(dT)15, 2 displayed an uncompetitive type of inhibition (Ki, 0.23 mM) [84]. Figure 17 shows the molecular structures of the compounds 2 and 3. Table 7 shows the specific activities of compounds 2 and 3 against the HIV-1 RT and protease. Other inhibitory compounds of HIV-1 RT are the dibenzylbutadiene lignans, anolignan A and anolignan B, isolated from Anogeissus acuminate. The activities of these molecules was obtained followed by bioassay-guided fractionation [85]. For the structural features of anolignan A and anolignan B, see Fig. 18. Min and co-workers isolated and reported three new kaempferol glycosides, crassirhizomosides A (1), B (2) and C (3), from the rhizome of Dryopteris crassirhizoma (Family: Aspidiaceae), together with a known kaempferol glycoside, sutchuenoside A (4) [81]. These kaempferol glycosides were characterized by traditional chemical and spectroscopic approaches. Min et al. [81] also reported the inhibitory activities of the 1–4 (see Fig. 19 for molecular structures) including kaempferol on HIV RT-associated DNA polymerase (RNA-dependent DNA polymerase (RDDP) and DNA-dependent DNA polymerase (DDDP)) and RNase H. Table 8 shows the inhibition and SAR of the kaempferol and kaempferol glycosides against HIV RT and its related RDDP and DDDP. Very recently, Li et al. [86] isolated and reported a new pinoresinol-type lignan, 9a-angloyloxypinoresinol, from the roots and rhizomes of Ligularia kanaitizensis, in addition to a known compound, 9a-hydroxypinoresinol. This new compound exhibited inhibition against the HIV-1 RT. Suppression of HIV promoter activity Uchiumi et al. [87] have established a sensitive assay system for screening HIV promoter-suppressor molecules using the pHIVLuc (a luciferase (Luc) reporter plasmid carrying the HIV promoter region) reporter plasmid. The inhibitory effect of coumarin and chalcone skeleton-containing natural products has been revealed to be greater than that of ellagitannins and on the other hand, naphthalene- or anthraquinone-related molecules did not exhibit suppression of TPA-induced HIV promoter activities. Previously, tannins and lignins were reported to be potent inhibitors of HIV expression [88] and adsorption to host cells [89]. Following this strategy, tannic acid was demonstrated to have inhibitory effects on 12-O-tetra-decanoyl phorbol-13-acetate (TPA)-induced HIV promoter. They found a putative tannic acid-responsive element between the positions –133 and –104 of the HIV promoter [90].

248

Fig. 17. Molecular structures of two potent HIV-1 RT inhibitory phlorotannins iso-

lated and reported by Ahn et al. (2004) from a marine brown alga Ecklonia cava (modified from Ahn et al. 2004) [84].

In addition, Uchiumi et al. [87] found that hirtellin A, hirtellin B, hirtellin C, tamarixinin B and cocciferin D1, which are classified as ellagitannin dimmers (from structures see Fig. 20), are very similar to tannic acid in inhibition of HIV-1 promoter activity.

249 Table 7. Inhibitory potentials of the compounds 2 and 3 (structures are in Fig. 17) isolated from the marine brown alga Ecklonia cava and reported to have potent HIV-1 RT activities. IC50 (mM)

Compounds

2 3 Nevirapinea Acetyl pepstatinb

RT

Protease

0.51 5.31 0.28 –

81.5 36.9 – 0.34

Source: Modified from Ahn et al. (2004) [84]. a Specific reference inhibitor for RT. b Specific reference inhibitor for protease.

Fig. 18. Structural features of Anolignan A and B [85].

Fig. 19. Structures of kaempferol acetylrhamnosides isolated from the rhizome of

Dryopteris crassirhizoma showed potent inhibitory activities against the human HIV1 RT (modified from Min et al., 2001) [81].

250 Table 8. The inhibition and SAR of the kaempferol and kaempferol glycosides against HIV RT and its related RDDP and DDDP isolated from the rhizome of Dryopteris crassirhizoma. IC50 (mM)

Compounds

1 2 3 4 Kaempferol Adriamycin

RDDP

DDDP

215 4500 240 405 110 46

25 4100 28 23 75 6

Note: Adriamycin was used as reference inhibitor for HIV-1 RT-related DNA polymerase. Source: Modified from Min et al. (2001) [81].

Fig. 20. Ellagitannins reported by Uchiumi and co-workers (2003), which suppressed the TPA-induced HIV promoter activity (adapted from Uchiumi et al., 2003) [87].

251 Furthermore, Uchiumi et al. [87] presented data showed that derivatives of 3-phenylcoumarin (for structures, see Fig. 21) and chalcone (for structures, see Fig. 22) suppressed TPA-induced HIV promoter activity more effectively than tannic acid. The glycyrrhisoflavone, glycycoumarin and licopyranocoumarin (for structures, see Fig. 21), which include isoflavone and 3-phenylcoumarin skeletons, also suppressed the HIV promoter more than ellagitannins [83,87].

Fig. 21. Coumarin derivatives suppressed also the TPA-induced HIV promoter ac-

tivity more effectively than tannic acid (modified from Uchiumi et al., 2003) [87].

Fig. 22. Chalcone derivatives suppressed also the TPA-induced HIV promoter ac-

tivity more effectively than tannic acid (modified from Uchiumi et al., 2003) [87].

252 Tetrahydroxymethoxychalcone, licochalcones A and B (for structures, see Fig. 22), having the chalcone skeleton, potently reduce the HIV promoter activity, as well [87]. Although, aspidin AB, BB (for structures, see Fig. 23) and chromene glucoside have hexagonal structures, but in the studies of Uchiumi et al. [87], they did not show any suppressive activities. This is possibly due to the absence of circular chain composed of phenols and saccharides. Since these circular structures, which are characteristic of ellagitannins, have been suggested to be responsible for the inhibition of poly(ADP-ribose) glycohydrolase (PARG) [91], the suppressive effects on the HIV promoter by these compounds suggest that poly(ADP-ribose)n degradation by PARG may be involved in the onset of HIV promoter activity [87]. Although liquiritigenin, liquiritin, cassiaoccidentalin A and cassiaoccidentalin C (see Fig. 23) have skeletons related to chalcones, but these compounds did not exhibited obvious suppressive activities against the HIV promoter [87]. The torosachrysone gentibioside and rubrofusarin gentiobioside (for structures, see Fig. 23), chrysophanol, emodin and aloe-emodin that have a naphthalene or anthraquinone (for structures, see Fig. 24) skeleton classified as acetogenins, show no suppressive effect; instead, they somewhat induce TPA-induced HIV promoter activities [87]. From the studies of Uchiumi et al. [87], it could be concluded that derivatives of coumarin and chalcone may be used as anti-HIV drugs, which can interfere with the viral gene expression regulatory mechanism.

Chemogenomics in anti-HIV drug discovery From very recent definitions, chemogenomics, an advanced and highthroughput method, which directs the drug discovery process based on the respective gene families, has been developed. By integrating all the available information within a protein family (sequence, structure–function relationship data, protein structure, etc.), chemogenomics can efficiently enable cross-SAR exploitation, that express early compound selection and discovery of best selectivity panel members [92]. In recent years numerous significant discoveries and advancements have been done on chemogenomics, which results large number of very informative review papers on these technologies, those are explaining different approaches utilized in chemogenomics, their appliances in drug discovery, etc., counting several real-life examples captivating quite a lot of biomolecular targets [92–125]. The spectacular amplification in the amount of data from protein structural biology has been obscured by the publicity surrounding the Human Genome Project [126]. There are more than 40,000 3D structures of biological molecules, of which more than 31,000 are of proteins and are now

253

Fig. 23. Flavones and naphthalenes suppressed also the TPA-induced HIV promoter activity more effectively than tannic acid (modified from Uchiumi et al., 2003) [87].

publicly available, especially through PDB (http://www.rcsb.org/pdb/) [75,127], and the trend of these number is increasing every year. This is due to the flow of interest in the public and private sectors to actively obtain representative structures for novel proteins. The combination of this huge

254

Fig. 24. Anthraquinones suppressed also the TPA-induced HIV promoter activity

more effectively than tannic acid (modified from Uchiumi et al., 2003) [87].

source of raw data and refined homology modeling (like, ModellerTM, etc.) tools enables the structures of a huge number of pharmaceutically significant protein targets to be predicted as well as the shapes and physical properties of potential ligand-binding sites [126]. The ability to map genomic data onto protein structures provides the framework linking three billion ATGC codes to drug-design chemistry [126]. Similar approaches can be and should be implemented toward the selection and discovery processes for the anti-HIV drug discovery pathway, as the amount of compounds screened and to be screen is huge and it is obvious that this sort of approach shall speed-up further in the anti-HIV drug discovery steps. The chemogenomic approaches also can be utilized to study the so-called ‘‘side effects’’ of some particular molecules of interest. This can significantly enhance the efficiency of the whole discovery process, by giving information to the early discovery of molecules having certain adverse effects or effects having on other non-targeted organs. Certainly, most of the members of the same family of proteins often exhibit similar biochemical and pharmacological characteristics and can share important practical aspects, such as in vitro assay conditions. Similarly, identifying and recycling such similarities in early-stage drug discovery can have obvious benefits in terms of efficiency [92]. These kinds of approach also can be useful to find more selective molecules against target of interest. Interestingly, a number of biomolecular targets of the HIV-1 virus are already publicly available largely, for example, in PDB (like HIV-1 protease, reverse transcriptase, integrase, some complexes with related DNA, etc.). Moreover within the PDB, approximately 760 structural data of HIV-1 related molecules or complexes are publicly available. Even some of the potential anti-HIV molecules have been complexed with the target molecules (e.g., targeted protein or DNA) and co-crystallized or have been coNMRed, which are also available from this sort of public databases, those can be used as ‘‘template’’ for the discovery of newer and more potential molecules against HIV. These large amounts of data could be easily freely downloadable and usable for further drug discovery processes for the development of novel anti-HIV molecules even from the natural sources.

255 Conclusions, perspectives and recommendations Although HIV infection is now primarily treated with RT and PR inhibitors, anti-HIV therapy must look toward and find new drugs with novel mechanism(s) of action to both improve efficacy and address the growing problem of drug resistance [128]. Since 1996, HAART was designed to rapidly control HIV replication. It has had a significant impact on patient health and progression of AIDS in developed countries, but its success has not been complete. HAART strategy still suffers from issues of patient compliance, cost, deleterious side effects and emerging drug resistance. Therefore, expansion of novel anti-HIV drugs and targets will be critical in the forthcoming years. In this context, discovering anti-HIV agents from natural sources and particularly from plants, may highlight the principle of a nutritional antioxidant antiretroviral diet [8,11]. In considerations of the fact that the majority of the HIV-infected patients are in developing countries, economically attractive additional anti-HIV effects of folk medicines are of great relevance and should be studied and relative folk medicines be identified and employed [11]. In addition, it should be concluded that conventional antiretroviral drugs such as AZT, ddI and ddC and recently discovered HAART regimens containing protease inhibitors might have serious side effects [129]. Unfortunately, only a few natural products from plants have received scrutiny for their potential anti-HIV effects [130,131]. Nevertheless, the discovery of very specific and less toxic antiviral agents is a current high priority in the search for more effective therapies against HIV [132]. In this respect, the evaluation of the mechanism of the anti-HIV effects of polyphenols, e.g., flavonoids, etc., may lead to the identification of new antiHIV agents useful in the treatment of patients with AIDS. The scientific neighborhood is searching enthusiastically for novel drugs and permutations for the treatment of HIV infection effective for first-line treatment, as well as against drug-resistant mutants [83]. A further comment should be done on the fact that, the HIV treatment antiviral drugs are quite expensive. Therefore, if anti-HIV treatment can be supported with a convenient diet, this might limit the cost for the whole treatment process. Here, we have reviewed that large number of promising anti-HIV molecules are present as the dietary supplements in different foods, vegetable, etc. Especially tea contains huge number of tannins and catechins, which are already proved to be effective against HIV in different stages of viral life cycles. These can be good and inexpensive sources of antiviral agents for the HIV-infected patients. Polyphenolics also have been proved to be very good antioxidants; so the patients could get rid of also from the oxidative stresses by reducing even the reactive oxygen species (ROS) and other oxidizing molecules. The mode of action of the plant derived molecules should also be studied besides, may be by utilizing chemo-genomic approaches [133].

256 Need for new anti-HIV (or AIDS) drugs are a global burning issue. In addition to evident financial and commercial impediments, HIV/AIDS patients face with diverse impenetrabilities connected with the presently official anti-HIV drugs. Adverse effects, the appearance of drug resistance [134] and the narrow spectrum of activity have restricted the therapeutic convenience of the different RT and protease inhibitors that are currently accessible on the market [26]. Large numbers of molecules until now reported to possess moderate to potent inhibitory profiles against HIV-1. Now it is time to prove their clinical efficacy and their specificity. Before going to clinical trials it is apparent to optimize the targeted molecule(s) for their safety profiles. So there are questions of in vitro, acute and chronic in vivo toxicity studies. As most of the potent inhibitors are synthetic in origin, so the amounts of the particular molecules are not the big issues. There are some reports published about the in vivo studies [135–142] of some potential molecules on different mammalian models, even some small trials on healthy subjects [143], but these data are not sufficient to go for clinical trials. There are needs for longterm chronic toxicity studies on higher animals and also genotoxic studies are extremely important also for the long-term treatments. Still there is a great need of developing new and chemo-diverse molecules against HIV-1 for effective and cheaper source of treatments for AIDS patients. Unmodified and/or purified traditional medicines, like Ayurveda, Unani, Chakma medicine, etc. [9], and ethnopharmacology might play a major role for the treatment of AIDS in cheaper way [144,145]. In the near future, these traditional medicines could be great source of anti-HIV treatment. It is very much anticipated that anti-HIV molecules as drugs and prophylactic preparations including the natural products or their analogs would soon be commercially available for regular clinical uses. The outcome and understandings with many of the anti-HIV natural products will encourage and stimulate even more scientists to look for new leads from natural resources [26]. Acknowledgments One of us (MTHK) is thankful for the award (#1056) from the UNESCOMCBN. References 1.

2. 3.

Cos P, Maes L, Berghe DV, Hermans N, Pieters L and Vlietinck A. Plant substances as anti-HIV agents selected according to their putative mechanism of action. J Nat Product 2004;67:284–293. Wang WK, Chen MY, Chuang CY, Jeang KT and Huang LM. Molecular biology of human immunodeficiency virus type 1. J Microbiol Immunol Infect 2000;33(3):131–140. Pani A, Loi AG, Mura M, Marceddu T, La Colla P and Marongiu ME. Targeting HIV: old and new players. Curr Drug Targets Infect Disord 2002;2(1):17–32.

257 4. 5. 6. 7.

8. 9. 10. 11.

12. 13.

14. 15.

16. 17. 18. 19. 20.

21.

22. 23. 24.

Bukrinskaya AG, HIV-1 assembly and maturation. Arch Virol 2004;149(6): 1067–1082. Bannwarth S and Gatignol A. HIV-1 TAR RNA: the target of molecular interactions between the virus and its host. Curr HIV Res 2005;3(1):61–71. Nielsen MH, Pedersen FS and Kjems J. Molecular strategies to inhibit HIV-1 replication. Retro Virology 2005;2(1):10. Manfredi R and Sabbatani S. A novel antiretroviral class (fusion inhibitors) in the management of HIV infection. Present features and future perspectives of enfuvirtide (T-20). Curr Med Chem 2006;13(20):2369–2384. Yu D, Morris-Natschke SL and Lee KH. New developments in natural products-based anti-AIDS research. Med Res Rev 2007;27(1):108–132. Khan MT, Ather A, Thompson KD and Gambari R. Extracts and molecules from medicinal plants against herpes simplex viruses. Antiviral Res 2005;67(2):107–119. Pomerantz RJ and Horn DL. Twenty years of therapy for HIV-1 infection. Nat Med 2003;9(7):867–873. Bailly F, Cotelle P, Anti-HIV activities of natural antioxidant caffeic acid derivatives: toward an antiviral supplementation diet. Curr Med Chem 2005;12(15): 1811–1818, and references cited therein. Barbaro G. HIV-associated cardiomyopathy etiopathogenesis and clinical aspects. Herz 2005;30(6):486–492. Barbaro G. Reviewing the cardiovascular complications of HIV infection after the introduction of highly active antiretroviral therapy. Curr Drug Targets Cardiovasc Haematol Disord 2005;5(4):337–343. Barbaro G, Lucchini A and Barbarini G. Highly active antiretroviral therapy in HIVassociated pulmonary hypertension. Minerva Cardioangiol 2005;53(2):153–154. Barbaro G, Scozzafava A, Mastrolorenzo A and Supuran CT. Highly active antiretroviral therapy: current state of the art, new agents and their pharmacological interactions useful for improving therapeutic outcome. Curr Pharm Des 2005;11(14):1805–1843. Atta ur R, Harvey K and Siddiqui RA. Interleukin-8: An autocrine inflammatory mediator. Curr Pharm Des 1999;5(4):241–253. Vlietinck AJ, Bruyne TD, Apers S and Pieters LA. Planta Med 1998;64:97. Matthee G, Wright AD and Konig GM. HIV reverse transcriptase inhibitors of natural origin. Planta Med 1999;65(6):493–506. Bedoya LM, Sanchez-Palomino S, Abad MJ, Bermejo P and Alcami J. Anti-HIV activity of medicinal plant extracts. J Ethnopharmacol 2001;77(1):113–116. Ng TB, Huang B, Fong WP and Yeung HW. Anti-human immunodeficiency virus (anti-HIV) natural products with special emphasis on HIV reverse transcriptase inhibitors. Life Sci 1997;61(10):933–949. Schinazi RF, Chu CK, Peck A, McMillan A, Mathis R, Cannon D, Jeong LS, Beach JW, Choi WB and Yeola S. Activities of the four optical isomers of 20 ,30 -dideoxy-30 thiacytidine (BCH-189) against human immunodeficiency virus type 1 in human lymphocytes. Antimicrob Agents Chemother 1992;36(3):672–676. Schinazi RF, Mead JR and Feorino PM. Insights into HIV chemotherapy. AIDS Res Hum Retroviruses 1992;8(6):963–990. Schinazi RF, In: Chu, C. K. C., H. G. Ed.; Natural Products As Antiviral Agents; Plenum Press: New York, 1992, pp.1–30. Jung M, Lee S and Kim H. Recent studies on natural products as anti-HIV agents. Curr Med Chem 2000;7(6):649–661.

258 25. Asres K and Bucar F. Anti-HIV activity against immunodeficiency virus type 1 (HIV-I) and type II (HIV-II) of compounds isolated from the stem bark of Combretum molle. Ethiop Med J 2005;43(1):15–20. 26. Asres K, Seyoum A, Veeresham C, Bucar F and Gibbons S. Naturally derived anti-HIV agents. Phytother Res 2005;19(7):557–581. 27. Kanyara JN and Njagi EN. Anti-HIV-1 activities in extracts from some medicinal plants as assessed in an in vitro biochemical HIV-1 reverse transcriptase assay. Phytother Res 2005;19(4):287–290. 28. el-Mekkawy S, Meselhy MR, Kusumoto IT, Kadota S, Hattori M and Namba T. Inhibitory effects of Egyptian folk medicines on human immunodeficiency virus (HIV) reverse transcriptase. Chem Pharm Bull 1995;43:641–648. 29. Park JC, Hur JM, Park JG, Hatano T, Yoshida T, Miyashiro H, Min BS and Hattori M. Inhibitory effects of Korean medicinal plants and camelliatannin H from Camellia japonica on human immunodeficiency virus type 1 protease. Phytotherapy Res 2002;16:422–426. 30. Park JC, Kim SC, Choi MR, Song SH, Yoo EJ, Kim SH, Miyashiro H and Hattori M. Anti-HIV protease activity from rosa family plant extracts and rosamultin from Rosa rugosa. J Med Food 2005;8(1):107–109. 31. Bessong PO, Obi CL, Andreola ML, Rojas LB, Pouysegu L, Igumbor E, Meyer JJ, Quideau S and Litvak S. Evaluation of selected South African medicinal plants for inhibitory properties against human immunodeficiency virus type 1 reverse transcriptase and integrase. J Ethnopharmacol 2005;99(1):83–91. 32. Paskaleva EE, Lin X, Li W, Cotter R, Klein MT, Roberge E, Yu EK, Clark B, Veille JC, Liu Y, Lee DY and Canki M. Inhibition of highly productive HIV-1 infection in T cells, primary human macrophages, microglia, and astrocytes by Sargassum fusiforme. AIDS Res Ther 2006;3:15. 33. Sangma C, Chuakheaw D, Jongkon N, Saenbandit K, Nunrium P, Uthayopas P and Hannongbua S. Virtual screening for anti-HIV-1 RT and anti-HIV-1 PR inhibitors from the Thai medicinal plants database: a combined docking with neural networks approach. Comb Chem High Throughput Screening 2005;8(5):417–429. 34. Alves CN, Pinheiro JC, Camargo AJ, Ferreira MMC, Romero RAF and da Silva ABF. A multiple linear regression and partial least squares study of flavonoid compounds with anti-HIV activity. J Mol Struct: THEOCHEM 2001;541(1–3):81–88. 35. Sangma C, Chuakheaw D, Jongkon N, Saenbandit K, Nunrium P, Uthayopas P and Hannongbua S. Virtual screening for anti-HIV-1 RT and anti-HIV-1 PR inhibitors from the Thai medicinal plants database: a combined docking with neural networks approach. Comb Chem High Throughput Screen 2005;8(5):417–429. 36. Sechi M, Sannia L, Carta F, Palomba M, Dallocchio R, Dessi A, Derudas M, Zawahir Z and Neamati N. Design of novel bioisosteres of beta-diketo acid inhibitors of HIV-1 integrase. Antivir Chem Chemother 2005;16(1):41–61. 37. Debnath AK. Generation of predictive pharmacophore models for CCR5 antagonists: study with piperidine- and piperazine-based compounds as a new class of HIV-1 entry inhibitors. J Med Chem 2003;46(21):4501–4515. 38. Guichou JF, Viaud J, Mettling C, Subra G, Lin YL and Chavanieu A. Structure-based design, synthesis, and biological evaluation of novel inhibitors of human cyclophilin A. J Med Chem 2006;49(3):900–910. 39. Kumar V, Sardana S and Madan AK. Predicting anti-HIV activity of 2,3-diaryl-1,3thiazolidin-4-ones: computational approach using reformed eccentric connectivity index. J Mol Model (Online) 2004;10(5–6):399–407.

259 40. Renner S, Ludwig V, Boden O, Scheffer U, Gobel M and Schneider G. New inhibitors of the Tat-TAR RNA interaction found with a ‘‘fuzzy’’ pharmacophore model. Chembiochem 2005;6(6):1119–1125. 41. Varnek A and Solov’ev VP. ‘‘In silico’’ design of potential anti-HIV actives using fragment descriptors. Comb Chem High Throughput Screen 2005;8(5):403–416. 42. Vilar S, Estrada E, Uriarte E, ;Santana L and Gutierrez Y. In silico studies toward the discovery of new anti-HIV nucleoside compounds through the use of TOPS-MODE and 2D/3D connectivity indices. 2. Purine derivatives. J Chem Inf Model 2005;45(2):502–514. 43. von Grotthuss M, Koczyk G, Pas J, Wyrwicz LS and Rychlewski L. Ligand.Info smallmolecule meta-database. Comb Chem High Throughput Screen 2004;7(8):757–761. 44. Min BS, Tomiyama M, Ma CM, Nakamura N and Hattori M. Kaempferol acetylrhamnosides from the rhizome of Dryopteris crassirhizoma and their inhibitory effects on three different activities of human immunodeficiency virus-1 reverse transcriptase. Chem Pharm Bull (Tokyo) 2001;49(5):546–550. 45. Lin YM, Anderson H, Flavin MT, Pai YH, Mata-Greenwood E, Pengsuparp T, Pezzuto JM, Schinazi RF, Hughes SH and Chen FC. In vitro anti-HIV activity of biflavonoids isolated from Rhus succedanea and Garcinia multiflora. J Nat Prod 1997;60(9):884–888. 46. Pengsuparp T, Cai L, Constant H, Fong HH, Lin LZ, Kinghorn AD, Pezzuto JM, Cordell GA, Ingolfsdottir K, Wagner H and Hughes SH. Mechanistic evaluation of new plant-derived compounds that inhibit HIV-1 reverse transcriptase. J Nat Prod 1995;58(7):1024–1031. 47. Kwon HC, Jung CM, Shin CG, Lee JK, Choi SU, Kim SY and Lee KR. A new caffeoyl quinic acid from aster scaber and its inhibitory activity against human immunodeficiency virus-1 (HIV-1) integrase. Chem Pharm Bull (Tokyo) 2000;48(11):1796–1798. 48. Kim HJ, Woo ER, Shin CG and Park H. A new flavonol glycoside gallate ester from Acer okamotoanum and its inhibitory activity against human immunodeficiency virus-1 (HIV-1) integrase. J Nat Prod 1998;61:145–148. 49. Ma C, Nakamura N, Hattori M, Kakuda H, Qiao J and Yu H. Inhibitory effects on HIV-1 protease of constituents from the wood of Xanthoceras sorbifolia. J Nat Prod 2000;63(2):238–242. 50. Beier RC and Oertli EH. Psoralen and other linear furocoumarins as phytoalexins in celery. Phytochemistry 1993;22:2595–2597. 51. Afek U, Sztejnberg A and Carmely S. 6,7–Dimethoxycoumarin, a citrus phytoalexin conferring resistance against Phytophthora gummosis. Phytochemistry 1986;25:1855–1856. 52. Zaat SAJ, van Brussel AAN and Tak T. Flavonoids induce Rhizobium leguminosarum to produce nodDABC gene-related factors that cause thick, short roots and root hair responses on common vetch. J Bacteriol 1987;169:3388–3391. 53. Laks PE and Pruner MS. Flavonoid biocides: structure/activity relations of flavonoid phytoalexin analogues. Phytochemistry 1989;28:87–91. 54. Snyder BA and Nicholson RL. Synthesis of phytoalexins in sorghum as a site-specific response to fungal ingress. Science 1990;248:1637–1639. 55. Kitamura K, Honda M, Yoshizaki H, Yamamoto S, Nakane H, Fukushima M, Ono K and Tokunaga T. Baicalin, an inhibitor of HIV-1 production in vitro. Antiviral Res 1998;37:131–140. 56. Xu HX, Wan M, Dong H, But PPH and Foo LY. Inhibitory activity of flavonoids and tannins against HIV-1 protease. Biol Pharm Bull 2000;23:1072–1076. 57. Likhitwitayawuid K, Sritularak B, Benchanak K, Lipipun V, Mathew J and Schinazi RF. Phenolics with antiviral activity from Millettia erythrocalyx and Artocarpus lakoocha. Nat Prod Res 2005;19(2):177–182.

260 58. Reddy VL, Reddy SM, Ravikanth V, Krishnaiah P, Goud TV, Rao TP, Ram TS, Gonnade RG, Bhadbhade M and Venkateswarlu Y. A new bis-andrographolide ether from Andrographis paniculata nees and evaluation of anti-HIV activity. Nat Prod Res 2005;19(3):223–230. 59. Dixon RA. The phytoalexin response: elicitation signaling and control of host gene expression. Bioll Rev 1986;61:239–291. 60. Nahrstedt A, The significance of secondary metabolites for interactions between plants and insects, Planta Med 1990;55:333–338, and references cited therein. 61. Robbins R. Medical and nutritional aspects of citrus bioflavonoids. In: Citrus Nutrition and Quality, Nagy S and Attaway J (eds), Washington, DC, American Chemistry Society, 1980, pp. 43–59. 62. Matsubara Y, Kumamoto H and Iizuka Y. Structure and hypotensive effect of flavonoid glycosides in Citrus unshiu peelings. Agric Biol Chem 1985;49:909–914. 63. Robak J and Gryglewski RJ. Flavonoids are scavengers of superoxide anions. Biochem Pharmacol 1988;37:837–841. 64. Stolk LML and Siddiqui AH. Minireview: biopharmaceutics, pharmacokinetics and pharmacology of psoralens. Gen Pharmacol 1988;19:649–653. 65. Jacobs M and Rubery PH. Naturally occurring auxin transport regulators. Science 1988;241:346–349. 66. Anonymous, United States Department of Agriculture. Agricultural Research Service website (http://www.ars.usda.gov/is/np/phenolics/intro.htm). 1999. 67. Liu J-S and Li L. Kadsulignans L-N, three dibenzocyclooctadiene lignans from Kadsura coccinea. Phytochemistry 1995;38(1):241–245. 68. Kashiwada Y, Aoshima A, Ikeshiro Y, Chen YP, Furukawa H, Itoigawa M, Fujioka T, Mihashi K, Cosentino LM, Morris-Natschke SL and Lee KH. Anti-HIV benzylisoquinoline alkaloids and flavonoids from the leaves of Nelumbo nucifera, and structure-activity correlations with related alkaloids. Bioorg Med Chem 2005;13(2):443–448. 69. Zhang SY, Meng L, Gao WY, Song NN, Jia W and Duan HQ. Advances on biological activities of coumarins. Zhongguo Zhong Yao Za Zhi 2005;30(6):410–414. 70. Liu S, Lu H, Zhao Q, He Y, Niu J, Debnath AK, Wu S and Jiang S. Theaflavin derivatives in black tea and catechin derivatives in green tea inhibit HIV-1 entry by targeting gp41. Biochim Biophys Acta 2005;1723(1–3):270–281. 71. Mahmood N, Piacente S, Burke A, Khan A and Pizza C. Antiviral Chem Chemother 1997;8:70–74. 72. Yamaguchi K, Honda M, Ikigai H, Hara Y and Shimamura T. Inhibitory effects of (-)epigallocatechin gallate on the life cycle of human immunodeficiency virus type 1 (HIV1). Antiviral Research 2002;53:19–34. 73. Nair PM, Kandaswami C, Mahajan S, Nair HN, Chawda R, Shanahan T and Schwartz SA, Grape seed extract proanthocyanidins downregulate HIV-one entry coreceptors, CCRtwob, CCRthree and CCRfive gene expression by normal peripheral blood mononuclear cells. Biol Res 2002;35:421–431. 74. Dyda F, Hickman AB, Jenkins TM, Engelman A, Craigie R and Davies DR. Crystal structure of the catalytic domain of HIV-1 integrase: similarity to other polynucleotidyl transferases. Science 1994;266(5193):1981–1986. 75. Berman HM, Westbrook J, Feng Z, Gilliland G, Bhat TN, Weissig H, Shindyalov IN and Bourne PE. Nucleic Acids Res 2000;28(1):235–242. 76. David C, Rowley M, Hansen ST, Rhodes D, Christoph A, Sotriffer H, Ni J, Andrew McCammon FD and Bushman WF. Thalassiolins A–C: new marine-derived inhibitors of HIV cDNA integrase. Bioorg Med Chem 2002;10:3619–3625.

261 77. Tewtrakul S, Nakamura N, Hattori M, Fujiwara T and Supavita T. Flavanone and flavonol glycosides from the leaves of Thevetia peruviana and their HIV-1 reverse transcriptase and HIV-1 integrase inhibitory Activities. Chem Pharm Bull 2002;50(5):630–635. 78. Mazumder A, Neamati N, Sunder S, Schulz J, Pertz H, Eich E and Pommier Y. Curcumin analogs with altered potencies against HIV-1 integrase as probes for biochemical mechanisms of drug action. J Med Chem 1997;40:3057–3063. 79. McDougall B, King PJ, Wu BW, Hostomsky Z, Reinecke MG and Robinson Jr. WE. Dicaffeoylquinic and dicaffeoyltartaric acids are selective inhibitors of human immunodeficiency virus type 1 integrase. Antimicrob Agents Chemother 1998;42(1):140–146. 80. Lin Z, Neamati N, Zhao H, Kiryu Y, Turpin JA, Aberham C, Strebel K, Kohn K, Witvrouw M, Pannecouque C, Debyser Z, De Clercq E, Rice WG, Pommier Y and Burke Jr. TR. Chicoric acid analogues as HIV-1 integrase inhibitors. J Med Chem 1999;42(8):1401–1414. 81. Min B-S, Tomiyama M, Ma C-M, Nakamura N and Hattori N. Kaempferol acetylrhamnosides from the rhizome of Dryopteris crassirhizoma and their inhibitory effects on three different activities of human immunodeficiency virus-1 reverse transcriptase. Chem Pharm Bull 2001;49(5):546–550. 82. Hopkins AL, Ren J, Milton J, Hazen RJ, Chan JH, Stuart DI and Stammers DK. Design of non-nucleoside inhibitors of HIV-1 reverse transcriptase with improved drug resistance properties. 1. J Med Chem 2004;47(24):5912–5922. 83. Kostova I. Coumarins as inhibitors of HIV reverse transcriptase. Curr HIV Res 2006;4(3):347–363. 84. Ahn M-J, Yoon K-D, Min S-Y, Lee JS, Kim JH, Kim TG, Kim SH, Kim N-G, Huh H and Kim J. Inhibition of HIV-1 reverse transcriptase and protease by phlorotannins from the Brown Alga Ecklonia Cava. Biol Pharm Bull 2004;27(4):544–547. 85. Anonymous, Encyclopedia of Chinese Materia Medica (Zhon Yao Dai Zi Ten). Jiangsu New Medical College, Ed.; Shanghai Science and Technology Publishers: Shanghai 1977, 2342. 86. Li YS, Wang ZT, Zhang M, Luo SD and Chen JJ. A new pinoresinol-type lignan from Ligularia kanaitizensis. Nat Prod Res 2005;19(2):125–129. 87. Uchiumi F, Hatano T, Ito H, Yoshida T and Tanuma S-I. Transcriptional suppression of the HIV promoter by natural compounds. Antiviral Res 2003;58:89–98. 88. Mizuno T, Uchino K, Toukairin T, Tanabe A, Nakashima H, Yamamoto N and Ogawara H. Inhibitory effect of tannic acid sulfate and related sulfates on infectivity, cytopathic effect, and giant cell formation of human immunodeficiency virus. Planta Med 1992;58:535–539. 89. Nakashima H, Murakami T, Yamamoto N, Sakagami H, Tanuma S, Hatano T, Yoshida T and Okuda T. Inhibition of human immunodeficiency viral replication by tannins and related compounds. Antiviral Res 1992;18:91–103. 90. Uchiumi F, Maruta H, Inoue J, Yamamoto T and Tanuma S. Inhibitory effect of tannic acid on human immunodeficiency virus promoter activity induced by 12-O-tetra decanoylphorbol-13-acetate in Jurkat T-cells. Biochem Biophys Res Commun 1996;220:411–417. 91. Aoki K, Maruta H, Uchiumi F, Hatano T, Yoshida T and Tanuma S. A macrocircular ellagitannin, oenothein B, suppresses mouse mammary tumor gene expression via inhibition of poly(ADP-ribose) glycohydrolase. Biochem Biophys Res Commun 1995;210:329–337. 92. Harris CJ and Stevens AP. Chemogenomics: structuring the drug discovery process to gene families. Drug Discov Today 2006;11(19–20):880–888.

262 93. Tlk 286. Drugs R D 2004;5(2):116–118. 94. Agrafiotis DK, Lobanov VS and Salemme FR. Combinatorial informatics in the postgenomics ERA. Nat Rev Drug Discov 2002;1(5):337–346. 95. Ashton GH, McGrath JA and South AP. Strategies to identify disease genes. Drugs Today 2002;38(4):235–244. 96. Birault V, Harris CJ, Le J, Lipkin M, Nerella R and Stevens A. Bringing kinases into focus: efficient drug design through the use of chemogenomic toolkits. Curr Med Chem 2006;13(15):1735–1748. 97. Bleicher KH. Chemogenomics: bridging a drug discovery gap. Curr Med Chem 2002;9(23):2077–2084. 98. Bredel M and Jacoby E. Chemogenomics: an emerging strategy for rapid target and drug discovery. Nat Rev Genet 2004;5(4):262–275. 99. Byvatov E and Schneider G. Support vector machine applications in bioinformatics. Appl Bioinformatics 2003;2(2):67–77. 100. Eguchi M, McMillan M, Nguyen C, Teo JL, Chi EY, Henderson Jr. WR and Kahn M. Chemogenomics with peptide secondary structure mimetics. Comb Chem High Throughput Screen 2003;6(7):611–621. 101. Eguchi Y. [Chemogenomics].. Nippon Yakurigaku Zasshi 2005;125(6):365–371. 102. Fielden MR, Pearson C, Brennan R and Kolaja KL. Preclinical drug safety analysis by chemogenomic profiling in the liver. Am J Pharmacogenom 2005;5(3):161–171. 103. Fradera X and Mestres J. Guided docking approaches to structure-based design and screening. Curr Top Med Chem 2004;4(7):687–700. 104. Gagna CE, Winokur D and Clark Lambert W. Cell biology, chemogenomics and chemoproteomics. Cell Biol Int 2004;28(11):755–764. 105. Guba W. Chemogenomics strategies for G-protein coupled receptor hit finding. Ernst Schering Res Found Workshop 2006(58):21–29. 106. Huang Y and Sadee W. Drug sensitivity and resistance genes in cancer chemotherapy: a chemogenomics approach. Drug Discov Today 2003;8(8):356–363. 107. Jacoby E. Chemogenomics: drug discovery’s panacea? Mol Biosyst 2006;2(5):218–220. 108. Jacoby E, Schuffenhauer A, Popov M, Azzaoui K, Havill B, Schopfer U, Engeloch C, Stanek J, Acklin P, Rigollier P, Stoll F, Koch G, Meier P, Orain D, Giger R, Hinrichs J, Malagu K, Zimmermann J and Roth HJ. Key aspects of the Novartis compound collection enhancement project for the compilation of a comprehensive chemogenomics drug discovery screening collection. Curr Top Med Chem 2005;5(4):397–411. 109. Jimonet P and Jager R. Strategies for designing GPCR-focused libraries and screening sets. Curr Opin Drug Discov Dev 2004;7(3):325–333. 110. Klabunde T and Hessler G. Drug design strategies for targeting G-protein-coupled receptors. Chembiochem 2002;3(10):928–944. 111. Klabunde T and Jager R. Chemogenomics approaches to G-protein coupled receptor lead finding. Ernst Schering Res Found Workshop 2006(58):31–46. 112. Kley N, Ivanov I and Meier-Ewert S. Genomics and proteomics tools for compound mode-of-action studies in drug discovery. Pharmacogenomics 2004;5(4):395–404. 113. Kubinyi H. Chemogenomics in drug discovery. Ernst Schering Res Found Workshop 2006(58):1–19. 114. Li Q and Xu W. Novel anticancer targets and drug discovery in post genomic age. Curr Med Chem Anticancer Agents 2005;5(1):53–63. 115. Mestres J. Computational chemogenomics approaches to systematic knowledge-based drug discovery. Curr Opin Drug Discov Dev 2004;7(3):304–313.

263 116. Mestres J. Representativity of target families in the Protein Data Bank: impact for family-directed structure-based drug discovery. Drug Discov Today 2005;10(23–24):1629–1637. 117. Miller GG and Voronina TA. [Perspective technologies for drug design].. Antibiot Khimioter 2005;50(2–3):52–63. 118. Savchuk NP, Balakin KV and Tkachenko SE. Exploring the chemogenomic knowledge space with annotated chemical libraries. Curr Opin Chem Biol 2004;8(4):412–417. 119. Scapin G. Protein kinase inhibition: different approaches to selective inhibitor design. Curr Drug Targets 2006;7(11):1443–1454. 120. Shen J, Xu X, Cheng F, Liu H, Luo X, Shen J, Chen K, Zhao W, Shen X and Jiang H. Virtual screening on natural products for discovering active compounds and target information. Curr Med Chem 2003;10(21):2327–2342. 121. Shokat K and Velleca M. Novel chemical genetic approaches to the discovery of signal transduction inhibitors. Drug Discov Today 2002;7(16):872–879. 122. ter Haar E, Walters WP, Pazhanisamy S, Taslimi P, Pierce AC, Bemis GW, Salituro FG and Harbeson SL. Kinase chemogenomics: targeting the human kinome for target validation and drug discovery. Mini Rev Med Chem 2004;4(3):235–253. 123. Vieth M, Higgs RE, Robertson DH, Shapiro M, Gragg EA and Hemmerle H. Kinomics-structural biology and chemogenomics of kinase inhibitors and targets. Biochim Biophys Acta 2004;1697(1–2):243–257. 124. Vivanco F, Lopez-Bescos L, Tunon J and Egido J. [Proteomics and cardiovascular disease].. Rev Esp Cardiol 2003;56(3):289–302. 125. Zheng XF and Chan TF. Chemical genomics: a systematic approach in biological research and drug discovery. Curr Issues Mol Biol 2002;4(2):33–43. 126. Caron PR, Mullican MD, Mashal RD, Wilson KP, Su MS and Murcko MA. Chemogenomic approaches to drug discovery. Curr Opin Chem Biol 2001;5(4):464–470. 127. Berman HM, Bhat TN, Bourne PE, Feng Z, Gilliland G, Weissig H and Westbrook J. The Protein Data Bank and the challenge of structural genomics. Nat Struct Biol 2000(7 Suppl):957–959. 128. Yu D, Wild CT, Martin DE, Morris-Natschke SL, Chen CH, Allaway GP and Lee KH. The discovery of a class of novel HIV-1 maturation inhibitors and their potential in the therapy of HIV. Expert Opin Investig Drugs 2005;14(6):681–693. 129. Vigouroux C, Gharakhanian S, Salhi Y, Nguyen TH, Adda N, Rozenbaum W and Capeau J. Diabetes and Metabolism 1999;25(3):383–392. 130. Kreis W, Kaplan MH, Freeman J, Sun DK and Sarin PS. Inibition of HIV replication by hyssop officinalis extracts. Antiviral Res 1990;14(6):323–337. 131. Leighton F, Cuevas A, Guasch V, Perez DD, Strobel P, San Martin A, Urzua U, Diez MS, Foncea R, Castillo O, Mizon C, Espinoza MA, Urquiaga I, Rozowski J, Maiz A and Germain A. Plasma polyphenols and antioxidants, oxidative DNA damage and endothelial function in a diet and wine intervention study in humans. Drugs Exp Clin Res 1999;25:133–141. 132. Lappara J, Michaud J and Masqualier J. Pharmacokinetic study of flavanolic oligomers. J Plant Med Phytotherarpy 1977;11:133–142. 133. Parsons AB, Lopez A, Givoni IE, Williams DE, Gray CA, Porter J, Chua G, Sopko R, Brost RL, Ho CH, Wang J, Ketela T, Brenner C, Brill JA, Fernandez GE, Lorenz TC, Payne GS, Ishihara S, Ohya Y, Andrews B, Hughes TR, Frey BJ, Graham TR, Andersen RJ and Boone C. Exploring the mode-of-action of bioactive compounds by chemical-genetic profiling in yeast. Cell 2006;126(3):611–625.

264 134. Aquaro S, Calio R, Balestra E, Bagnarelli P, Cenci A, Bertoli A, Tavazzi B, Di Pierro D, Francesconi M, Abdelahad D and Perno CF. Clinical implications of HIV dynamics and drug resistance in macrophages. J Biol Regul Homeost Agents 1998;12(1–2 Suppl):23–27. 135. Charvat TT, Lee DJ, Robinson WE and Chamberlin AR. Design, synthesis, and biological evaluation of chicoric acid analogs as inhibitors of HIV-1 integrase. Bioorg Med Chem 2006;14(13):4552–4567. 136. Olsuf’eva EN and Preobrazhenskaia MN. [Structure-activity relationships in a series of semisynthetic polycyclic glycopeptide antibiotics].. Bioorg Khim 2006;32(4):339–359. 137. Olszewski A, Sato K, Aron ZD, Cohen F, Harris A, McDougall BR, Robinson Jr. WE, Overman LE and Weiss GA. Guanidine alkaloid analogs as inhibitors of HIV-1 Nef interactions with p53, actin, and p56lck. Proc Natl Acad Sci USA 2004;101(39):14079–14084. 138. Sun Z, Chen B, Zhang S and Hu C. Four new eudesmanes from Caragana intermedia and their biological activities. J Nat Prod 2004;67(12):1975–1979. 139. Richter S, Parolin C, Gatto B, Del Vecchio C, Brocca-Cofano E, Fravolini A, Palu G and Palumbo M. Inhibition of human immunodeficiency virus type 1 tat-trans-activationresponsive region interaction by an antiviral quinolone derivative. Antimicrob Agents Chemother 2004;48(5):1895–1899. 140. Stalmeijer EH, Van Rij RP, Boeser-Nunnink B, Visser JA, Naarding MA, Schols D and Schuitemaker H. In vivo evolution of X4 human immunodeficiency virus type 1 variants in the natural course of infection coincides with decreasing sensitivity to CXCR4 antagonists. J Virol 2004;78(6):2722–2728. 141. Hanna L. Calanolide A: a natural non-nucleoside reverse transcriptase inhibitor. Beta 1999;12(2):8–9. 142. de Baar MP, Janssens W, de Ronde A, Fransen K, Colebunders R, Kestens L, van der Groen G and Goudsmit J. Natural residues versus antiretroviral drug-selected mutations in HIV type 1 group O reverse transcriptase and protease related to virological drug failure in vivo. AIDS Res Hum Retroviruses 2000;16(14):1385–1394. 143. Wang LX, Heredia A, Song H, Zhang Z, Yu B, Davis C and Redfield R. Resveratrol glucuronides as the metabolites of resveratrol in humans: characterization, synthesis, and anti-HIV activity. J Pharm Sci 2004;93(10):2448–2457. 144. Vlietinck AJ and Vanden Berghe DA. Can ethnopharmacology contribute to the development of antiviral drugs? J Ethnopharmacol 1991;32(1–3):141–153. 145. De Clercq E. Perspectives for the chemotherapy of AIDS. Anticancer Res 1987;7(5B):1023–1038.

265

Ganoderma lucidum and its pharmaceutically active compounds Bojana Boh1, Marin Berovic2,, Jingsong Zhang3 and Lin Zhi-Bin4 1

Faculty of Natural Sciences and Engineering, University of Ljubljana, Vegova 4, 1000 Ljubljana, Slovenia 2 Department of Chemical and Biochemical Engineering, Faculty of Chemistry and Chemical Technology, University of Ljubljana, Askerceva 5, 1001 Ljubljana, Slovenia 3 Institute of Edible Fungi, Shanghai Academy of Agriculture Sciences, Shanghai, P.R. China 4 Department of Pharmacology, Peking University Health Science Center, Beijing 10083, P.R.China Abstract. Ganoderma lucidum is a wood-degrading basidiomycete with numerous pharmacological effects. Since the mushroom is very rare in nature, artificial cultivation of fruiting bodies has been known on wood logs and on sawdust in plastic bags or bottles. Biotechnological cultivation of G. lucidum mycelia in bioreactors has also been established, both on solid substrates and in liquid media by submerged cultivation of fungal biomass. The most important pharmacologically active constituents of G. lucidum are triterpenoids and polysaccharides. Triterpenoids have been reported to posses hepatoprotective, anti-hypertensive, hypocholesterolemic and anti-histaminic effects, anti-tumor and anti-engiogenic activity, effects on platelet aggregation and complement inhibition. Polysaccharides, especially b-Dglucans, have been known to possess anti-tumor effects through immunomodulation and antiangiogenesis. In addition, polysaccharides have a protective effect against free radicals and reduce cell damage caused by mutagens. Keywords: Ganoderma lucidum, cultivation, wood logs, sawdust, solid-state cultivation, submerged cultivation, triterpenoids, polysaccharides, b-D-glucans, pharmacological effects, anti-cancer effects, immunomodulation.

Introduction Ganoderma is a white rot wood-degrading basidiomycete with hard fruiting bodies. G. lucidum (W.Curt.:Fr.) Lloyd and Ganoderma applanatum (Pers.) Pat. (Aphyllophoromycetideae) are two species most often reported as a source of medicinal compounds. In Asian traditional medicine, the fruiting body of G. lucidum (Fig. 1), called Ling-Zhi in Chinese and Reishi in Japanese language, has been used for treatment of several diseases for thousands of years, as reported in Shen Nong’s Materia Medica [1,2]. However, an increasing systematic research (Fig. 2) into the Ganoderma active compounds elucidates its numerous pharmacological effects, such as antitumor, immunomodulatory, cardiovascular, respiratory, antihepatotoxic and central nervous system effects. Modern uses of Ganoderma therefore include treatment of coronary heart diseases, arteriosclerosis, hepatitis, arthritis, nephritis, bronchitis, asthma, hypertension, cancer and gastric ulcer [1,3]. Publications also report on Ganoderma antiallergenic constituents [4], immunomodulatory

Corresponding author: Tel: 386-1-2419510. Fax: 386-1-4760-300. E-mail: [email protected] (M. Berovic). BIOTECHNOLOGY ANNUAL REVIEW VOLUME 13 ISSN 1387-2656 DOI: 10.1016/S1387-2656(07)13010-6


266

Fig. 1. Fruiting body of Ganoderma lucidum (MZKI G97) originally isolated from the Slovenian forest.

600 Nonpatent Patent

Number of documents

500

400

300

200

100

0 1980

1985

1990

1995

2000

2005

Year

Fig. 2. Number of new documents on Ganoderma in the Chemical Abstracts Plus database. An increasing growth of publications indicates an accelerated research in Ganoderma mushrooms. The larger proportion of patents vs. non-patent publications during the last years suggests that a shift happened from basic to more applied research and development of new cultivation technologies, pharmaceutical products and nutriceutical formulations.

267 action [5,6], antitumor activity [7], cardiovascular effects [8], liver protection and detoxification, and effects on nervous system [9]. New reports emphasize its potential in treatment of viral, especially HIV infections [10–15]. G. lucidum is very scarce in nature. As the demands of international markets for G. lucidum fruiting bodies and/or mycelium biomass are in constant increase, an artificial cultivation has become essential. Successful farming on wood logs and in bags filled with wood or straw substrates has been known for decades, especially in China. Biotechnological cultivation in bioreactors on solid substrates, or with submerged liquid substrate cultivation has been developed and introduced for small and pilot-plant production [16–20]. The quality and content of physiologically active substances vary from strain to strain, and also depend on location, culture conditions [21], the growth stage of the fungus [22], the processing procedures, and formulation preparation [23]. Diverse groups of chemical compounds with pharmacological activity have been isolated from the mycelium and fruiting body of Ganoderma species: triterpenoids, polysaccharides, proteins, amino acids, nucleosides, alkaloids, steroids, lactones, fatty acids and enzymes [1,3]. The most important pharmacologically active constituents of Ganoderma mushrooms are triterpenoids and polysaccharides. G. lucidum cultivation methods As G. lucidum is very scarce in nature, artificial cultivation has become essential to meet the demands of international markets (Fig. 3). The main traditional G. lucidum fruiting body cultivation methods remain sawdust cultivation in bags or bottles and cultivation on natural logs. Both cultivation technologies depend on the same essential environmental factors, including temperature, humidity and oxygen [24]. During the spawn run, mycelia grow at 10–381C, with the optimum mycelial incubation temperature between 251C and 321C. The optimum moisture content of sawdust substrate is 65–70% and that of log is around 40%. pH 4.2–5.3 is regarded as optimum. The mycelial growth does not necessarily need light. Oxygen is indispensable to mycelial growth since G. lucidum is a strict aerobe. In the next cultivation stage of primordia formation, G. lucidum fruits and develops at 20–341C, with the optimum temperature 27–321C. The humidity of the growing room should be maintained at about 90% during primordial induction, 70–80% during cap formation and 30–40% during the final stage of fruit body development. Light (50–450 lux) is required during primordial formation and fruit body development. After the cap is formed, the growing room has to be well ventilated. Chen [25] published detailed information on substrate formulation for G. lucidum. As G. lucidum is a lignin-degrading white-rot fungus on hardwoods, woody tissue, such as sawdust, is a natural substrate. Thiamin

268 Unsterilised long wood logs (outdoors) Sterilised short wood logs (indoors) Fruiting bodies cultivation Sawdust in bags

Sawdust in bottles Ganoderma lucidum cultivation

Sawdust in beds or trays Mycelia cultivation in bioreactors

Solid state cultivation Submerged cultivation in liquid media

Fig. 3. Main cultivation methods for the production of G. lucidum fruit bodies and

mycelia.

contained in fresh, unprocessed coarse bran is required for mushroom formation. A low content of sugar (1% sucrose) triggers formation and activation of lignin-decomposing enzymes. Calcium appears to encourage mushroom differentiation. Water-logging in substrate prevents air exchange and cuts off oxygen supply. If the sawdust particle size is too fine, proper air exchange is impeded. On the other hand, rough wood chips in substrate may puncture the bag and invite contamination. Cultivation of fruit bodies on natural wood logs Cultivation on long unsterilized logs In the past, natural logs as long as one meter were used without sterilization in growing Ganoderma species in China. Fruiting body cultivation on long wood logs took much labor. Long incubation periods (2–3 years) were required to obtain mature fruiting bodies on such substrates [24,26]. Cultivation on short sterilized logs Since the late 1980s, new trends have been developed using short logs. Almost all Ganoderma spp. natural log growers adopted the short-log cultivation in China, Japan, the United States and elsewhere. High yield in a shorter cultivation time enabled quicker turnover of the capital. Short-log cultivation

269 takes only 4–5 months for mycelial incubation, and the fruiting body can be cropped in the same year. Detailed procedures on short-log cultivation were published elswere [24,26–28]. The main cultivation stages of growing G. lucidum on short natural logs enclosed in ventilated synthetic bags during spawn run include [26]: Preparation of wood logs. Most broad-leaf hardwoods can be used, the

standard log size is 15 cm in diameter, and 15–24 cm long. Moisture content is 35–40%. Enclosing logs into bags, sterilization. Heat-sealed polypropylene or polyethylene bags with micro-filter windows are used. Spawning. A variety of spawns, such as pure culture liquid mycelial spawn, grain spawn and sawdust-bran spawn can be used. Usually 5–10 g of spawn is used for each log. Spawn run is carried out in darkness, and less oxygen is required. Special attention should be given to ensure proper mycelial colonization in the log. Lack of oxygen or poor aeration, such as water-logging, results in poor mycelial growth and slow growth rate. Primordia initiation. Ganoderma spp. primordia are usually formed 50–60 days after spawning. Brief exposure to very little light triggers Ganoderma spp. primordia initiation. Oxygen is also conducive to primordia formation. Embedding in soil. Colonized logs are embedded vertically directly in soil after primordia formation, leaving the primordia above the ground level. To retain moisture, the soil is covered with chopped straw. Maintaining suitable growth parameters. The most crucial factor during primordia initiation is to have high humidity, preferably 90–95%, while the most crucial factor during pileus differentiation in fruiting, is increase in ventilation to reduce CO2 build-up from the drastic increase in respiration from Ganoderma spp. fruiting. Differentiation of Ganoderma spp. fruiting is highly sensitive to CO2 concentration, which determines whether antler-shaped fruiting bodies (CO2 4 0.1%), or fruiting bodies with a well-formed pileus (CO2o0.1%) will be produced. CO2 concentration at 0.04–0.05%, as close to fresh air as possible (0.03% CO2), should be maintained for production of pileated mushrooms (mushrooms with caps). Air humidity can be supplied by a fine mist (1–2, or 3–4 times/day). Harvesting mushrooms. From primordia formation to fruiting body for harvest takes approximately 25 days. Fruiting maturity is indicated by the disappearance of the undifferentiated white growth at the edge of the fruiting body. Cultivation is then continued at reduced air humidity of 60–85% for additional 7–10 days to encourage further growth in pileate thickness and firmness. Harvest is done by cutting the stipe (stalk); only 2 cm of the stipe remains with the pileus.

270 Post-harvest treatment includes immediate drying under the sun or with

heat (601C) for 2–3 days. Improper drying lowers the quality of the product.

Cultivation of fruit bodies on sawdust substrates Sawdust substrates in sterilizable bags (synthetic log cultivation) According to Royse [29], most cultivation of G. lucidum is on supplemented sawdust contained in heat-resistant polypropylene bottles or bags. Sawdust of hardwoods is usually supplemented with rice bran (10%) and CaCO3 (3%), moistened with water and filled (700 g) into plastic bags. A plastic collar then is fitted onto each bag and closed with a cotton plug. After heat treatment (95–1001C for 5 h) the substrate is allowed to cool overnight and then inoculated with grain or sawdust spawn. The inoculated substrate is incubated for 3–4 weeks or until the spawn has fully colonized the substrate. Mushroom production is initiated by maintaining air temperature at about 281C with relative humidity in the range 85–90%. Basidiocarps begin to appear in about 1–2 weeks after initiation. Approximately 2–3 months after the appearance of primordia, mushrooms are ready to harvest. A mushroom is considered mature when the whitish margin around the edge of the basidiocarp has turned red. The substrate may yield another harvest of mushrooms after removal of the first flush. Chen [25] recommended the following substrate formulation for Ganoderma cultivation: oak sawdust 80%, fresh unprocessed coarse wheat bran 18%, supplemented with sucrose 1%, calcium carbonate (or calcium sulphate) 1%, with approximately 67– 70% water. For 500 g dry substrate (containing 400 g oak sawdust and 90 g wheat bran), 1 l of water containing 5 g of sucrose and 5 g of calcium carbonate was added. Using this formulation, scale-up cultivation by mushroom growers in the United States and Canada has been successful. Another extensive article on Ganoderma cultivation in bags (synthetic logs) is available in [30]. Several publications describe G. lucidum cultivation in bags under unconventional conditions. A Japanese patent [31] claimed cultivation of G. lucidum fruit bodies in an antler form in bags. Nascimento [32] grew G. lucidum on hardwood chips and sawdust of two Chilean native red wood trees, Nothofagus obliqua and Nothofagus alpine, in polypropylene bags. No difference was observed in the fruiting stage between the two species of tree wood. Gonzalez-Matute et al. [33] studied the sunflower seed hulls as a main nutrient source for the cultivation of G. lucidum in bags (synthetic log system). The study concluded that sunflower seed hulls could be used as the main energy and nutritional source in substrates for G. lucidum cultivation, and the addition of 5% malt to the substrate improved the mushroom growth rate.

271 Yang et al. [34] utilized stillage grain from a rice-spirit distillery in the cultivation of G. lucidum in polypropylene bags. Due to its high content of carbohydrate and nitrogen, stillage grain was considered as a nutritive substrate for mycelial growth. Sawdust supplemented with stillage grain at a ratio 4:1 at a water content of 60% was optimal for the production of fruiting body. Hsieh et al. [35] used the soy residue from the waste of tofu manufacturing to culture G. lucidum in polypropylene plastic bags. The fruiting bodies were fully developed only for the C/N ratios of 70 and 80. Sawdust substrates in bottles and pots Kim [36] grew twenty-one isolates of nine Ganoderma species (including G. lucidum) in 2 l sterilizable plastic bottles on solid substrates based on sawdust. The substrate was prepared by mixing oak sawdust and wheat bran (8:2, v/v), and the water was added to 65% of the total volume. The cultivation room was maintained at 28–311C and 85% of relative humidity until primordial formation and ventilated once or twice for 10–20 min afterwards. After the pileus was formed, ventilations were frequent (5–6 times/day) and the relative humidity was controlled at 80–85%. A Japanese patent [37] claimed an arrangement for cultivating G. lucidum or other mushrooms in bottle containers, where a negative voltage was applied to the container body for activation and production of mushrooms of excellent quality in a short period. Another Japanese patent described a cultivation method for Ganoderma in pots by protecting a nursery bed against the infiltration of bacteria by a plastic cover. In a work by Shigeru [39], finely crushed thinned citrus fruits and residue of citrus fruit juice were utilized in the culture medium, mixed with rice bran at a weight ratio of about 10:2 for G. lucidum in bottles. Ganoderma was cultured at about 25–301C and about 60–90% humidity to develop fruit bodies. Sawdust substrates in trays or beds Chen [28] reported on G. lucidum cultivation on trays or beds in North America. The article stated that wood-chip or sawdust beds were labor saving, if contamination was avoided. The substrate, 12 cm deep, composed of wood chips supplemented with sawdust, was spread evenly over cultivation trays or beds. Colonized sawdust, grain, or liquid spawn, 0.5 cm thick or more, was seeded on the surface under still air and covered with a plastic sheet. After 3–5 days, a layer of white mycelia became visible and begun to penetrate the substrate. When primordia were formed after 1–2 months, the plastic cover sheet was removed. The growing space was maintained at onethird diffuse light, 85–95% relative humidity and 251C. Air circulation or aeration was provided 3–5 times/day 5–10 min each time.

272 Cultivation of G. lucidum mycelia in a bioreactor Solid-state cultivation A Slovenian patent [40] claimed a process of growing G. lucidum on a solid cultivation substrate using the solid-state cultivation in a horizontal stirred bioreactor. Beech sawdust was used as a solid cultivation substrate. The process enabled a precise leading and monitoring of the fungal growth at sterile conditions. Large quantities of biomass could be prepared according to the process to yield products applicable to pharmacy. The biomass could also be used as a solid inoculum for the further cultivation of G. lucidum. A World patent application with a Chinese priority [41] described a method for propagating fungi and producing fungal metabolites with medicinal activities, using a solid-state fermentation process and cultivation in bottles. The invention also described substrates for small- and large-scale fungal cultivation of G. lucidum, Cordyceps sinensis, Antrodia camphorata, Trametes versicolor and Agaricus blazei. Chen [28] reported that in North America, Ganoderma mycelial preparations for human consumption are produced by solid-state fermentation on grain- or soy-based substrates. Submerged cultivation in liquid media Authors use different substrate compositions for the submerged cultivation of G. lucidum mycelia. An example given in [28] consisted of sucrose 50.0 g, ammonium succinate 3.2 g, KH2PO4 1.0 g, MgSO4 7H2O 0.3 g, FeSO4 7H2O 13.0 mg, ZnSO4 7H2O 4.0 mg, yeast or malt extract 10.0 g, adjusted to pH 5.2 with concentrated ammonia, water to 1 l. An optimized medium for a shake-flask submerged culture of G. lucidum, reported by Chang et al. [42] consisted of 1.88 g/l CaCO3, 71.4 brown sugar, 12.1 g/l malt extract, 2.28 g/l yeast extract, 18.4 g/l skim milk, 3.44 g/l safflower seed oil, 3.96 g/l olive oil, at pH 6.5. Compared to an unoptimized substrate, the mycelium formation was markedly improved from 1.70 g/l to 18.70 g/l, and the polysaccharide production increased from 0.140 g/l to 0.420 g/l. Hsieh et al. [43] studied production of polysaccharides by G. lucidum in shake flasks under various limitations of nutrients, including carbon-source, nitrogen-source, phosphate-source, magnesium-source and dissolved oxygen. Different responses of polysaccharide production were observed under different limitations of nutrients. A Slovenian patent [44] described a procedure for G. lucidum inoculum preparation in a shaked culture, and the production of mycelia by submged fermentation in a bioreactor. According to the patent description, potato dextrose agar of the total area of 100–200 mm2 overgrown by the fungus

273 G. lucidum was transferred into a 500 ml Erlenmayer flask containing 100 ml of substrate. The vegetative substrate contained the filtrate of 300 g/l of peeled cooked potato, 20 g/l of glucose and 2% v/v of olive oil, and was filled with distilled water of pH 5.8 to the total volume of 1 l. The culture was shaken in this medium for 80–160 h at the temperature of 20–301C at 80–160 rotations per minute. In a bioreactor, the substrate was sterilized at the temperature of 110–1301C and with mixing 200–400 rotations per minute, for about half an hour. After cooling down to 301C, the sterile substrate in a bioreactor was inoculated with 17% v/v of the vegetative inoculum containing mycelium of G. lucidum, produced in a shaken culture of the age of 120–170 h. The growth of the mycelium lasted for 160 h during which the concentration of dissolved oxygen was maintained by aeration of 6–15 l/min at 200–600 rotations/min, maximal redox potential during growing amounting to 410–460 mV and minimal partial oxygen pressure from 25 to 33% v/v at pH from 4.10 to 4.30. Yang and Liau [45] studied the influence of cultivation parameters on polysaccharide formation by G. lucidum in submerged cultures. The substrate consisted of glucose 50 g/l; K2HPO4 0.5 g/l; KH2PO4 0.5 g/l; MgSO4 7H20 0.5 g/l; yeast extract 1 g/l; and ammonium chloride 4 g/l. Optimal temperature was 30–351C and the pH 4–4.5. Polysaccharide concentration reached 1.6 mg/ml. Agitation and aeration influenced the formation and secretion of polysaccharides. The optimal rotating speed was 150 rpm in 7-day flask cultures, while the agitation speed employed in fermenter culture greatly affected the production rate and maximum concentration of polysaccharide. Although higher speeds enhanced mixing efficiency and polysaccharide release, higher shear stress had a detrimental effect on mycelial growth and polysaccharide formation. Lee [46] reported that pH control substantially affected mycelial cell growth and exopolysaccharide production of the mycelial cultivation of G. lucidum. The fermentor was a concentric draught-tube internal loop type which can greatly reduce shear stress as compared to an impeller-type fermentor. Five percent (v/v) of the culture was inoculated into the fermentor and cultivated at 251C with air being supplied at a rate of 2.5 vvm under batch conditions. Compared to the uncontrolled pH cultivation, the bistage pH control technique, in which pH was shifted from 3 to 6 at the initial phase of the exponential growth, increased exopolysaccharide production from 4.1 g/l to 20.1 g/l, retained the desirable morphologies of the mycelia during cultivation, and resulted in low viscosity and yield stress of the culture broth. Fang and Zhong [47] investigated the effects of initial pH on simultaneous production of ganoderic acid and a polysaccharide by G. lucidum. An initial pH value, varying within the range 3.5–7.0, had a significant effect on the cell growth and product biosynthesis. At an initial pH of 6.5, a maximum in biomass of 17.370.12 g/l by dry weight was obtained, as well as a maximal

274 specific production of ganoderic acid of 1.2070.03 mg/100 mg dry weight and total production of 207.972.7 mg/l. Lowering the initial pH from 6.5 to 3.5 gradually led to a higher production of extracellular polysaccharide and a higher specific production of intracellular polysaccharide. The same research group [48] studied the effects of nitrogen source and initial glucose concentration in submerged fermentation of G. lucidum for simultaneous production of bioactive ganoderic acids and polysaccharides. The cells could not grow well when either yeast extract or peptone was used as the sole nitrogen source. However, a combined addition of 5 g/l of yeast extract and 5 g/l of peptone was optimal for the cell growth and metabolite production. Initial glucose concentration within 20–65 g/l greatly affected the cell growth and product biosynthesis. The highest levels of cell density (16.7 g dry weight/l), intracellular polysaccharide (1.19 g/l) and ganoderic acid (212.3 mg/l) were obtained at an initial glucose concentration of 50 g/l. Fang et al. [49] also reported on a significance of inoculation density control in production of polysaccharide and ganoderic acid by submerged culture of G. lucidum. Control of inoculation density was significant for cell growth, morphology, and production of polysaccharide and ganoderic acid. A maximal cell concentration of 15.7 g dry cell weight/l was obtained at an inoculation density of 330 mg dry weight/l. For inoculation density within the range 70–670 mg dry weight/l, a large inoculation density led to a small pellet size and high production of extracellular and intracellular polysaccharides, while a relatively big pellet size and high accumulation of ganoderic acid were observed at a low inoculation density. It was also shown that small pellet size resulted in high polysaccharide production, while large pellet size led to high production of ganoderic acid. In a work of Berovic et al. [20] G. lucidum was cultivated in a liquid substrate based on potato dextrose and olive oil. The influences of inoculum and oxygen partial pressure in batch and fed-batch cultivation in a 10 l laboratory-stirred tank reactor were studied. The cultivation conditions were temperature of cultivation, T ¼ 301C; mixing, N ¼ 300 min1; aeration, Qg ¼ 10 l min1; average values of pH, 5.8–4.2; oxygen partial pressure, 70–80% and redox potential, Eh ¼ 300–400 mV. Fungal biomass was found to be oxygen and shear sensible. Using a 17% (wet weight) 6 days old vegetative inoculum, 9.6 g ll of dry biomass in batch cultivation and 15.2 g ll in fed-batch process were obtained. Extracellular (9.6 g ll) and intracellular (6.3 g ll) polysaccharide fractions were isolated. Extracellular polysaccharide fraction and four intracellular polysaccharide fractions were obtained. Polysaccharides were further separated by ion-exchange, gel and affinity chromatography. The isolated polysaccharides were mainly b-D-glucans. Immunostimulatory effects of isolates were tested on induction of cytokine (tumor necrosis factor a (TNF-a) and interferon g (IFN-g)) synthesis in primary cultures of human peripheral blood mononuclear cells (PBMC) isolated from a buffy coat. The TNF-a-inducing activity was comparable with

275 romurtide, which has been used as a supporting therapy in cancer patients treated with radiotherapy and/or chemotherapy. Tang and Zhong [50] studied the effects of carbon source and initial sugar concentration on the production of ganoderic acids and polysaccharided in a fed-batch G. lucidum cultivation process in shake flasks and in a stirred bioreactor. Sucrose as a carbon source was suitable for the extracellular polysaccharide production although the cells could not grow well. Lactose was beneficial for the cell growth and production of ganoderic acid and intracellular polysaccharides. When the initial lactose concentration exceeded 35 g/l, the ganoderic acid accumulation was decreased. The ganoderic acid production was remarkably improved by pulse feeding of lactose, when its residual concentration was between 10 g/l and 5 g/l. Submerged fermentation of G. lucidum is viewed as a fast and cost-effective alternative for efficient production of polysaccharides and ganoderic acids from G. lucidum. However, submerged cultivation of mycelia is characterized by an increase in broth viscosity with time, which is a consequence of increased cell concentration and the accumulation of extracellular polysaccharides that dramatically alter the rheological characteristics of fermentation broth, and create a series of problems that have to be solved, especially oxygen supply. Oxygen affects cell growth, cellular morphology, nutrients uptake and metabolite biosynthesis. Tang et al. [51] reported on the effects of oxygen supply on the submerged cultivation of G. lucidumin in a 3.5 l agitated bioreactor with two six-bladed turbine impellers. Aeration was through a ring sparger with a pore size of 0.8 mm. Fermentation was conducted at 301C in the dark. The cultivation medium consisted of 35 g/l lactose, 5 g/l peptone, 2.5 g/l yeast extract, 1 g/l KH2PO4 H2O, 0.5 g/l MgSO4 7H2O and 0.05 g/l Vitamin B1. The results showed that an initial volumetric oxygen transfer coefficient (KLa) value within the range 16.4–96.0 h1 had a significant effect on the cell growth, cellular morphology and metabolites biosynthesis. An increase of initial KLa led to a bigger size of mycelia aggregates and a higher production of ganoderic acids. Fang et al. [49] studied the significance of inoculation density and pellet size in the submerged culture of G. lucidum on production of polysaccharides and ganoderic acids. Inoculum sizes of 70, 170, 330 and 670 mg dry cell weight (DW)/l were tested. Inoculation density significantly affected the cultivation process. Small pellet size resulted in high polysaccharide production, while large pellet size led to high production of ganoderic acid. Pellets with diameters smaller than 1.2 mm, 1.2–1.6 mm and larger than 1.6 mm had ganoderic acid content of 0.98, 1.27 and 1.62 mg/100 mg DW, respectively. Some authors reported on submerged cultivation of G. lucidum mycelia on unconventional substrates, including liquid waste materials, such as thin stillage and deproteinated cheese whey. Hsieh et al. [52] produced G. lucidum polysaccharides by reusing thin stillage (from a wine manufacturing plant) in a shake flask culture. By

276 adjusting the pH 5, a 60% thin stillage was used successfully to grow the mycelia of G. lucidum with the highest cell concentration of 7.8 g/l and polysaccharide production of 7.50 g/l. Molasses addition produced the highest mycelia growth rate and a cell concentration increase to 12.7 g/l. Glucose addition led to increase overall polysaccharides production up to 3.69 g/l. Polysaccharides in the range of molecular weight from 10,000 to 200,000 Da were also found at almost three times production with thin stillage only. Lee et al. [53,54] utilized deproteinated cheese whey for cultivating mycelia of the G. lucidum in a bioreactor by submerged cultivation and concluded that cultivation of G. lucidum mycelia could offer a potential cost-effective solution for an alternative utilization of the deproteinated cheese whey. Main pharmacologically active compounds in G. lucidum The most important pharmacologically active constituents of Ganoderma mushrooms are triterpenoids and polysaccharides (Fig. 4). Triterpenoids from Ganoderma mushrooms Over 150 triterpenoids were found in Ganoderma spp., such as ganoderic (highly oxygenated C30 lanostane-type triterpenoids), lucidenic, ganodermic, ganoderenic, ganolucidic and applanoxidic acids, lucidones, ganoderals and ganoderols [5–64]. Representative examples are shown in Figs. 5–13. Boh et al. [65] reported that the quantity of triterpenoids differed in older and younger parts of G. applanatum fruiting bodies. The highest amount of triterpenoid acids was found in the tubes (6.4 mg/g of air-dried weight), followed by the younger dark context layer of the pileus (2.5 mg/g), the older white context layer (0.6 mg/g) and the upper surface of the fruiting body (0.6 mg/g). Triterpenoids have numerous pharmacological effects, as summarized below: Anti-hepatotoxic and hepatoprotective effects Hirotani et al. [55] successfully isolated ganoderic acids R and S from the cultured mycelia and proved their strong anti-hepatotoxic activity in the galactosamine-induced cytotoxic test with primary cultured rat hepatocytes. Kim et al. [66] reported on beta-glucuronidase-inhibitory and hepatoprotective effect of G. lucidum. Anti-tumor effects Isolated ganoderic acids Z, Y, X, W, V and T from Ganoderma mycelia demonstrated cyctotoxic activitiy in vitro on hepatoma cells [67].

277 Anti-hepatotoxic and hepatoprotective Anti-tumor

Anti-angiogenic

Anti-hypertensive Effects of triterpenoids Hypocholesterolemic

Anti-histaminic Ganoderma lucidum pharmacological effects

Platelet aggregation

Complement inhibition

Anti HIV Anti-tumor effect through immunomodulation Effects of polysaccharides

Anti-tumor effect through anti-angiogenesis Cell protection from free radicals and peroxidation

Fig. 4. Main pharmacological effects of G. lucidum.

H3C OAc CH3 O

COOH O

CH3

CH3 O CH3 O

O H3C

CH3

Fig. 5. Ganodermic acid F (12b-acetoxy-3,7,11,15,23-pentaoxo-5a-lanost-8-en-26-

oic acid).

278 COOH

CH3 CH3

CH3

CH3 OAc

CH3 AcO CH3

CH3

Fig. 6. Ganodermic acid R ((24E)-3a,15a-diacetoxy-5a-lanosta-7,9(11),24-triene26-oic acid).

O

CH3

CH3

CH3

COOH

O CH3 OH CH3 O

OH H3C

CH3

Fig. 7. Ganoderenic acid A ((20E)-7b,15a-dihydroxy-3,11,23-trioxo-5a-lanosta-8,20-

dien-26-oic acid). COOH

H3C O

CH3

O CH3 CH3 O

O

O H3C

CH3

Fig. 8. Lucidenic acid D1 (4,4,14a-trimethyl-3,7,11,12,15-pentaoxo-5a-chol-8-en-

24-oic acid).

Lin et al. [68] reported that a triterpene fraction from G. lucidum significantly inhibited growth of human hepatoma Huh7 cells, probably due to the oxidative stress induction. The same triterpenoid extract only had little influence on normal human liver cell line.

279 CH3

COOH

CH3 CH3

O

O CH3

OH CH3 O CH3

CH3

Fig. 9. Ganolucidic acid A (15a-hydroxy-3,11,23-trioxo-5a-lanost-8-en-26-oic acid).

CH3

O

CH3

O

CH3

COOH

CH3 O

CH3

OH

O H3C

CH3

Fig. 10. Applanoxidic acid A ((20E)-15a-hydroxy-7a,8a-epoxy-3,11,23-trioxo-5a-la-

nosta-9(11),20-dien-26-oic acid).

CH3

O

CH3 O CH3 O CH3 HO

OH CH3

CH3

Fig. 11. Lucidone (3b,7b-dihydroxy-4,4,14a-trimethyl-11,15,20-trioxo-5a-pregn-8-

ene).

Gao et al. [69] isolated three new lanostante-type triterpene aldehydes, named lucialdehydes A-C, from the fruiting bodies of G. lucidum. Lucialdehydes B, C showed cytotoxic effects on Lewis lung carcinoma (LLC), T-47D, Sarcoma 180, and Meth-A tumor cell lines. Lucialdehyde C exhibited the most potent cytotoxicity against the tested cell lines with ED50 values of 10.7, 4.7, 7.1 and 3.8 mg/l, respectively.

280 CHO

CH3 CH3

CH3

CH3 CH3 O CH3

CH3

Fig. 12. Ganoderal A ((24E)-3-oxo-5a-lanosta-7,9(11),24-triene-26-al).

CH3 CH3

CH2OH CH3

CH3 CH3 HO CH3

CH3

Fig. 13. Ganoderol B (ganodermadiol-5a-lanosta-7,9(11),24-triene-3b,26-diol).

Six new highly oxygenated lanostane-type triterpenes isolated from Ganoderma spores also showed direct cytotoxicity in vitro on the Meth-A and LLC tumor cell lines [70]. It has also been suggested that the triterpeneenriched fraction, WEES-G6, prepared from mycelia of G. lucidum inhibited the growth of human hepatoma Huh-7 cells. Treatment with WEES-G6 caused a rapid decrease in the activity of cell growth regulative protein, PKC, and the activation of JNK and p38 MAP kinases, which resulted in a prolonged G2 cell cycle phase and strong growth inhibition of the hepatoma cells [71]. The alcohol extract of G. lucidum also showed that it inhibited cell proliferation in a dose- and time-dependent manner, which might be mediated through up-regulation of p21/Waf1 and down-regulation of cyclin D1. Furthermore, it can directly induce apoptosis in MCF-7 cells, which might be mediated through up-regulation of a pro-apoptotic Bax protein and not by the immune system (Hu et al., 1999a). Two alcohol extracts (I and III) from G. lucidum spores strongly inhibited the growth of HeLa cells. Moreover, extract III was shown to be capable of blocking the cell cycle at the transition from G1 to S phase and inducing a marked decrease of intracellular calcium level. These results imply that the effective extract might influence the cell

281 cycle and cellular signal transduction by altering the calcium transport system [72]. Yu et al. [73] reported on the correlation between intracellular triterpenes from mycelia of G. lucidum in different growth stages of submerged cultivation and the inhibition effect on K562 tumor cells. The results showed that intracellular triterpenes were produced mainly in the later period of fermentation. Intracellular triterpenes from different stages of fermentation varied in types, quantity and relative proportion, and exhibited different effects on tumor cell inhibition. Optimum culture conditions for producing intracellular triterpenes inhibitory to K562 cells were shake flask cultures. The results of Liu et al. [74] indicated that the triterpenoids fraction of G. lucidum might be a useful ingredient in the treatment of benign prostatic hyperplasia. The ethanol extract of G. lucidum showed inhibitory activity on both isozymes (types 1 and 2) of 5a-reductase and suppression effects of ventral prostate growth induced by testosterone in castrated rat, but not induced by dihydrotestosterone. Activity-guided fractionation and TLC analysis suggested that the active principles in vivo were triterpenoids. Mueller et al. [75] results indicate that G. lucidum extract has a profound activity against leukemia, lymphoma and multiple myeloma cells and may be a novel adjunctive therapy for the treatment of hematologic malignancies. In their work, G. lucidum extract was screened for its anti-proliferative activity using a panel of 26 human cancer cell lines.

Anti-angiogenic effect Kimura et al. [76] reported that anti-tumor and anti-metastatic activities of a triterpenoid fraction of G. lucidum, containing ganoderic acid F, were due to the inhibition of tumor-induced angiogenesis. The triterpenoid fraction (100 and 200 mg/kg) of the fruit bodies of G. lucidum inhibited primary solidtumor growth in the spleen, liver metastasis and secondary metastatic tumor growth in the liver in intrasplenic Lewis lung carcinoma (LLC)-implanted mice. In addition, the triterpenoid fraction (800 mmg/ml) inhibited angiogenesis induced by Matrigel (a soluble basement membrane extract of the Engelbreth-Holm-Swarm tumor) supplemented with vascular endothelial growth factor (VEGF) and heparin in an in vivo model.

Anti-hypertensive effects Morigiwa et al. [57] found out that some G. lucidum triterpenes inhibited angiotensin converting enzyme, while Kabir et al. [77] reported on dietary effects of G. lucidum on blood pressure and lipid levels in spontaneously hypertensive rats.

282 Hypocholesterolemic effects Lin and Shiao [60] reported on inhibitory effects of ganoderic acid Mf and ganodermic acid T-O on cholesterol synthesis. Anti-histaminic effects Kohda et al. [78] studied biologically active constituents of G. lucidum and found out that triterpenes, such as ganodermic acids C and D, inhibit histamine release. Platelet aggregation effects Wang et al. [79,80] reported that ganodermic acid S exhibited amphipathic effect on the platelet aggregation. At high concentration of ganodermic acid S platelet aggregation occurred, while at a low concentration the aggregation was inhibited. The inhibition was concentration and time dependent. Su et al. [81,82] reported that collagen-induced platelet aggregation by ganodermic acid S was due to blocking of Ca mobilization through the thromboxane A2-dependent pathway in human platelets response to collagen, and that ganodermic acid S elevated prostaglandin E(1)-induced cyclic AMP in human platelets. Complement inhibition effect Min et al. [83] reported that ganoderiol F, ganodermanondiol, ganodermanontriol from spores of G. lucidum exhibited strong anti-complement activity, through which these substances can affect the humoral immune system in the host defense. Anti-HIV activity In 1997, Kim et al. [12] reported that the water-soluble extract of G. lucidum inhibited cytopathic effect of HIV-1, and Hattori et al. [13] reported on inhibitory effects of components from G. lucidum on the growth of human immunodeficiency virus (HIV) and its protease activity. El-Mekkway et al. [14] studied anti-HIV-1 and anti-HIV-1 protease substances from G. lucidum. Ganoderiol F and ganodermanontriol were active as anti-HIV-1 agents in concentration 7.8 mg/ml. Several other Ganoderma triterpenoids moderately inhibited HIV-1 protease activity at concentrations of 0.17–0.23 mM. Similar results with another set of G. lucidum triterpenoids were reported by Min et al. [15]. Ganoderic acid b, ganodermanondiol, ganodermanontriol, ganolucidic acid A and lucidumol B showed strong anti-HIV-1 protease activity with IC50 vaules of 20–90 mM.

283 Polysaccharides from Ganoderma fruit bodies and mycelia In recent years, a lot of scientific attention has been focussed on Ganoderma polysaccharides, which represent a structurally diverse class of biological macromolecules with a wide range of physicochemical properties. Studies have shown that the most active immunomodulatory polysaccharides are water-soluble b-1-3-D and b-1-6-D glucans, that can be precipitated by ethanol. Their prevailing structure is b-1-3 D-glucopyronan with 1–15 units of b-1-6 monoglucosyl side chains. Their 1,3-linked backbone, relatively small side chains and an organized helical structure are beneficial for the immunostimulation [84]. Other immunomodulatory polysaccharides have been reported, especially glycopeptides [85] and proteoglycans [86]. Reports on the pharmacological activity of Ganoderma polysaccharides mainly focus on anti-tumor effects, which are linked to immunomodulation, although other effects, such as regulation and protection of cells, have also been observed. Bioactive water-soluble polysaccharides have been isolated from the fruiting bodies and from the mycelial biomass cultivated in liquid culture. Few have been isolated from the culture medium. Some water insoluble anti-tumor polysaccharides were also identified [87]. Anti-tumor effects and immunological mechanisms In 1971 Sasaki et al. [88] reported on anti-tumor polysaccharides from some Polyporaceae, including G. applanatum. The study of anti-tumor effects of G. lucidum and their mechanisms had become matters of great interest. Since 1980s, numerous pharmacological investigations demonstrated that the hot water extract of G. lucidum polysaccharides inhibited the tumor growth in several tumor-bearing mice models [89–95], but the mechanism of the antitumor effect of G. lucidum was not well understood. A hypothesis was set up that G. lucidum polysaccharides exert anti-tumor activity by reinforcing body’s intrinsic immune-system defending functions. This hypothesis was used as a direction to guide further research to designing protocols and experiments to confirm its validity. Further research [96–99] demonstrated that in vivo water-soluble extract of G. lucidum and G. lucidum polysaccharides inhibited the growth of S-180 sarcoma and Lewis lung carcinoma implanted in mice. However, when added to the cultures of S-180 or HL-60 tumor cells, neither G. lucidum water extract nor G. lucidum polysaccharides inhibited the proliferation and induced apoptosis of the tumor cells, even at the very high concentration (400 mg/l) of G. lucidum polysaccharides, which indicated that either G. lucidum or its active fraction had no direct cytotoxicity on tumor cells. In the 1990s it became evident that Ganoderma polysaccharides indeed affect the immune system. A number of reports showed that polysaccharides stimulated immune functions both in vivo and in vitro, and that macrophages

284 were involved in this mechanism [100]. The proliferation of cancer cells was not affected by Ganoderma polysaccharides alone, but were significantly inhibited by the conditioned medium from polysaccharide-activated blood mononuclear cells [101–106]. Lei et al. [107] studied antagonistic effects of G. lucidum polysaccharides on the immunosuppressive response induced by cyclosporin A, hydrocortisone and anti-tumor agents, and reported on enhancement of cell-mediated immune functions and augmentation of cytokine production. Other immunomodulation effects, i.e., enhancement of unspecific immune functions, were reported [108], where Ganoderma polysaccharides affected intracellular free calcium and oxygen free radicals in murine peritoneal macrophages. Anti-tumor activity of cytotoxic drug such as cyclophosphamide was enhanced by oral administration of G. lucidum polysaccharides. G. lucidum polysaccharides induced HL-60 cells apoptosis by boosting macrophage activity [96,97]. G. lucidum polysaccharides (0.8, 3.2 and 12.8 mg/l) promoted not only the maturation of cultured murine bone marrow-derived dendritic cells (DC) through increasing co-expression of CD11c and I-A/I-E molecules on DC surface and enhancing IL-12 p40 production as well as its mRNA expression in DC, but also the immune response initiation induced by DC, such as cytotoxicity of specific cytotoxic T lymphocytes (CTL) induced by DC pulsed with P815 tumor antigen during the stage of antigen presentation by augmenting protein and mRNA expression of IFN-g and granzyme B [109–110]. Shao et al. [111] investigated immune receptors for polysaccharides from G. lucidum. The study, which was designed to identify and characterize the immune receptors for polysaccharides, demonstrated that G. lucidum polysaccharides activated BALB/c mouse B cells and macrophages, but not T cells, in vitro. Chien et al. [112] reported that a fucose-containing glycoprotein fraction, isolated from G. lucidum, increased the population and cytotoxicity of CD56+ NK-cell in human umbilical cord blood mononuclear cells. Lei and Lin [113–114] reported that G. lucidum promoted the mixed lymphocyte reaction (MLC), and its active polysaccharides increased the DNA synthesis of spleen cells in MLC through the enhancement of DNA polymerase induction in mice. The anti-tumor effect of G. lucidum polysaccharides was mediated by cytokines released from activated T lymphocytes and macrophages [102]. In vitro, G. lucidum polysaccharides induced B lymphocyte activation and proliferation [113,115–117]. However, studies on antibody production induced by G. lucidum were varied in different authors’ reports, in which G. lucidum polysaccharides having a backbone consisting of 1,4-linked a-D-glucopyranosyl residues and 1,6-linked b-D-galactopyranosyl residues with branches at O-6 of glucose residues and O-2 of galactose residues had little effect on serum IgG and complement (C3) levels after intraperitoneal

285 injection at a dose of 25 mg/kg for 4 days [116], and G. lucidum oral infusion significantly suppressed IgG1 and increased IgG2a production in mice with airway inflammation [118]. These works suggested that G. lucidum involved intrinsic immunological mechanism in the anti-tumor effect. Further studies confirmed that endogenous immunological mechanism play an important role in the anti-tumor effect of G. lucidum by serologic pharmacological method. After treatment with water extract from G. lucidum or G. lucidum polysaccharides by injection or oral infusion, mice serum was collected and tested. The results demonstrated that in vitro G. lucidum extract-treated serum inhibited proliferation of S-180 tumor cells and induced their apoptosis. Meanwhile, TNF-a and IFN-g level of G. lucidum extract-treated serum significantly increased [86,92]. Similarly, G. lucidum polysaccharides B-treated serum had the same effects in HL-60 cells [91,96,103]. Addition of G. lucidum polysaccharides to the culture of macrophage and T lymphocyte promoted TNF-a and IFN-g production and their mRNA expression in a dose-dependent manner [86,92]. Mononuclear cell conditioned media with G. lucidum polysaccharides (PSG-MNC-CM) containing TNF-a and IFN-g suppressed the proliferation clonogenicity of both the HL-60 and the U937 leukemic cell lines, induced their apoptosis and triggered their differentiation. But either G. lucidum polysaccharides alone or normal mononuclear cell-conditioned media without G. lucidum polysaccharides (MNC-CM) had no such effects even at higher dose of 400 mg/ml [102]. Antibody neutralization studies confirmed that adding antibody of anti-TNF-a or anti-IFN-g in the G. lucidum-treated serum markedly antagonize tumor-inhibiting effect of the G. lucidum-treated serum [119]. These findings revealed that G. lucidum inhibited tumor cells proliferation and induced their apoptosis via enhancing endogenous immune system function such as secreting anti-tumor cytokines TNF-a and IFN-g. G. lucidum polysaccharides synergize cytokines to induce immunological effector cells. Some reports demonstrated that G. lucidum polysaccharides could enhance the cytotoxicity of CTL and natural killer (NK) cells in mice and lymphokine-activated killer (LAK) cells derive from human cord blood [120–121]. Cytokine-induced killer (CIK) cells have been shown to generate effector cells with higher proliferative capacity, increased cytotoxicity and fewer side effects than LAK cells [122]. By synergizing cytokines, G. lucidum polysaccharides (400 mg/l or 100 mg/l) could decrease the amount of cytokine in lymphokine-activated killer (LAK) cells and CIK cells culture, but had no significant effect on the proliferation, cytotoxicity or phenotype of LAK cells and CIK cells induced by cytokines in higher doses alone. The activity of G. lucidum polysaccharides was relevant to enhancing IL-2, TNF production, protein and mRNA expression of granuzyme B and perforin in CIK cells, and mostly be blocked by anti-CR3 (complement receptor type 3), which suggested that the effect of G. lucidum polysaccharides on CIK cells was

286 possibly mediated primarily through complement receptor type 3 (CR3) [123–124]. Several other authors also reported on enhancement of cytokine production, both from macrophages and T lymphocytes, especially tumor necrosis factor-a (TNF-a) and interferon-g (IFN-g). Both G. lucidum and G. lucidum polysaccharides promoted proliferation of lymphocytes induced by concanavalin A or lipopolysaccharide, and potentiated interlukins (IL-1, IL-2, IL-3 and IL-6), tumor necrosis factor a (TNF-a) and interferon g (IFN-g) production and their mRNA expression in T lymphocyte and macrophage [86,92,102,115,125]. Chen et al. [126] studied the effect of G. lucidum polysaccharides on cytokine expression in mouse splenocytes. One of the fractions (F3) has activated the expression of IL-1, IL-6, IL-12, IFN-g, TNF-a, GM-CSF, G-CSF and M-CSF. Together with previous studies, the mode of action on macrophages was proposed, where F3 binds to TLR4 receptor and activates extracellular signal-regulated kinase, c-Jun N-terminal kinase and p38 to induce IL-1 expression. Boh et al. [104] reported on in vitro testing of G. lucidum extracellular and intercellular polysaccharide fractions on the induction of cytokine synthesis in primary cultures of human mononuclear cells from a buffy coat of healthy donors. Water-soluble polysaccharides, isolated from the mycelia produced by submerged cultivation in a stirred tank bioreactor, induced 3.0–630 pg/ml of TNF-a and 1.23–2.18 pg/ml of IFN-g. Zhu and Lin [124] studied the interaction between G. lucidum polysaccharides (Gl-PS) and cytokines, and explored mechanisms of Gl-PS acting on proliferation and anti-tumor activity of cytokine-induced killer (CIK) cells. The results suggested that 400 mg/ml or 100 mg/ml of Gl-PS promoted CIK cells proliferation and cytotoxicity were relevant to enhancing IL-2, TNF production, protein and mRNA expression of granzyme B and perforin in CIK cells through synergizing cytokines in decreasing doses of IL-2 and antiCD3 by 75% and 50%. Wang et al. [128] screened various G. lucidum strains and studied their anti-tumor and immunostimulation properties. The research on anti-proliferation of tumor cells of ethano1 extract and immunostimulation of water extract of G. lucidum was carrried out using K562 cells and macrophages. The results showed that tumor cells were inhibited by the ethanol extract, and the macrophages were activated by the water extract of G. lucidum. In the work of Zhang et al. [129] polysaccharide fractions from G. lucidum strains significantly inhibited the proliferation of leukemia cells. Effects of fractions of fruit body on stimulating the proliferation of spleen lymphocyte and T and B cells and activating the NK activity was stronger than that of fractions of mycelia. Fractions of fruit body and mycelia had similar effect on stimulating the proliferation of T and B cells in peripheral blood mononuclear cells. Fractions of fruit body and mycelia induced peripheral blood

287 mononuclear cells to release TNF-a in dose-dependent manner. At low concentration fractions of mycelia had similar capacity to induce the production of TNF-a with fractions of fruit body; however, at high concentration, fractions of mycelia were better than that of fruit body. The above studies confirmed that water extracts of G. lucidum and G. lucidum polysaccharides possess anti-tumor activity in vivo, but have no direct cytotoxic effect on tumor cells, which indicates that they exert antitumor activity mediated by intrinsic immunological mechanism involved in their immuno-modulatory action such as enhancing anti-tumor function of DC, CTL, promoting anti-tumor cytokines production and potentiating cytokines activity. Anti-tumor effects through immonomodulation and anti-angiogenesis Some polysaccharides and peptides of G. lucidum have shown anti-tumor effects by inhibiting angiogenesis. Investigations revealed that administrating intragastrically G. lucidum polysaccharides 50, 100, 200 mg/kg markedly inhibited xenograft (human lung carcinoma cell PG) in BALB/c immunedefault nude mice in vivo, and the G. lucidum polysaccharides-treated serum potently inhibited PG cell proliferation but not G. lucidum polysaccharides 0.1–100 mg/l alone in vitro. Because BALB/c nude mice have congenital T lymphocyte deficiency, and above G. lucidum polysaccharides could not affect macrophages function, there may be other anti-tumor mechanisms of G. lucidum polysaccharides except immunological mechanisms. A study by Cao and Lin [130,131] found out that G. lucidum polysaccharides and G. lucidum polysaccharides-treated serum have anti-angiogenic effect on chick chorioallantoic membrane. Further research confirmed that G. lucidum polysaccharides inhibited proliferation of human umbilical cord vascular endothelial (HUVEC) in a dose-dependent fashion, and decreased vascular endothelial growth factor (VEGF) secretion in human lung cancer cells in hypoxia, so anti-angiogenic effect may be the new anti-tumor mechanism of G. lucidum polysaccharides. Kimura et al. [76] found that the triterpenoid fraction of the fruit bodies of G. lucidum at the concentration of 800 mg/l inhibited angiogenesis induced by Matrigel – a soluble basement membrane extract of Engelbreth–Holm–Swam(EHS) tumor – supplemented with vascular endothelial growth factor (VEGF) and heparin in an in vivo model. Song et al. [132] studied anti-angiogenic and inhibitory activity on inducible nitric oxide production of the 70% ethanol extract of G. lucidum fresh fruiting bodies. The extract showed significant anti-angiogenic activity, which was detected using a chick embryo chorioallantoic membrane assay. Stanley et al. [133] examined the effect of G. lucidum on angiogenesis related to prostate cancer and found that G. lucidum inhibited the early event in angiogenesis, capillary morphogenesis of the human aortic endothelial cells. Cao and Lin [131] reported that G. lucidum polysaccharides peptide inhibited the growth

288 of vascular endothelial cell and the induction of vascular endothelial growth factor (VEGF) in human lung cancer cells. The proliferation of human umbilical cord vascular endothelial cell (HUVEC) culture was inhibited by Gl-PP in a dose-dependent fashion, but not because of cytotoxicity. Flow cytometric studies revealed that Gl-PP treatment of HUVECs could induce cell apoptosis directly. Therefore, inducing cell apoptosis by Gl-PP might be the mechanism of inhibiting HUVEC proliferation. Human lung carcinoma cells PG when exposed to high dose of Gl-PP in hypoxia for 18 h resulted in a decrease in the secreted VEGF. These findings supported the hypothesis that the key attribute of the anti-angiogenic potential of Gl-PP is that it may directly inhibit vascular endothelial cell proliferation or indirectly decrease growth factor expression of tumor cells. Regulation and protection of cells Several works suggested that G. lucidum also positively affects and protects living cells. Cao and Lin [106,109] found out that G. lucidum polysaccharides had an effect on regulation of maturation and function of dendritic cells. You and Lin et al. [134] reported on protective effects of G. lucidum glycopeptides on injury of macrophages induced by reactive oxygen species. Shi et al. [135] studied aqueous extracts of eight mushroom species and found out that G. lucidum had a potential for protecting cellular DNA from oxidative damage. Zhang et al. [136] reported on in vitro and in vivo protective effect of G. lucidum polysaccharides on pancreatic islets damage induced by alloxan. The effect was dose-dependent and visible as increased serum insulin and reduced serum glucose levels when pretreated intragastrically for 10 days in alloxan-induced diabetic mice. The mechanism was based on polysaccharide scavenging ability to protect the pancreatic islets from free radicals. It was found that the pancreas homogenates had higher lipid peroxidation products in alloxan-treated mice than in animals treated with G. lucidum polysaccharides. In a study by Lakshmi et al. [137] methanolic extract of G. lucidum had a dose-dependent protective effect on hepatic cells damage, caused by benzo[a]pyrene, and a significant anti-mutagenic activity in vivo in rats by restoring anti-oxidant defense. Peptidoglycanes and proteins One of the oldest known proteins isolated from G. lucidum is LZ-8, for which immunomodulatory and immunosuppressive activities were reported [138]. From G. lucidum mycelia produced by submerged fermentation, Tian and Zhang [139] purified and characterized a proteinase A inhibitor with a molecular mass of 38 kDa. The purification was carried out by ethanol precipitation (50–80%), ACA44 gel filtration and Source 30Q anion

289 exchange. Its carbohydrate content was about 70%. The linkage between the glycan and the core protein backbone might be O-linkage. By investigating the interaction between the inhibitor and a variety of proteinases, it was indicated that the inhibitor was more specific against yeast proteinase A than other proteinases. The inhibitor showed a remarkable heat stability. A bioactive fraction (GLPG) was extracted and purified from the mycelia of G. lucidum by EtOH precipitation and DEAE-cellulose column chromatography by Liu et al. [140]. GLPG was a proteoglycan with a carbohydrate:protein ratio of 10.4:1. The product had anti-virus effects. Possible mode of action of antitherpetic activities of a proteoglycan isolated from the mycelia of G. lucidum was studied in vitro. Its antiviral activities against herpes simplex virus type 1 and type 2 were investigated by the cytopathic effect inhibition assay in cell culture. Although the precise mechanism was not defined yet, the work suggested that GLPG inhibited viral replication by interfering with the early events of viral adsorption and entry into target cells. Thus, the GLPG proteoglycan is a potential candidate for anti-herpes simplex virus agents. A ribonuclease with a molecular mass of 42 kDa and with an N-terminal sequence distinct from other mushroom ribonucleases was isolated from fresh fruiting bodies of G. lucidum by Wang et al. [141]. In the purification process, the ribonuclease was adsorbed on DEAE-cellulose and Q-Sepharose, and unadsorbed on CM-Sepharose. The optimum pH of 4.0 was low compared with those reported for other mushroom ribonucleases. A temperature of 601C was required for optimal enzyme activity. The ribonuclease was unique among mushroom ribonucleases in that it exhibited the highest potency toward poly(U), followed by poly(A). Its activity toward poly(G) and poly(C) was about one-half of that toward poly(A) and one-quarter of that toward poly(U). Wang et al. [142] isolated a 15-kDa protein from fruiting bodies of G. lucidum, named ganodermin. The isolation procedure utilized chromatography on DEAE-cellulose, Affi-gel blue gel, CM-Sepharose and Superdex 75. Ganodermin expressed an anti-fungal activity by inhibiting the mycelial growth of Botrytis cinerea, Fusarium oxysporum and Physalospora piricola.

Other compounds Polysaccharides and triterpenes have been most thoroughly investigated from G. lucidum and related species. However, other active compounds have also been described, such as adenosine with anti-platelet aggregation effect, lectins with mitogenic effect, alkaloids, fatty acids, vitamins and essential minerals. A structured list of chemical constituents of G. lucidum was given in [87].

290 Formulations, market products and clinical trials Several formulations have been developed, patented and used as nutraceuticals, nutriceuticals and pharmaceuticals [143], mainly with Ganoderma fruiting bodies, spores and their water or ethanol extracts, rarely with purified active compounds. Several products have undergone clinical trials and became available commercially as a syrup, injection, tablet, tincture, or bolus of powdered medicine and additives [144]. For example, Zhang and Li [145] reported on a clinical investigation of Green Valley Lingzhi capsule in treatment of 130 patients suffering from type 2 diabetes mellitus. After 2 months of a treatment, the formulation showed a synergistic effect in hypoglycemic action combined with regular hypoglycemic drugs, and significantly decreased the clinical symptoms, compared to a control group treated only with regular hypoglycemic drugs. Shi and Qing [146] published the results of a clinical observation assessment of 547 medium and late phases cancer patients treated by a formulation Chinese G. lucidum Essence. The study showed that the death rate of the patients in the long-term treatment was significantly lower. A continuous 2–3-month active treatment with a daily dosage of 4–6 g of G. lucidum essence was proposed, with further dosage of 2 g/day continuously after the third month of therapy. Short-term treatments were less successful. Sliva et al. [147] reported that G. lucidum spores and unpurified fruiting body inhibited invasion of breast and prostate cancer cells by a common mechanism and could have potential therapeutic use for cancer treatment. In their study, they investigated the effect of G. lucidum on highly invasive breast and prostate cancer cells. Spores or dried fruiting body inhibited constitutively active transcription factors AP-1 and NF-jB in breast MDA-MB231 and prostate PC-3 cancer cells. Furthermore, Ganoderma inhibition of expression of uPA and uPA receptor (uPAR), as well as secretion of uPA, resulted in the suppression of the migration of MDA-MB-231 and PC-3 cells. Noguchi et al. [148] performed a phase I study of a methanol extract of G. lucidum in men with mild symptoms of bladder outlet obstruction. Male volunteers of age 450 were enrolled. The overall administration was well tolerated with no adverse effects. Statistically significant reductions in International Prostate Symptoms Score (I-PSS) versus placebo were observed at the 6 mg and 60 mg dose. The study concluded that the extract of G. lucidum was well tolerated, a significant improvement in I-PSS was observed and a 6 mg dose of the extract was recommended for the phase II trial. A study by Chen et al. [149] evaluated the effects of G. lucidum polysaccharides on patients with advanced colorectal cancer. Forty-seven patients were enrolled and treated with oral G. lucidum at 5.4 g/day for 12 weeks. Selected immune parameters were monitored using various immunological methods throughout the study. In 41 assessable cancer patients, treatment

291 with G. lucidum tended to increase mitogenic reactivity to phytohemagglutinin, counts of CD3, CD4, CD8 and CD56 lymphocytes, plasma concentrations of interleukin (IL)-2, IL-6 and interferon (IFN)-g, and NK activity, whereas plasma concentrations of IL-1 and tumor necrosis factor (TNF)-a were decreased. For all of these parameters, no statistical significance was observed when a comparison was conducted between baseline and those values after a 12-week treatment with G. lucidum. The changes of IL-1 were correlated with those for IL-6, IFN-g, CD3, CD4, CD8 and NK activity (po0.05) and IL-2 changes were correlated with those for IL-6, CD8 and NK activity. The results indicated that G. lucidum may have potential immunomodulating effect in patients with advanced colorectal cancer; however, further studies are needed to explore the benefits and safety of G. lucidum in cancer patients. Clinical studies also showed that G. lucidum preparations exerted synergistic therapeutic effect when used in conjunction with radiation and chemotherapy, reduced the following side effects: leukopenia, thrombocytopenia, anemia, nausea, vomiting, appetite loss, anti-infection deficiency and immunosuppression, enhanced tolerance for radiation and chemotherapy, constitution and immunity in cancer patients to potentiate the therapeutic efficacy and ameliorate adverse toxicity of radiation and chemotherapy [149–153]. The therapeutic efficacy of G. lucidum on cancer patients is mediated not only by immuno-modulation, such as enhancing function of DC, CTL and other immunological effector cells to kill tumor cells, and promoting anti-tumor cytokines production and activity, but also by antiangiogenesis. G. lucidum abates or avoids the toxicity caused by other therapies via stimulating hematopoiesis and antagonizing damage resulted from radiation and chemotherapy. It is obvious that the effect of G. lucidum on radiation and chemotherapy exactly compensates for the deficit of the two therapies by strengthening internal immune function to resist external malign factors caused by tumor or toxicity of radiation and chemotherapy. Although preparations of G. lucidum have been commonly used in China as self-medication for the prevention and treatment of various medical diseases including cancer, liver diseases, hypertension, hyperlipidemia and coronary disease, there were, until recently, no reports on their side effects. However, in 2004, Yuen et al. [127] reported on a case of hepatotoxicity, probably due to a formulation of G. lucidum. A patient was a 78-year-old Chinese lady in whom an intake of a formulation of G. lucidum powder caused a significant hepatotoxicity in which the liver biochemistry mimiced that of acute cholangitis. The patient took regular self-medications including oral calcium and multivitamin tablets daily. She also had regular intake of G. lucidum as a health supplement for at least 1 year, but she started to take a new powder formulation 4 weeks before the onset of symptoms. The liver toxicity was therefore most likely due to the ingredients of the powder G. lucidum formulation.

292 Conclusion Reports on isolated compounds from G. lucidum are very convincing; there is abundant evidence that triterpenoids, polysaccharides and proteoglycans are effective. In most cases, extracts of partly-purified preparations have been used for in vitro or in vivo testing. Synergistic effects of mixtures of active components have been known; however, their biological activities need further assessment before they can be accepted not only by the traditional Asian medicine, but also by the Western science and medicine. Modern biotechnological cultivation methods in bioreactors enable fast, efficient and economical production of G. lucidum biomass in sufficient quantities for potential future pharmaceutical industrial production. References 1.

Leung SWS. Lingzhi (Ganoderma) research – the past, present and future perspectives. In: Ganoderma: Genetics, Chemistry, Pharmacology and Therapeutics, Zhi-Bin Lin (ed), Proceedings of International Symposium on Ganoderma Research, Shanghai, October 21–23, Beijing, Medical University Press, 2002, pp. 1–9. 2. Kim HW and Kim BK. Recent advances on the biologically active triterpenoids of Ganoderma lucidum. In: Ganoderma: Genetics, Chemistry, Pharmacology and Therapeutics, Zhi-Bin Lin (ed), Proceedings of International Symposium on Ganoderma Research, Shanghai, October 21–23, Beijing, Medical University Press, 2002, pp. 10–19. 3. Jong SC and Birmingham JM. Medicinal benefits of the mushrooms Ganoderma. Adv Appl Microbiol 1992;37:101–134. 4. Tasaka K, Akagi M, Miyoshi K, Mio M and Makino T. Anti-allergic constituents in the culture medium of Ganoderma lucidum. 1. Inhibitory effect of oleic acid on histamine release. Agents Action 1988;23:153–156. 5. Kino K, Sone T, Watanabe J, Yamashita A, Tsuboi H, Miyajima H and Tsunoo H. Immunomodulator, LZ-8, prevents antibody-production in mice. Int J Immunopharmacol 1991;13:1109–1115. 6. van Der Hem LG, Van Der Vliet J A, Bocken C F M, Kino K, Hoitsma A J and Tax WJM. Ling zhi-8 – studies of a new immunomodulating agent. Transplantation (Baltimore) 1995;60:438–443. 7. Maruyama H, Yamazaki K, Murofushi S, Konda C and Ikekawa T. Antitumor activity of Sarcodon aspratus (Berk.) S. Ito and Ganoderma lucidum (Fr.) Karst. J PharmacobioDyn 1989;12:118–123. 8. Lee SY and Rhee H M. Cardiovascular effects of mycelium extract of Ganoderma lucidum – inhibition of sympathetic outflow as a mechanism of its hypotensive action. Chem Pharm Bull 1990;38:1359–1364. 9. Liu GT, Bao X, Niu S, Li Z and Sung Z. Some pharmacological actions of the spores of Ganoderma lucidum and the mycelium of Ganoderma capense (Lloyd) Teng cultivated by submerged fermentation. Chin Med J 1979;92:469–500. 10. Hirose K, Muto S, Niimura K, Ohara M, Oguchi Y, Matsunaga K, Kadochi J, Sugita N, Furushu T, Yoshikumi C and Takahashi M. Antiviral agent. Japanese Patent JP 63316734, 1988.

293 11. Mizumoto K, Yamashita A, Kii M and Sumio H. Antiretrovirus agent. Japanese Patent JP 2032026, 1990. 12. Kim HW, Shim MJ, Choi EC and Kim BK. Inhibition of cytopathic effect of human immunodeficiency virus-1 by water-soluble extract of Ganoderma lucidum. Arch Pharm Res 1997;20:425–431. 13. Hattori M, El-Mekkawy S and Meselhy R. Inhibitory effects of components from Ganoderma lucidum on the growth of human immunodeficiency virus (HIV) and the protease activity. In: Proceedings of the 1st International Symposium on Ganoderma Lucidum in Japan, Mizuno T, Ide N and Hasegawa Y (eds), November 17–18, 1997, pp. 128–135. 14. El-Mekkawy S, Meselhy M R, Nakamura N, Tezuka Y, Hattori M, Kakiuchi N, Shimotohno K, Kawahata T and Otake T. Anti-HIV-1 and anti-HIV-1 protease substances from Ganoderma lucidum. Phytochemistry 1998;49:1651–1657. 15. Min BS, Nakamura N, Miyashiro H, Bae K W and Hattori M. Triterpenes from the spores of Ganoderma lucidum and their inhibitory activity against HIV-1 protease. Chem Pharm Bull 1998;46:1607–1612. 16. Triratana S, Thaithatgoon S and Gawgla M. Cultivation of Ganoderma lucidum in sawdust bags. In: Science and Cultivation of Edible Fungi, Maher MJ (ed), Proceedings of the 13th International Congress on the Science Cultivation of Edible Fungi, Dublin, September 1–6, 1991, Rotterdam, A. A. Balkema, pp. 567–572. 17. Kohlmu¨nzer S, Wegiel J and Sitarz J. Polysaccharides in mycelial culture of Ganoderma applanatum (Pers.) Pat. Herba Hung 1989;28:87–94. 18. Lin SQ, Wang SZ, Lin SG and Lin ZB. Studies on Ganoderma submerged fermentation and its product extraction technique. In: Ganoderma: Genetics, Chemistry, Pharmacology and Therapeutics, Zhi-Bin Lin (ed), Proceedings of International Symposium on Ganoderma Research, Shanghai, October 21–23, 2002, Beijing, Medical University Press, pp. 93–97. 19. Habijanic J and Berovic M. The relevance of solid-state substrate moisturing on Ganoderma lucidum biomass cultivation. Food Technol Biotechnol 2000;38:225–228. 20. Berovic M, Habijanic J, Zore I, Wraber B, Hodzar D, Boh B and Pohleven F. Submerged cultivation of Ganoderma lucidum biomass and immunostimulatory effects of fungal polysaccharides. J Biotechnol 2003;103:77–86. 21. Mizuno T, Wang G, Zhang J, Kawagishi H, Nishitoba T and Li J. Reishi, Ganoderma lucidum and Ganoderma tsugae: bioactive substance and medicinal effects. Food Rev Int 1995;11:151–166. 22. Miyahara R, Yoshimoto T and Asawa K. Chemical structures and changes of extracts during growth of reishi (Ganoderma lucidum). Mokuzai Gakkaishi 1987;33:416–422. 23. Chan K. Linking chemical and biological characteristics in assuring the quality of Chinese medicinal materials and OTC products. In: L Ganoderma: Genetics, Chemistry, Pharmacology and Therapeutics, Zhi-Bin Lin (ed), Proceedings of International Symposium on Ganoderma Research, Shanghai, October 21–23, 2002, Beijing, Medical University Press, pp. 24–34. 24. Lee JH. Cultivation of reishi (Ganoerma lucidum) [online]. MushWorld – Cultivation, 2000-12-24, available at http://www.mushworld.com/sub_en.html 25. Chen AW. A fresh look at an ancient mushroom Ganoderma lucidum (Reishi) [online]. MushWorld – Cultivation, 2003-03-18, available at http://www.mushworld.com/ sub_en.html 26. Chen AW. Natural-log cultivation of the medicinal mushroom Ganoderma lucidum (Reishi) [online]. MushWorld – Cultivation, 2004-01-09, available at http://www.mushworld.com/ sub_en.html

294 27. Sukarno N. Development of Ganoderma lucidum on soft and hard wood logs and determination of organic germanium and ganoderic acid content of the fruiting body produced [online]. MushWorld – Cultivation, 2004-10-06, available at http:// www.mushworld.com/sub_en.html 28. Chen AW. Cultivation of the medicinal mushroom Ganoderma lucidum (Curt.: Fr.) P. Karst (Reishi) in North America (3) [online]. MushWorld – Cultivation, 2002-02-01, available at http://www.mushworld.com/sub_en.html 29. Royse DJ. Specialty mushrooms. In: Progress in New Crops, Janick J (ed), Arlington, VA, ASHS Press, 1996, pp. 464–475. 30. Chen AW. Growing Ganoderma mushrooms. In: Mushroom Grower’s Handbook I, Part III: Mushrooms Worldwide, Chapter 11: Mushrooms for the tropics. ISSN 17391377, [online]. MushWorld, 2004, pp. 224–234, available at http://www.mushworld.com/ service/handbook/english/eng-book1/chapter11-01_p.224.pdf 31. Yukinori S, Tetsuji S and Yasushi S. Cultivation of Ganoderma lucidum Karst. Patent JP11146728, JMC KK, 1999-06-02. 32. Furci George-Nascimento GM. Ganoderma lucidum (Curt.:Fr.) P. Karst grown indoors on native Chilean hardwoods [online]. MushWorld – Cultivation, 2005-07-30, available at http://www.mushworld.com/sub_en.html 33. Gonzalez-Matute R, Figlas D, Devalis R, Delmastro S and Curvetto N. Sunflower seed hulls as a main nutrient source for cultivating Ganoderma lucidum. Micologia Aplicada Int. 2002;14:19–24. 34. Yang FC, Hsieh C and Chen HM. Use of stillage grain from a rice-spirit distillery in the solid state fermentation of Ganoderma lucidum. Process Biochem 2003;39:21–26. 35. Hsieh C and Yang FC. Reusing soy residue for the solid-state fermentation of Ganoderma lucidum. Bioresour Technol 2004;91:105–109. 36. Kim HK. Comparison of characteristics of Ganoderma lucidum according to geographical origins: consideration of growth characteristics (1) [online]. MushWorld – World Mushroom, 2001-09-01, available at http://www.mushworld.com/sub_en.html 37. Takashi M. Apparatus for cultivating Ganoderma lucidum Karst and other mushrooms and their cultivation. Patent JP11155366, 1999. 38. Kiyoshi K, Hamajirou S and Yoshiaki A. Cultivation method for Ganoderma lucidum (fr.) Karst. Patent JP10084772, 1998. 39. Shigeru Y. Cultivation of Reishi. Patent JP3083521, 1991. 40. Habjanic J and Berovic M. Process of cultivation of fungus Ganoderma lucidum on a solid cultivation substrate. Patent SI 20923, 2002. 41. Li PJ and Shen CG. Method for propagating fungi using solid state fermentation. Patent WO0220727, 2002. 42. Chang MY, Tsai GJ and Houng JY. Optimization of the medium composition for the submerged culture of Ganoderma lucidum by Taguchi array design and steepest ascent method. Enzyme Microb Technol 2006;38:407–414. 43. Hsieh C, Tseng MH and Liu CJ. Production of polysaccharides from Ganoderma lucidum (CCRC 36041) under limitations of nutrients. Enzyme Microb Technol 2006;38:109–117. 44. Zore I, Berovic M, Boh B, Hodzar D and Pohleven F. Procedure for preparation of inoculum for growing of fungus Ganoderma lucidum by submersion fermentation. Patent SI 9700014, 1998. 45. Yang FC and Liau CB. The influence of environmental conditions on polysaccharide formation by Ganoderma lucidum in submerged cultures. Process Biochem 1998;33:547–553.

295 46. Lee KM, Lee SY and Lee HY. Bistage control of pH for improving exopolysaccharide production from mycelia of Ganoderma lucidum in an air-lift fermentor. J Biosci Bioeng 1999;88:646–650. 47. Fang QH and Zhong JJ. Effect of initial pH on production of ganoderic acid and polysaccharide by submerged fermentation of Ganoderma lucidum. Process Biochem 2002;37:769–774. 48. Fang QH and Zhong JJ. Submerged fermentation of higher fungus Ganoderma lucidum for production of valuable bioactive metabolites – ganoderic acid and polysaccharide. Biochem Eng J 2002;10:61–65. 49. Fang QH, Tang Yj and Zhong JJ. Significance of inoculation density control in production of polysaccharide and ganoderic acid by submerged culture of Ganoderma lucidum. Process Biochem 2002;37:1375–1379. 50. Tang YJ and Zhong JJ. Fed-batch fermentation of Ganoderma lucidum for hyperproduction of polysaccharide and ganoderic acid. Enzyme Microb Tech 2002;31:20–28. 51. Tang YJ and Zhong JJ. Role of oxygen supply in submerged fermentation of Ganoderma lucidum for production of Ganoderma polysaccharide and ganoderic acid. Enzyme Microb Technol 2003;32:478–484. 52. Hsieh C, Hsu TH and Yang FC. Production of polysaccharides of Ganoderma lucidum (CCRC36021) by reusing thin stillage. Process Biochem 2005;40:909–916. 53. Lee H, Song M, Yu Y and Hwang S. Production of Ganoderma lucidum mycelium using cheese whey as an alternative substrate: response surface analysis and biokinetics. Biochem Eng J 2003;15:93–99. 54. Lee H, Song M, Yu Y and Hwang S. Optimizing bioconversion of deproteinated cheese whey to mycelia of Ganoderma lucidum. Process Biochem 2003;38:1685–1693. 55. Hirotani M, Ino C, Furuya T and Shiro M. Ganoderic acids T, S, and R, new triterpenoids from the cultured media of Ganoderma lucidum. Chem Pharm Bull 1986;34:2282–2285. 56. Kikuchi T, Kanomi S, Murai Y, Kadota S, Tsubono K and Ogita Z. Constituents of the fungus Ganoderma lucidum (Fr.) Karst. I. Structures of ganoderic acids C2, E, I, and K, lucidenic acid F and related compounds. Chem Pharm Bull 1986;34: 3695–3712. 57. Morigiwa A, Kitabatake K, Fujimoto Y and Ikekawa N. Angiotensin converting enzyme inhibitory triterpenes from Ganoderma lucidum. Chem Pharm Bull 1986;34:3025–3028. 58. Nishitoba T, Sato H and Sakamura S. Triterpenoids from the fungus Ganoderma lucidum. Phytochemistry 1987;26:1777–1784. 59. Nishitoba T, Goto S, Sato H and Sakamura S. Bitter triterpenoids from the fungus Ganoderma applanatum. Phytochemistry 1989;28:193–197. 60. Lin LJ and Shiao MS. Seven new triterpenes from Ganoderma lucidum. J Nat Prod 1988;51:918–924. 61. Lin LJ, Shiao MS and Yeh SF. Triterpenes from Ganoderma lucidum. Phytochem 1988;27:2269–2271. 62. Shiao MS, Lin LJ, Yeh SF and Chou CS. Two new triterpenes of the fungus Ganoderma lucidum. J Nat Prod 1987;50:886–890. 63. Chairul, Tokuyama T, Hayashi Y, Nishizawa M, Tokuda H, Chairul SM and Hayashi Y. Applanoxidic acids A, B, C and D, biologically active tetracyclic triterpenes from Ganoderma applanatum. Phytochem 1991;30:4105–4109. 64. Chairul, Hayashi Y and Chairul SM. Lanostanoid triterpenes from Ganoderma applanatum. Phytochem 1994;35:1305–1308.

296 65. Boh B, Hodzar D, Dolnicar D, Berovic M and Pohleven F. Isolation and quantification of triterpenoid acids from Ganoderma applanatum of Istrian origin. Food Technol Biotechnol 2000;1:11–18. 66. Kim DH, Shim SB, Kim NJ and Jang IS. Beta-glucuronidase-inhibitory and hepatoprotective effect of Ganoderma lucidum. Biol Pharm Bull 1999;22:162–164. 67. Toth JO, Luu B and Ourisson G. Les acides ganoderiques T a Z: triterpenes cytotoxiques de Ganoderma lucidum (Polyporaceae). Tetrahedron Lett 1983;24: 1081–1084. 68. Lin SB, Li CH, Chen YR, Kan LS and Lee SS. Triterpene extract from Ganoderma lucidum inhibits growth of hepatoma Huh7 cells: involvement of oxidative stress induction. In: Ganoderma: Genetics, Chemistry, Pharmacology and Therapeutics, Zhi-Bin Lin (ed), Proceedings of International Symposium on Ganoderma Research, Shanghai, October 21–23, 2002, Beijing, Medical University Press, pp. 176–182. 69. Gao JJ, Min BS, Ahn EM, Nakamura N, Lee HK and Hattori M. New triterpene aldehydes, lucialdehydes A-C, from Ganoderma lucidum and their cytotoxicity against murine and human tumor cells. Chem Pharm Bull (Tokyo) 2002;50:837–840. 70. Min BS, Gao JJ, Nakamura N and Hattori M. Triterpenes from the spores of Ganoderma lucidum and their cytotoxicity against meth-A and LLC tumor cells. Chem Pharm Bull (Tokyo) 2000;48:1026–1033. 71. Lin SB, Li CH, Lee SS and Kan LS. Triterpene-enriched extracts from Ganoderma lucidum inhibit growth of hepatoma cells via suppressing protein kinase C, activating mitogen-activated protein kinases and G2-phase cell cycle arrest. Life Sci 2003;72:2381–2390. 72. Zhu HS, Yang XL, Wang LB, Zhao DX and Chen L. Effects of extracts from sporoderm-broken spores of Ganoderma lucidum on HeLa cells. Cell Biol Toxicol 2000;16:201–206. 73. Yu SP, Zhang JS, Tang QJ, Shi XM, Liu YF, Yan Y and Pan YL. Correlation between intracellular triterpenes from mycelia of Ganoderma lucidum in different growth stages and inhibition effect on tumor cells. Mycosystema 2004;23:548–554. 74. Liu J, Shimizu K, Konishi F, Noda K, Kumamoto S, Kurashiki K and Kondo R. Antiandrogenic activities of the triterpenoids fraction of Ganoderma lucidum. Food Chem 2007;100:1691–1696. 75. Mueller CI, Kumagai T, Kelly JO, Seeram NP, Heber D and Koeffler HP. Ganoderma lucidum causes apoptosis in leukemia, lymphoma and multiple myeloma cells. Leukemia Res 2006;30:841–848. 76. Kimura Y, Taniguchi M and Baba K. Antitumor and antimetastatic effects on liver of triterpenoid fractions of Ganoderma lucidum: mechanism of action and isolation of an active substance. Anticancer Res 2002;22(6A):3309–3318. 77. Kabir Y, Kimura S and Tamura T. Dietary effect of Ganoderma lucidum mushroom on blood pressure and lipid levels in spontaneously hipertensive rats. J Nat Sci Vitaminol 1988;34:433–438. 78. Kohda H, Tokumoto W, Sakamoto K, Fujii M, Hirai Y, Yamasaki K, Komoda Y, Nakamura H, Ishihara S and Uchida M. The biologically active constituents of Ganoderma lucidum (Fr.) Karst histamine release-inhibitory triterpenes. Chem Pharm Bull 1985;33:1367–1374. 79. Wang CN, Chen JC, Shiao MS and Wang CT. The aggregation of human platelet induced by ganodermic acid S. Biochim Biophys Acta 1989;986:151–160. 80. Wang CN, Chen JC, Shiao MS and Wang CT. The inhibition of human platelet function by ganodermic acids. Biochem J 1991;277:189–197.

297 81. Su CY, Shiao MS and Wang CT. Predominant inhibition of ganodermic acid S on the thromboxane A2 -dependent pathway in human platelets response to collagen. Biochim Biophys Acta 1999;1437:223–234. 82. Su C, Shiao M and Wang C. Potentiation of ganoderemic acid S on prostaglandin E(1)induced cyclic AMP elevation in human platelets. Thromb Res 2000;99:135–145. 83. Min BS, Gao JJ, Hattori M, Lee HK and Kim YH. Anticomplement activity of triterpenoids from the spores of Ganoderma lucidum. Planta Med 2001;67:811–814. 84. Fang JN, Bao XF and Yuen WH. Studies on the polysaccharides from spores of Ganoderma lucidum. In: Ganoderma: Genetics, Chemistry, Pharmacology and Therapeutics, Zhi-Bin Lin (ed), Proceedings of International Symposium on Ganoderma Research, Shanghai, October 21–23, 2002, Beijing, Medical University Press, pp. 98–103. 85. Lin SQ, Wang SZ, Lin ZB and Lin YX. Purification and identification of glycopeptides from Ganoderma lucidum fruit bodies cultivated with grass and wood log. In: Ganoderma: Genetics, Chemistry, Pharmacology and Therapeutics, Zhi-Bin Lin (ed), Proceedings of International Symposium on Ganoderma Research, Shanghai, October 21–23, 2002, Beijing, Medical University Press, pp. 109–114. 86. Zhang QH and Lin ZB. Effect of Ganoderma lucidum polysaccharides B on TNF-a and INF-g production and their mRNA expression. J Beijing Med Univ 1999;31:179–183. 87. Russell R and Paterson M. Ganoderma – A therapeutic fungal biofactory. Phytochem 2006;67:1985–2001. 88. Sasaki T, Arai Y, Ikekawa T, Chihara G and Fukuoka F. Antitumor polysaccharides from some Polyporaceae, Ganoderma applanatum (Pers.) Pat. and Phellinus linteus (Berk. Et Curt) Aoshima. Chem Pharm Bull 1971;19:821–826. 89. Lee SS, Chen FD, Chang SC, Wei YH, Liu I and Chen C. In vivo antitumor effect of crude extracts from the mycelium of Ganoderma lucidum. Bull Chinese Oncol Soc 1984;5:22–27. 90. Hwang SF, Liu KJ, Kuan YH, Tung KS, Su CH and Tung TC. The inhibitory effect on artificial pulmonary metastasis of murine S-180 Sarcoma cells by orally administered Ganoderma lucidum. J Chim Oncol Soc 1989;5:10–15. 91. Furusawa E, Chou SC, Furasawa S, Hirazum A and Dang Y. Antitumor activity of Ganoderma lucidum, and edible mushroom, on intraperitoneally implanted Lewis lung carcinoma in synergeneic mice. Phytother Res 1992;6:300–304. 92. Zhang Q and Lin ZB. The antitumor activity of Ganoderma lucidum (Curt,:Fr.) P. Karst. (Lingzhi) (Aphylophoromycetidae) polysaccharides is related to tumor necrosis factor-a and interferon-g. Int J Med Mushroom 1999;1:207–215. 93. Lu H, Kyo E, Uesaka T, Katoh O and Watanabe H. Prevention of development of N,N0 -dimethylhydrazine-induced colon tumors by a water-soluble extract from cultured medium of Ganoderma lucidum (Rei-shi) mycelia in male ICR mice. Int J Mol Med 2002;9:113–117. 94. Liu X, Yuan JP, Chuang CK and Chen XJ. Antitumor activity of the sporodermbroken germinating spores of Ganoderma lucidum. Cancer Lett 2002;182:155–161. 95. Lu H, Kyo E, Uesaka T, Katoh O and Watanabe H. A water-soluble extract from cultured medium of Ganoderma lucidum (Rei-shi) mycelia suppresses azoxymethaneinduction of colon cancers in male F344 rats. Oncol Rep 2003;10:375–379. 96. Hu YH and Lin ZB. Polysaccharides isolated from mycelia of Ganoderma lucidum induced HL-60 cell apoptosis by enhancing macrophage activity. Chin Pharmacol Bull 1999;5:27–30. 97. Hu YH and Lin ZB. Effects of polysaccharides isolated from mycelia of Ganoderma lucidum on HL-60 cell apoptosis. Acta Pharm Sin 1999;34:268–271.

298 98. Zhang NQ.. Basic Theory of Traditional Chinese Medicine, 1st edn, Vol. I, Beijing, People Sanitation Press, 1990. pp. 199–205. 99. Lin ZB. Modern Research of Ganoderma lucidum, 2nd edn, Beijing, Beijing Medical University Press, 2001 (Chapter 1 and 8). 100. Lei LS, Li MC and Sun LS. Ganoderma lucidum and its components on the function of macrophages. In: Ganoderma: Genetics, Chemistry, Pharmacology and Therapeutics, Zhi-Bin Lin (ed), Proceedings of International Symposium on Ganoderma Research, Shanghai, October 21–23, 2002, Beijing, Medical University Press, pp. 20–23. 101. Shiuh S, Yau H, Chieh F, Sheng Y and Kuang Y. Antitumor effects of Ganoderma lucidum. J Chin Med 1995;6:1–12. 102. Wang SY, Hsu ML, Hsu HC, Tzeng CH, Lee SS, Shiao MS and Ho CK. The antitumor effect of Ganoderma lucidum is mediated by cytokines released from activated macrophages and T lymphocyes. Int J Cancer 1997;70:699–705. 103. Zhang QH and Lin ZB. The antitumor activity of Ganoderma lucidum (Curt; Fr.) P. Karst (Ling Zhi) (Aphyllophoromycetideae) polysaccharides is related to tumor necrosis factor-a and interferon-g. Int J Med Mushroom 1999;1:207–215. 104. Boh B, Berovic M, Wraber B, Hodzar D, Habijanic J, Pohleven F and Zore I. Ganoderma lucidum (W.Curt.:Fr.) Lloyd and G. applanatum (Pers.) Pat. (Aphyllophoromycetideae) from Slovenian habitats: cultivation, isolation, and testing of active compounds. Int J Medic Mushroom 2004;6:15–32. 105. Lee SS, Lee PL, Chen CF, Wang SY and Chen KY. Antitumor effects of Ganoderma lucidum. In: Ganoderma: Genetics, Chemistry, Pharmacology and Therapeutics, Zhi-Bin Lin (ed), Proceedings of International Symposium on Ganoderma Research, Shanghai, October 21–23, 2002, Beijing, Medical University Press, pp. 46–57. 106. Cao LZ and Lin ZB. Regulation of Ganoderma lucidum polysaccharides on cytotoxic T lymphocytes induced by dendric cells in vitro. In: Ganoderma: Genetics, Chemistry, Pharmacology and Therapeutics, Zhi-Bin Lin (ed), Proceedings of International Symposium on Ganoderma Research, Shanghai, October 21–23, 2002, Beijing, Medical University Press, pp. 122–129. 107. Lei L, Lin Z, Chen Q, Li R and He Y. Antagonistic effect of Ganoderma lucidum polysaccharide on the immunosuppressive response induced by cyclosporin A, hydrocortisone and antitumor agents. Chin J Pharmacol Toxicol 1993;7:183–185. 108. Li M, Lei L and Liang D. Effect of Ganoderma polysaccharide on intracellular free calcium in murine peritoneal macrophages. Chin Pharm J 1999;34:805–807. 109. Cao LZ and Lin ZB. Regulation on maturation and function of dendritic cells by Ganoderma lucidum polysaccharides. Immunol Lett 2002;83:163–169. 110. Cao LZ and Lin ZB. Regulation effect of Ganoderma lucidum polysaccharides on cytotoxic T-lymphocytes induced by dendritic cells in vitro. Acta Pharmacol Sin 2003;24:321–326. 111. Shao BM, Dai H, Xu W, Lin ZB and Gao XM. Immune receptors for polysaccharides form Ganoderma lucidum. Biochem Bioph Res Commun 2004;323:133–141. 112. Chien CM, Cheng JL, Chang WT, Tien MH, Tsao CM, Chang YH, Chang HY, Hsieh JF, Wong CH and Chen ST. Polysaccharides of Ganoderma lucidum alter cell immunophenotypic expression and enhance CD56+ NK-cell cytotoxicity in cord blood. Bioorg Med Chem 2004;12:5603–5609. 113. Lei LS and Lin ZB. Effects of Ganoderma polysaccharides on the activity of DNA polymerase in spleen cells stimulated by alloantigents in mice in vitro. J Beijing Med Univ 1991;23:329–333. 114. Lei LS and Lin ZB. Effects of Ganoderma polysaccharides on the MLC reaction. Basic Med Clin 1992;12:59–60.

299 115. Cao LZ and Lin ZB. Comparison of the effects of polysaccharides from wood-cultured and bag-cultured Ganoderma lucidum on murine spleen lymphocyte proliferation in vitro. Acta Pharmacol Sin 2003;38:92–97. 116. Bao XF, Wang XS, Dong Q, Fang JN and Li XY. Structural features of immunologically active polysaccharides from Ganoderma lucidum. Phytochem 2002;59:175–181. 117. Zhang JS, Tang QJ, Zimmerman-Kordmann M, Reuter W and Fan H. Activation of B lymphocytes by GLIS, a bioactive proteoglycan from Ganoderma lucidum. Life Sci 2002;71:623–638. 118. Liu YH, Tsai CF, Kao MC, Lai YL and Tsai JJ. Effectiveness of Dp2 nasal therapy for Dp2-induced airway inflammation in mice: using oral Ganoderma lucidum as an immunomodulator. J Microbiol Immunol Infect 2003;36:236–242. 119. Lin ZB and Wang PY. The Pharmacological Study of Ganoderma Spores and Its Active Components, J Peking Univ (Health Sciences), 2006;38(5):541–547. 120. Xu X, Hou GH, Cao RH and Xie WG. Study on mechanism of Ganodermaa lucidum polysaccharides enhancing activity of LAK cells from human cord blood. Chinese J Cancer Biotherapy 1997;4:236. 121. Lin ZB and Zhang HN. Anti-tumor and immunoregulatory activities of Ganoderma lucidum and its possible mechanisms. Acta Pharmacol Sin 2004;25:1387–1391. 122. Lu PH and Negrin RS. A novel population of expended human CD3+CD56+ cells derived from T cells with potent in vivo antitumor activity in mice with severe combined immunodeficiency. J Immunol 1994;153:1687–1696. 123. Zhu XL and Lin ZB. Effects of Ganoderma lucidum polysaccharides on proliferation and cytotoxicity of cytokine-induced killer cells. Acta Pharm Sin 2005;26:1130–1137. 124. Zhu XL and Lin ZB. Modulation of cytokines production, granzyme B and perforin in murine CIK cells by Ganoderma lucidum polysaccharides. Carbohyd Polym 2006;63:188–197. 125. Wang YY, Khoo KH, Chen ST, Lin CC, Wong CH and Lin CH. Studies on the immuno-modulating and antitumor activities of Ganoderma lucidum (Reishi) polysaccharides: functional and protewmic analyses of a fucose-containing glycoprotein fraction responsible for the activities. Bioorg Med Chem 2002;10:1057–1062. 126. Chen HS, Tsai YF, Lin S, Lin CC, Khoo KH, Lin CH and Wong CH. Studies on the immuno-modulating and anti-tumor activities of Ganoderma lucidum (Reishi) polysaccharides. Bioorg Med Chem 2004;12:5595–5601. 127. Yuen MF, Ip P, Ng WK and Lai CL. Hepatotoxicity due to a formulation of Ganoderma lucidum (lingzhi). Letters to the editor. J Hepatol 2004;41:685–690. 128. Wang XX, Tang QJ, Zhang JS, Yang Y, Liu YF, Jia W and Liu F. Effect of various grades of Ganoderma lucidum on tumor inhibition and immunostimulation. Acta Edul Fung 2005;12:48–51. 129. Zhang JS, Jia W, Xing ZT, Tang QJ, Liu YF, Yang Y, Zhou CY and Liu F. Comparison of bioactivity of fruiting body and mycelia of Ganoderma lucidum and their purified fractions. Mycosystema 2004;23:85–92. 130. Cao QZ and Lin ZB. Antitumor and anti-angiogenic activity of Ganoderma lucidum polysaccharides peptide. Acta Pharmacol Sin 2004;25:833–838. 131. Cao QZ and Lin ZB. Ganoderma lucidum polysaccharides peptide inhibits the growth of vascular endothelial cell and the induction of VEGF in human lung cancer cell. Life Sci 2006;78:1457–1463. 132. Song YS, Kim SH, Sa JH, Jin C, Lim CL and Park EH. Anti-angiogenic and inhibitory activity on inducible nitric oxide production of the mushroom Ganoderma lucidum. J Ethnopharmacol 2004;90:17–20.

300 133. Stanley G, Harvey K, Slivova V, Jiang J and Sliva D. Ganoderma lucidum suppresses angiogenesis through the inhibition of secretion of VEGF and TGF-b1 from prostate cancer cells. Biochem Biophys Res Com 2005;330:46–52. 134. You YH and Lin ZB. Protective effects of Ganoderma lucidum polysaccharides peptide on injury of macrophages induced by reactive oxygen species. Acta Pharmacol Sin 2002;23:878–891. 135. Shi YL, James AE, Benzie IFF and Bushwell JA. Mushroom derived preparations in the prevention of H2O2-induced oxidative damage to cellular DNA. Teratogen Carcin Mut 2002;22:103–111. 136. Zhang HN, He JH, Yuan L and Lin ZB. In vitro and in vivo protective effect of Ganoderma lucidum polysaccharides on alloxan-induced pancreatic islets damage. Life Sci 2003;73:2307–2319. 137. Lakshmi B, Ajith TA, Jose N and Janardhanan KK. Antimutagenic activity of methanolic extract of Ganoderma lucidum and its effect on hepatic damage caused by benzo[a]pyrene. J Ethnopharmacol 2006;107:297–303. 138. Van der Hem LG, van der Vliet JA, Bocken CF, Kino K, Hoitsma HJ and Tax WJ. Ling Zhi-8: studies of a new immunomodulating agent. Transplantation 1995;60:438–443. 139. Tian YP and Zhang KC. Purification and characterization of a novel proteinase A inhibitor from Ganoderma lucidum by submerged fermentation. Enzyme Microb Technol 2005;36:357–361. 140. Liu J, Yang F, Ye LB, Yang XJ, Timani KA, Zheng Y and Wang YH. Possible mode of action of antiherpetic activities of a proteoglycan isolated from the mycelia of Ganoderma lucidum in vitro. J Ethnopharmacol 2004;95:256–272. 141. Wang HX, Ng TB and Chiu SW. A distinctive ribonuclease from fresh fruiting bodies of the medicinal mushroom Ganoderma lucidum. Biochem Biophys Res Com 2004;314:519–522. 142. Wang HX and Ng TB. Ganodermin, an antifungal protein from fruiting bodies of the medicinal mushroom Ganoderma lucidum. Peptides 2006;27:27–30. 143. Chang ST and Buswell JA. Mushroom nutriceuticals. World J Microbiol Biotechnol 1996;12:473–476. 144. Wasser SP and Weiss AL. Medicinal Mushrooms – Ganoderma lucidum, Reishi Mushroom, Haifa, Pedeifus Publishing House, 1997. 145. Zhang CY and Li YM. Clinical investigation of Green Valley Lingzhi capsule on type 2 diabetes mellitus. In: Ganoderma: Genetics, Chemistry, Pharmacology and Therapeutics, Zhi-Bin Lin (ed), Proceedings of International Symposium on Ganoderma Research, Shanghai, October 21–23, 2002, Beijing, Medical University Press, pp. 194–198. 146. Shi KG and Quing LH. The follow-up observation assessment of medium and late phases cancer treated by Chinese Ganoderma lucidum essence (CGLE). In: Ganoderma: Genetics, Chemistry, Pharmacology and Therapeutics, Zhi-Bin Lin (ed), Proceedings of International Symposium on Ganoderma Research, Shanghai, October 21–23, 2002, Beijing, Medical University Press, pp. 188–193. 147. Sliva D, Labarrere C, Slivova S, Sedlak M, Lloyd FP and Ho NW. Ganoderma lucidum suppresses motility of highly invasive breast and prostate cancer cells. Biochem Biophys Res Commun 2002;298:603–612. 148. Noguchi M, Kakuma T, Tomiyasu K, Konishi F, Kumamoto S, Kondo R and Matsuoka K. Phase I study of a methanol extract of Ganoderma lucidum, edible and medicinal mushroom, in men with mild symptoms of bladder outlet obstruction. Urology 2005;66(suppl. 3A):21.

301 149. Chen X, Hu ZP, Yang XX, Huang M, Gao Y, Tang W, Chan SY, Dai X, Ye J, Ho PC, Duan W, Yang HY, Zhu YZ and Zhou SF. Monitoring of immune responses to a herbal immuno-modulator in patients with advanced colorectal cancer. Int Immunopharmacol 2006;6:499–508. 150. Ghafa MA, Golliday E, Bingham J, Mansukhani MM, Anastasiasis AG and Katz AE. Regression of prostate cancer following administration of Genistein Combined Polysaccharide (GCP), a nutritional supplement: a case report. J Altern Complement Med 2002;8:493–497. 151. Gao Y, Zhou S, Jiang W, Huang M and Dai X. Effects of ganopoly (a Ganoderma lucidum polysaccharide extract) on the immune functions in advanced-stage cancer patients. Immunol Invest 2003;32:201–215. 152. Gao Y, Tang W, Dai X, Chen G, Ye J, Chen E, Koh HL, Li X, and Zhou S, Effects of water-soluble Ganoderma lucidum polysaccharides on the immune functions of patients with advanced lung cancer. J Med Food 2005;8:159–168. 153. Chan WK, Lam DT, Law HK, Wong WT, Leung KMW, Lau YL and Chan GC. Ganoderma lucidum mycelium and spore extracts as natural adjuvants for immunotherapy. J Altern Complement Med 2005;11:1047–1057.


303

Citric acid production Marin Berovic1, and Matic Legisa2 1

Department of Chemical, Biochemical and Ecology Engineering, Faculty of Chemistry and Chemical Technology, University of Ljubljana, Askerceva 5, 1001 Ljubljana, Slovenia 2 National Institute of Chemistry, Hajdrihova 19, 61000 Ljubljana, Slovenia Abstract. Citric acid is a commodity chemical produced and consumed throughout The World. It is used mainly in the food and beverage industry, primarily as an acidulant. Although it is one of the oldiest industrial fermentations, its World production is still in rapid increasing. Global production of citric acid in 2007 was over 1.6 million tones. Biochemistry of citric acid fermentation, various microbial strains, as well as various substrates, technological processes and product recovery are presented. World production and economics aspects of this strategically product of bulk biotechnology are discussed. Keywords: citric acid biosynthesis, microbial strains, biochemistry, substrates, production processes, product recovery, economic aspects

Introduction Citric acid is the main organic acid produced today by fermentation. The history of citric acid actually started in 1784 with W. Scheele [1] who first isolated it from the lemon juice as calcium citrate, which treated with sulphuric acid gave citric acid in the liquid phase. In 1838, Libieg considered that citric acid is actually three carboxylic acid and in 1880 Grimoux and Adam [2] synthesized citric acid from glycerolderived 1,3 dichloroacetone for the first time chemically [1]. Wehmer in 1893 was the first who observed the presence of citric acid as by-product of calcium oxalate produced by a culture of Penicillium glaucum fermenting sugar [1,2]. The result of this fermentation had encouraged him to patent the process for citric acid production [3]. On this base in 1894 the first industrial fermentation, using open-tray system was built. Ten years later the factory was closed, as the fermentation was considered too long and frequent contamination occurred [4]. After Wehmer several other researchers followed [5,6], but reasonable advance in citric acid production appeared with Zahorsky in 1913, who first patented a new strain – Aspergillus niger [7]. Following the fundamental investigations by Thom and Currie 1916 [8], Currie 1917 [9] opened the way for industrial citric acid fermentation using a new microorganism. His most important finding was that Aspergillus niger could grow well at low pH values

Corresponding author: Tel: +386 1 2419510. Fax: + 386 1 4760 300.

E-mail: [email protected] (M. Berovic). BIOTECHNOLOGY ANNUAL REVIEW VOLUME 13 ISSN 1387-2656 DOI: 10.1016/S1387-2656(07)13011-8


304 around 2.5–3.5. This low pH prevented contamination which was common in Wehmer’s process. In 1928 beet molasses, as a cheap sugar source was first used in Czechoslovakia. Difficulties, however, were encountered with this source due to its trace metals content. Using a patent of Mezzadroli 1938 [10], this problem was effectively solved by using potassium hexacianoferrate as a chelating agent for the trace metals in beet molasses substrate. Originally first industrial citric acid fermentations were carried out as surface cultures. The introduction of the submerged fermentation was a significant improvement. Among the studied preceding the commercial implementation of submerged fermentation the work of Perquin 1938 [11]; it should be mentioned as the first one for its shill and precision comparable to Currie’s work on the surface process [9]. In Japan in the 1960s, a new process emerged by using n-alkanes as a carbon source. A yeast of the genus Candida, which produced appreciable amounts of citric and isocitric acid was used [12,13]. The world production of this ‘2-hydroxy-propan-1,2,3-three carboxy acid’, by fermentation, is rapidly increasing. Although in South America, Mexico and Greece there still exists some factories where citric acid is isolated from unripe citrus fruits, today over 99 per cent of the world’s output of citric acid is produced microbially by various fermentation processes, substrates and microorganisms. The traditional method of preparing citric acid by extraction from the juice of lemons, limes and pineapple wastes is still in practice in the developing world, but its production is not significant, as it comprises less world production. Various chemical syntheses of citric acid have appeared in the chemical and patent literature since the first one based on the reaction of glycerol-derived 1.3 dichloroacetone with cyanide by Grimoux and Adam in 1880 [2]. However none of these has reached a commercial status competitive with fermentation processes. Uses and occurrence 60% of citric acid product is mainly used in the food and beverage industry, because of its general recognition as safe having pleasant taste, high water solubility and chelating and buffering properties. Citric acid is used extensively in carbonated beverages to provide taste and complement fruit and berry flavours. It also increases the effectiveness of antimicrobial preservatives. The amount of acid used depends on the flavour of the product. It may usually vary from 1.5 to 5 per cent [1–14]. In jam and jellies it is used for taste and for pH adjustment in the final product. For optimum gelation, pH has to be adjusted in very narrow limits [15]. Citric acid is usually added as a 50 per cent solution to assure good

305 distribution through the batch. In confectionery industry 0.5–2.0 per cent is used as flowing agent [16]. The chelating and pH adjusting properties of citric acid enable it to optimize the stability of frozen food products by enhancing the action of antioxidants and inactivating enzymes. It also helps to prolong the shellfire of frozen fish and shellfish [17]. Citric acid also inhibits colour and flavour deterioration in frozen fruit [18]. Amounts in concentration of 0.005–0.02 per cent citric acid are used as an antioxidant synergism in fats, oils and fat containing foods [16]. As a flavour adjunct, citric acid is used in sherbets and ice creams [16]. Temperament of total citric acid production is used in pharmaceutical industry as oral pharmaceutical liquids, elixirs and suspensions to buffer and maintain stability of active ingredients and to enhance the activity of preservatives. Addition of 0.02 per cent citric acid to liquid dosage forms complexes with trace iron and copper ions and retards degradation of active ingredients [19]. Citric acid is a standard ingredient in cosmetic formulations for pH adjustment, and in antioxidant systems as a metallic-ion chelator [20]. The detergent-building properties of citrate enables it to be used as a rapidly biodegradable environmentally acceptable phosphate substitute in non-phosphate detergent powders [21]. Citric acid-based metal cleaning formulations efficiently remove metal oxidation products from the surface of ferrous and non-ferrous metals [22]. Citrates have been reported to assist in platting of copper [23], nickel [24], chromium, lead [25] and various heavy metals [16]. Various other uses of citric acid and its salts and esters were reported also in photography as a component of printing plate emulsions in various bleaches, fixers and stabilizers [26] in oil well treatment and cements [16], in textile industry [27], in paper industry [28] and the tobacco industry [29]. Citric acid is also a preferred nucleating or blowing agent in polymeric foams for food and beverage use and its esters are used as plasticizers in the preparation of polymer compositions [16]. As the by-products of citric acid fermentation various enzymes (amylolytic, pectolytic, etc.) were referred [29].

Strains for citric acid production Many strains excrete traces of citric acid as a metabolite of primary metabolism. It is a result of some severe irregularity of metabolism caused by genetic deficiency or by metabolic imbalances. In the history of citric acid fermentation, in the last hundred years, various strains of genera fungi, yeast and bacteria were reported such as: Penicillium luterum, P. purpurogenum, P. restrictum, P. janthinellum, P. citrinum, Paecilomyces divaricatum, Mucor piriformis, Trichoderma viride, Sacharomycopsis lipolitica, Arthrobacter paraffineus, Corynebacterium sp. and others [30,31].

306 However, only mutants of Aspergillus and yeast genus Candida have almost exclusively been utilized. Apart from Aspergillus niger the following species of Aspergillus have been reported: Aspergillus niger, A. wentii, A. awamori, A. foetidus, A. fenicis, A. fonsecalus, A. fumaricus, A. luchensis, A. saitoi and A. usumii. From the genus Candida the following have to be mentioned: Candida lipolytica, C. tropicalis, C. guilliermondii, C. intermedia, C. parapsilosis, C. zeylanoides, C. fibriae, C. subtropicalix, C. oleophila. Mutants of A. niger, A. wentii and on paraffine substrats Candida lipolitica are used in industrial production [32]. For industrial citric acid production, filamentous fungus A. niger is far the most used microorganism. In the second-half of the 20th century, progress in life sciences and accumulating knowledge about metabolic events stimulated several research groups to study the biochemical basis of citric acid accumulation by A. niger. They investigated why, and under which circumstances, citric acid is accumulated so that productive strains might be improved further and yields increased. Although a number of biochemical events were found to be responsible for citric acid overflow, differences existed amongst individual high-producing strains. Research during the last two decades, has resulted in generally accepted theory that describes the metabolic pathways used and the regulation events that are significant during citric acid accumulation [33]. Biochemistry There have been many theories proposed to explain the phenomena of citric acid accumulation by A. niger [34–42], but so far no complete explanation is available. It can be said that citric acid accumulates by an induced abnormality in the metabolism of the mould during the operation of the tricarboxylic acid cycle (TCA) postulated by Krebs in 1937, under its original name ‘‘citric acid cycle’’ [47] The TCA cycle is a cyclic sequence of reactions of almost universal occurrence in mitochondria in aerobic organisms. It is catalysed by multienzyme system, that accepts the acetyl group of acetyl-Co enzyme A as fuel and dismembers it to yield carbon dioxide and hydrogen atoms [40,43]. On each turnaround within the TCA cycle, one molecule of acetic acid (two carbon atoms) enters as acetyl-Co enzyme, condenses with a molecule of the four-carbon compound oxaloacetic acid to form citric acid, the six-carbon compound. Citric acid is then degraded through a reaction sequence, that yields two molecules of CO2 and regenerates the four-carbon oxaloacetic acid. Another turn of the cycle may now start by the reaction of the oxaloacetic acid with another molecule of acetyl-Co enzyme A. Thus, in each turn of the cycle one molecule of acetic acid enters, two molecules of ATP and CO2 are formed and a molecule of oxaloacetate is utilized to form citrate, but is regenerated at the end of the cycle [44].

307 The degree of involvement of TCA cycle during the accumulation is indeed controversial. Shu et al. in 1954 found that 40 per cent of the citric acid was formed from recycled dicarboxylic acid [45]. In contrast to that other researchers were able to demonstrate some degree of recycling [46,47]. It was also considered [48] that the citric acid accumulated came from the disappearance of aconitase and isocitric dehydrogenase, since prior to the accumulation of citric acid all enzymes of TCA cycle were present. Citric acid is excreted from the cells in response to unfavourable intracellular condition caused by increased levels of tricarboxylic acids (TCA). A crucial prerequisite for overflow of citric acid from A. niger cells is therefore increased at the level of Krebs cycle intermediates caused by anaplerotic reactions. Extensive studies have revealed that there are three main metabolic events that replenish TCA intermediates and predispose the cell to product overflow. Fast uptake of glucose based on simple diffusion. Unrestricted metabolic flow through glycolysis, making precursors for

synthesis of the TCA cycle intermediates readily available. Uncoupled NADH re-oxidation resulting in lower levels of ATP and

therefore decreased anabolic reaction. Only the activities of certain enzymes of individual A. niger cells can lead to such intracellular conditions. Glucose uptake rate has been identified as an important factor in the rate of citric acid production [49,50]. By using mathematical modelling, it was shown that glycolytic reactions of A. niger are limited by the supply of the initial substrate and the removal of the final product. Two glucose carriers have been identified, the first, a high-affinity carrier that is expressed at all times, and the second, a low-affinity carrier that is expressed only in the presence of high concentrations of glucose [51]. However, Mischak et al. [52] and Torres et al. [51] reported that both glucose carriers are inhibited by citric acid under production conditions. If the entry of glucose is primarily via the glucose carriers, the effect of citric acid concentration on the observed glucose uptake should be pronounced. By contrast, a simple diffusion model fits all the observed data under citric acid excretion conditions, explaining the observed relationship between specific uptake rate and glucose concentration, which would not exist under carrier-saturated conditions [53]. Finally, because simple diffusion is an inevitable physical process, it is not capable of being regulated directly by the organisms; this may, in itself, account for dramatic overproduction of citric acid under the conditions used in this process. The simple nature of this mechanism also explains the similarity of the uptake relationship from the different sources, despite the use of different strains and growing conditions.

308 However, the de-regulated metabolic flux through glycolysis is a prerequisite for rapid synthesis of citric acid. In glycolysis, the reactions catalysed by hexokinase, phosphofructokinase and pyruvate kinase are virtually irreversible. The activities of these enzymes are regulated by reversible binding of allosteric effectors or by covalent modification. Normally in eukaryotic organisms, the phosphofructokinase is the most important control element. However, in Aspergillus niger, during the growth on high sugar concentrations that are needed for rapid citric acid formation, the control of glycolysis is shifted from 6-phosphofructo-1-kinase level to the glyceraldehyde step [54]. In the literature, two attempts can be found to influence the efficiency of this pathway by genetic modification of the enzymes involved. Firstly, by disrupting trehalose-6-phosphate synthase gene (ggsA), the synthesis of trehalose-6-phosphate, a potent inhibitor of glycolysis was prevented, yet citric acid accumulation improved only slightly [55]. Secondly, moderate overexpression of the key regulatory enzymes, 6-phosphofructo-1-kinase and pyruvate kinase did not enhance acid production [56]. De-regulated glycolysis leading to strong anaplerosis is characteristic for a productive phase of A. niger cells, however, significant physiological changes are taking place in the cells during the early stages of growth in high initial sucrose or glucose medium that have crucial impact on overall productivity and yield. In fact, there could be no citric acid detected in the substrate during the first 24 h of growth [57], while a relatively slow accumulation rate initiates only during the second day of fermentation, followed by a sudden increase in specific productivity afterwards [57–59]. By flux distribution experiments pentose phosphate (PP) pathway was found to be predominant during the germination of spores, followed by a switch to glycolysis [60,61]. Initial phases of growth are characterized also by polyol formation and glucosamine accumulation. Polyols especially glycerol, whose intracellular concentration can reach up to 175 mM, may play an important role as an osmoregulator in A. niger cells during the growth in high sucrose medium [61]. Because the enhanced glycolytic flux is a prerequisite for increased anaplerosis, understanding the mechanism of the switch in carbohydrate metabolism from the PP pathway to glycolysis is of crucial importance. Although the initial inhibition at the level of TCA cycle enzymes has been extensively studied in the past, some authors in their recent reviews on citric acid accumulation by A. niger exclude the hypothesis of an inhibition of TCA cycle in the phase where acid accumulation starts [62]. However by measuring intracellular citrate concentration in the cells, low levels of citrate were recorded in germinating spores followed by a constant rise up to 10 mM before 24 h of fermentation [61]. An inhibition of NADP-dependent isocitrate dehydrogenase (ICDH) by glycerol was initially proposed to trigger an increase of intracellular citrate [63] on the basis of kinetic measurements performed on un-purified enzyme in

309 the homogenate [63,64]; however, later tests on partially purified enzyme showed no inhibition by glycerol [64]. NADP-specific isocitrate dehydrogenase was found to be predominantly located in mitochondrial compartment when glucose was used as sole carbon source [65], while only minor activities of NAD-dependent enzyme were detected in A. niger cells [66]. However, NADP-isocitrate dehydrogenase was found to be inhibited by citrate [66] and reduced metabolic flow at the early stages of fermentation through the TCA cycle at the stage of ICDH, could be predicted from studies based on the distribution of different marked C13 atoms in glucose by Peksel et al. [54]. At the early stages of citrate formation, their model indicated a substrate cycle of oxaloacetate to pyruvate was substantial in comparison to the flux of oxaloacetate to citrate. Simultaneously, a significant pyruvate/ phosphoenolpyruvate substrate cycle was predicted. Later in the fermentation there was decreased operation of the pyruvate/phosphoenolpyruvate substrate cycle and a net flux to citrate. The question of what triggers the initial increase in citrate concentration remains unexplained. It might be that another substance structurally related to citrate and metabolically formed from glycerol could cause initial deactivation of mitochondrial TCA cycle enzyme or the mass action effect of intermediates from glucose might cause the increase. The initial increase in intracellular citrate concentration could cause a decrease in glucose degradation through the PP pathway. It was shown that 6-phosphogluconate dehydrogenase, one of the regulatory enzymes of the oxidative step of PP pathway, is inhibited by citrate with apparent Ki value of 0.8 mM [67]. The direct conversion of hexoses to pyruvate via glycolysis becomes predominant during the productive phase of citric acid accumulation that starts after about 24 h and accelerates after 40–50 h of growth in a batch system. Regulation of the central part of hexose metabolism takes place at several levels: at the transcriptional level, by regulating the activity of allosteric enzymes by specific effectors and as revealed recently [68,69], even by posttranslational modification. In the glycolytic flux, 6-phosphofructo-1-kinase (EC 2.7.1.11) is the most important control element. It catalyses essentially irreversible reaction of glycolysis, the phosphorylation of fructose-6-phosphate using Mg-ATP to form fructose-1,6-bisphosphate and releasing Mg-ADP. Six organic allosteric ligands either increase or decrease substrate-binding affinity and concomitantly determine overall enzyme activity [70]. The enzyme attracted the interest of investigators, due to its ability to maintain a high glycolytic flux in spite of elevated intracellular concentrations of citrate, a well-known inhibitor of PFK1, which was reported to reach concentrations between 4 [57] and 10 mM [61]. Recently, another enzyme exhibiting PFK1 activity was isolated from A. niger mycelium with molecular mass of 49 kDa. A fragment of identical

310 size could also be obtained in vitro by the proteolytic cleavage of the purified native PFK1 with proteinase K, which regained its activity after the phosphorylation of the protein molecule by catalytic subunit of cAMP-dependent protein kinase. The native enzyme as a sole PFK1 enzyme could be isolated only from the early stages of growth on a minimal medium, while a 49 kDa fragment seemed to appear later and was activated concurrently with a sudden change in the growth rate. There is a strong evidence that the native PFK1 enzyme undergoes spontaneous posttranslational modification at the early stages of the fungal development. By measuring kinetic parameters of both PFK1 forms found in A. niger cells, ATP proved to be a strong inhibitor of the short PFK1 fragment, but the negative effect of ATP seemed to be suppressed by physiological concentrations of fructose-2,6-bisphosphate. The same effector significantly increased the Vmax and the affinity of the fragmented protein towards the substrate, while it does not affect the maximal velocity of the native protein [68]. In A. niger about 6 mM of fructose-2,6-bisphosphate were detected under citric acid excreting conditions [56]. The studies on fructose-2,6-bisphosphate formation showed that its synthesis is stimulated after the transfer of A. niger mycelium from low (1 per cent) to high (14 per cent) initial sucrose medium simultaneously with a rapid increase in cAMP level [71]. Other PFK1 stimulators, AMP and ammonium ions, increased the activity of the shorter fragment more intensely than the activity of the native protein, while citrate, a well-known allosteric inhibitor of eukaryotic PFK1 enzymes, showed moderate inhibition of the native enzyme, while no inhibition of the fragment could be observed by concentrations up to 10 mM [69]. Kinetic data so far obtained support the hypothesis that the posttranslational modification is needed for the formation of a highly active PFK1 enzyme insensitive to normal feedback control by citrate. Another phenomenon taking place at the early stages of A. niger growth in a high citric acid yielding medium seems to be relevant for the development of high citric acid yielding mycelium. Namely, the shorter fragment is inactive immediately after proteolytic cleavage and must be phosphorylated to regain activity [68,72]. PKA was found to be capable of the appropriate phosphorylation, which led to the re-activation of 49 kDa fragment [68]. Kinases are normally under the tight control of specific regulatory subunits and cyclic AMP is known to induce PKA enzyme. In A. niger strain A60, a spontaneous increase in the concentration of cyclic AMP was recorded after 24 h of growth in a citrate yielding medium. Further analyses have shown that the amount of cAMP formed depends on the initial concentration of sucrose in the medium. Under higher sucrose conditions the cAMP peak appeared earlier and was higher, while in lower sucrose media a flattened peak was observed later in fermentation [73]. A spontaneous increase in cyclic AMP concentration could be caused by intracellular acidification since a drop of intracellular pH is known to stimulate the RAS-adenylate cyclase signalling

311 pathway in a number of fungal species, including Saccharomyces cerevisiae [74]. In the citrate accumulating strain (A60) grown in a high sucrose medium, intracellular acidification was indeed recorded at the early stages of growth [61,75], while in another A. niger strain (NW131) no change in cytoplasmic and vacuolar pH could be detected by P31 NMR technique during the growth of immobilized cells [76]. Significant differences in membrane H+-ATPase activities of both strains were described. In the A60 (NRRL 2270; ATCC 11 414) strain, the activity of proton pumps was a quarter of that in A158 (CBS 120,49; N400), a strain which is related to the NW131 strain [75]. Moreover, under identical growth conditions strain A158 extruded protons more rapidly into the medium when ammonium ions were used as a sole nitrogen source than the A60 strain, indicating that proton pumps of the latter strain perhaps cannot extrude all the protons that are released into the cytosol after initial increase in intracellular citric acid concentration and ammonium assimilation [75]. Citric acid, which can reach a concentration of up to 10 mM in the cells [61], dissociates at neutral pH values (pK3 ¼ 5.4) releasing two protons. However under citric acid accumulating conditions, the ammonium salts are the preferred source of nitrogen. A. niger consumes ammonium very rapidly and it is normally depleted from the medium between 40 and 50 h of fermentation, which is well before the fungus stops growing [58]. The amount of protons excreted from the biomass appeared to be directly related to the initial ammonium concentration [58]. It is worth noting that ammonium ions are taken up by an uniport mechanism, however after the incorporation of NH+ 4 ion as an amino group, two protons are released which must be pumped back to the medium by H+-ATPases to maintain electroneutrality in the cells. Stoichiometric modelling of the early stages of the fermentation revealed that ammonium ions combine with a carbon-containing metabolite inside the cells in a ratio 1:1, to form an organic nitrogen compound, which is immediately excreted by the mycelium. The compound was proven to be glucosamine [58]. Characteristically, the maximal rate of acid overflow was recorded only after the depletion of ammonium from the medium, although increase in dry biomass was observed at later phases of growth as well. The enzyme responsible for glucosamine formation must be glucosamine-6-phosphate deaminase, which catalyses amination of fructose-6-phosphate to produce glucosamine-6-phosphate. Since the enzyme competes for the same substrate (fructose-6-phosphate) as PFK1, rapid accumulation of glucosamine must significantly decrease the metabolic flux through glycolysis at the early stages of growth; however, better understanding of the phenomenon must await more detailed characterization of deaminase kinetics. Aspergillus niger is well known for its strong extra and intracellular proteolytic activity [77]. Although proteases are normally strictly compartmentalized in the cells and are activated from their pre-pro forms only in the vacuoles, some leakage through the tonoplast into the cytosol must occur,

312 since cleavage of the native PFK1 enzyme takes place. In A. niger cells increased protein degradation was reported under the manganese-deficient conditions, which was also reflected by increased intracellular proteinase activities [78,79]. For efficient citric acid fermentation, lack of trace metal ions, particularly Mn2+ ions from the medium is of major importance. Much has been speculated about the principal physiological role of manganese ions in citric acid overflow in past, which seem to affect metabolism on various levels [62]. Whether manganese ions are somehow involved in increased cytosolic protease activities, and concomitantly, in the posttranslational modification of PFK1, will have to await further investigation. During the idiophase, the phase of maximal product formation [80], no significant inhibition of the TCA cycle could be observed and an increased level of all tricarboxylic acids, with an exception of 2-oxo-acids, can be detected in the mycelium [81]. The only plausible explanation of the phenomenon is accelerated glucose metabolism, which was confirmed also by testing mutants with increased citric acid productivity in respect to the parental strain [82]. However, another two enzymatic reactions that appear in A. niger cells play a noteworthy role in citric acid overflow. Cytosolic pyruvate carboxylase [83] and malate dehydrogenase isoenzyme [84] are catalysing the conversion of pyruvate first into oxaloacetate and finally into malate. In the late phase of fermentation, carbon dioxide fixation by pyruvate carboxylase became an important anaplerotic reaction [85], while increased concentration of malate in the cytosol finally serves as a counter ion for citrate export from the mitochondrial compartment by a tricarboxylic acid carrier [86]. The formation of citric acid is dependent on strong aeration; dissolved oxygen tensions higher than those required for the vegetative growth of A. niger stimulate citric acid fermentation [87,88]. The biochemical basis of this observation is related to the presence of an alternative, cyanide-resistant respiratory pathway, which is required for the re-oxidation of glycolytically produced NADH, when high oxygen tension is maintained. The enzyme responsible for the additional respiratory pathway is an alternative oxidase, which catalyses reduction of oxygen to water without the translocation of protons across the inner mitochondrial membrane, and thus functions as a non-energy-conserving member of the respiratory electron chain. The alternative respiration seems to be constitutively present in citric acid producing strains [89,90]. The alternative oxidase is synthesized in the cytosol and translocated into the mitochondria [91]. Although aox-1 gene encoding alternative oxidase from A. niger cells has been isolated, cloned and characterized [92], no transformants carrying multiple gene copies or strains with disrupted gene were prepared and tested for intracellular ATP concentrations and/or citric acid overflow. However, it is generally accepted that the presence of uncoupled NADH re-oxidation results in lower levels of ATP and therefore decreases anabolic reactions.

313 After citrate first accumulates in the cytosol, it must pass the plasma membrane to be excreted into the substrate. It was assumed that citrate, a charged metabolite cannot cross the lipid bilayer without support of the transport protein and an active, pH driven, H+-symport-dependent system was proposed that was functional only under the manganese deficient growth conditions [93]. However, recent thermodynamic calculations presented for citrate overflow from A. niger cells at the pH value 3 of the substrate suggest that a passive transport step suffices for citrate excretion [94]. Future perspectives In the last several decades enough knowledge on biochemical mechanisms leading to citric acid overflow have accumulated to generally understand the phenomenon, however many details still remain unexplained. Recently, several genomes of Aspergillus species have been fully sequenced and the information published: A. fumugatus [95], A. nidulans [96], A. oryzae [97], while the information of several other genomes including A. niger will be released in near future. On the basis of data from sequence analyses, and physiological information published from A. niger and related filamentous fungi, in silico model of the central carbon metabolism of A. niger has been constructed [98] and is regularly updated. Application of the stoichiometric model together with recent discoveries on the posttranslational modification of the key regulatory enzyme of glycolysis will present a powerful tool for further improvement of the primary metabolism in A. niger that will result in stronger anaplerosis and increased productivity. Influence of the trace metals In citric acid technology, absence of iron and manganese in the fermentation substrate plays the most crucial role [99,100]. Trace element nutrition is specially highlighted by the fact that an optimal nutrient medium for citric acid fermentation will not allow high production unless the trace elements content is carefully controlled [101]. However, if the trace element nutrition is correct, other factors (sugar concentration, phosphate and the others) have only less pronounced effects [102]. Iron ions in higher concentration than 1.5 mg/l strongly affect cellular morphology, by promoting unproductive filamentous mycelial growth form [103,104]. In further insights into the importance of metal ions, the presence of manganese ions to citric acid fermentation was reported by Clark et al. [105]. As little as 1 mg/l of manganese could completely ruined the production yield and caused organism’s morphology to switch from microbial pellets, known as citric acid productive form, to unproductive filamentous growth. In contrast, the most recent research by Berovic et al. [106] found that when fungal biomass reaches its stationary phase even in a case when fed

314 media contains unusually high amounts of manganese ions up to 200 mg/l, the presence of heavy metal ions do not affect on mycelial growth nor citric acid biosynthesis. Manganese deficiency lower than 107 M raised chitin and reduced bglucan production. Manganese levels also affect lipid synthesis, which in turn affects cell membrane composition [107]. It also exhibits effects on DNA synthesis of A. niger. Under manganese limitation, DNA formation was not inhibited but RNA synthesis was impaired [78]. On the other hand manganese deficiency in A. niger cultivation also results in significantly lower lipid levels due primarily to reduction of triglycerides and with little effect on free acids and sterols [108]. In anyway the influence of manganese ions on A. niger is very complex and it represents the most critical metal ion in citric acid fermentation [101,104]. In recent articles, the attention to A. niger metal ion tolerance was related to action of elevated manganese ion concentration and effects of copper and zinc antagonism to iron and manganese [109,110] and to various genetic manipulation for metal resistance strain improvement [111–113].

Substrates Most processes are based on molasses, although the use of cleaner sources is gaining ground. Whatever the source, its cost and preparation to permit optimal fermentation conditions are two important aspects of the technology in citric acid production. The basic substrate for citric acid fermentation in plants using the surface method of fermentation is beet or cane molasses. Plants using submerged fermentation can use not only beet or cane molasses, but a substrate of higher purity such as hydrolysed starch, technical and pure glucose, refined or raw sugar, purified and condensed beet or cane juice. This is because use of a pure substrate may result in increase in yield, or reduction in fermentation time [114].

Molasses Molasses is a widely used substrate, coming in a variety of qualities. Highquality molasses is usually demanded for citric acid production. Cane and beet molasses are not identical in composition; often one type will be preferred to the other. They are sometimes mixed to take advantage of the additional nutrients arising from the differences in composition. Besides substrate type (sugar beet, sugar cane), the chemical composition of molasses depends on many factors such as soil and climate conditions, fertilization type, crop method, time and conditions of storage, production technology, technical equipment of plant, etc. [114].

315 Beet molasses Beet molasses consists of about 65–80 per cent dry substance and 20–25 per cent water. The main ingredient of molasses is sucrose, 44–54 per cent by weight. Other sugars (carbohydrates), which can be found in higher amounts are inverted sugar 0.4–1.5 per cent, raffinose 0.5–2.0 per cent and kestose and neokestose 0.6–1.6 per cent. Raffinose is a natural part of sugar beet, while kestose is the result of microbial action during sugar beet treatment. Other sugars in molasses are arabinose, xylose and mannose in amounts of 0.5–1.5 per cent. All sugars (except sucrose) are included in the non-nitrogen organic substances of molasses. Products of chemical and thermal sugar decomposition (melanoidines, caramel) and organic acids also belong to this group. Caramel consists of sugar anhydride and colouring matters; mela-noidines are made in hot solution as the result of a reaction between reducing sugars and amino acids. In addition to the non-volatile dark coloured compounds, there are about 40 volatile compounds as aliphatic aldehyde, methylglyoxal, diacetyl, acetoin, acetone, oxymethylfurfurol and others [114]. The non-volatile organic acids present in molasses are glutaric, malonic, succinic, aconitic, malic and lactic acid; the remainder are oxalic, citric and tartaric acid. These can all react with calcium to form insoluble salts that can influence the precipitation and recovery of the citric acid crystals. Molasses contain such volatile acids as formic, acetic, propionic, butyric and valeric acid. Almost all organic acids, volatile and non-volatile, are from potassium or calcium salts. Molasses containing higher amounts (over 1 per cent) of volatile acids are normally too dark to be used as feedstock for the citric acid fermentation. Nitrogen compounds contained in molasses are mostly betaine (about 60–70 per cent of total nitrogen), amino acids (20–30 per cent of nitrogen), protein (3–4 per cent of nitrogen) and traces of nitrogen in ammonium nitrate and amide. The amino acids content in molasses depends on the soil and climate conditions and beet cultivation. Betaine comes from beet and is not used by microorganisms as a nitrogen source. The content of mineral substances in beet molasses amounts to 8.5–14.0 per cent [114]. Besides these factors, one of the most relevant parameters for high yielding citric acid fermentation is also the amount of particular microelements in different molasses. The pH of molasses depends on the sugar extraction technology. It was considered that a neutral, or slightly alkaline molasses gave the best citric acid yields. Citric acid production needs molasses with low buffer ability, to make possible the required rapid fall of medium pH during fermentation [114]. Cane molasses Cane molasses differs from beet molasses in its chemical composition. It contains less sucrose and more inverted sugar, has lower content of nitrogen and raffinose, more intensive colour and lower buffer capacity.

316 Beet and cane molasses can also contain other substances, which appear in small amounts, but are often crucial in deciding whether the molasses are suitable for use in citric acid biosynthesis. These are pesticides, fungicides and herbicides used in beet and cane cultivation and also substances used for defoaming in sugar production process. All have mostly toxic properties and negatively affect molasses usability. In general beet molasses is more suitable for citric acid fermentation than cane molasses. It is especially relevant in submerged fermentation where the quality of the substrate is more important for productivity and fermentation yield. The microflora of molasses can be an agent of negative influence on yield and productivity of fermentation. Molasses will always contain a certain number and type of microorganisms, sometimes the count can be higher than 10,000 g1 of molasses. The most common microorganisms in molasses are species of Bacillus, sometimes yeasts of Candida species, and very rarely, moulds of Penicillium, Aspergillus and other species [114]. The basic operation in molasses preparation is a treatment for heavy metal ions removal. Potassium ferrocyanide or other complex compounds are commonly used. Another compound complexing with heavy metals is the sodium salt of ethylene-diamineacetic acid (EDTA). Other heavy metal complexing compounds can also be used, e.g., sodium polyphosphates, potassium rhodanate, 2,4-dinitrophenols and 8-oxyquinoline. Molasses media are sometimes purified by ionites, especially on cation exchanger. Not all microelements should be removed during this process, as some of them are necessary for growth of the Aspergillus niger mycelium [114]. Sucrose Refined sugar of beet or cane is almost pure sucrose, which Aspergillus niger strains ferment very well [115]. Preparation of a refined sugar solution as a fermentation medium is based on its diluting with water to a concentration of 15–22 per cent, adding necessary nutrients (NH4NO3, KH2PO4, MgSO4) and acidifying with sulphuric acid to pH 2.6–3.0 [116]. Batch medium is sterilized in the fermentation vessel. All the ingredients of the fermentation medium are added straight into the bioreactor or are prepared separately by diluting in hot water (85–951C) and then pumped into the bioreactor. In this case, sugar is diluted to 50–60 per cent concentration and pumped into the fermenter that has had an exact amount of sterile water added, resulting in a total sugar concentration of 15–22 per cent. Syrups Syrups of beet or cane sugar can also be used as basic substrate for the submerged citric acid fermentation. The great advantage with this substrate is its purity; however, the quality of the syrups deteriorates rapidly during

317 storage. Because of this they can only be used during the sugar campaign season and only if the citric acid plant is not too far from the sugar factory because of the large transport costs. Preparation of the syrups for fermentation entails dilution with water to a sugar concentration of 15–20 per cent, addition of necessary nutrients (NH4NO3, KH2PO4, MgSO4, (NH4)2C2O4), acidification with hydrochloric or sulphuric acid to pH 4–5 and sterilization at 1211C for 0.5–1 h [117]. Starch The production of citric acid from sources of starch such as corn, wheat, tapioca and potato is widely used. The suitability of these substrates for citric acid fermentation depends on their purity and method of hydrolysis. Acid hydrolysis, enzymatic hydrolysis, or a combination of the two, are used. Preparation of starch substrates for fermentation is based on their enzymatic liquefaction and saccharification to a defined hydrolysis level. Additional nutrients are added, depending on which starch is used. The pH is adjusted to 3–4 using hydrochloric or sulphuric acid and the medium is sterilized at 1211C for 0.5–1 h. Good citric acid yields have been also obtained using dextrose syrup, obtained by enzymatic hydrolysis of starch. This method is now employed also in industrial scale. In this case it is especially important to restrict the amount of heavy metals below critical levels; heavy metals should therefore be removed by ion exchange. When using an Aspergillus niger strain resistant to higher concentrations of heavy metals, practically the same yield may be obtained on decationized and non-decationized dextrose syrup [118]. Hydrol This is a paramolasses obtained as a by-product during crystalline glucose production from starch. Because of the high glucose content (40–45 per cent) and high purity coefficient it is a very good substrate for citric acid production. Preparation of hydrol for fermentation involves dilution to a sugar concentration of 15–18 per cent, addition of necessary nutrients and adjustment of pH with hydrochloric or sulphuric acid to 3.0–4.0. The solution is sterilized at 1211C for 0.5 h and cooled to 32–351C [119]. Alkanes The low price of alkanes, coupled with the ability of many organisms to utilize them, produced major changes in the fermentation industry during the 1960s and 1970s. Citric acid production, using Candida lipolytica, is a typical example and has been the subject of many patents [120,121]. There are few

318 industrial citric acid processes that are based on alkanes. In these processes, isocitric acid would also be produced at concentrations that would cause product recovery problems, as well as reduced citric acid yields [122]. A fourfold increase in price since 1973 no longer makes alkanes a cheap substrate. Oils and fats For citric acid production, oils are now being used as principal carbon source in a manner analogous to the previous use of alkanes. With palm oil as carbon source, a yield of citric acid of 145 per cent using a mutant of Candida lipolytica has been reported [123]. There are examples of oil being added in small concentrations to Aspergillus niger fermentation [124] and even being used as a sole carbon source for Aspergillus niger fermentation. It was found that citric acid could be produced on these substrates with good yield [125]. These oils and fats may replace alkanes in several fermentations, but it is unlikely that they will remain at their current low prices. Production processes Although in citric acid, industrial scale production in past surface or emerged production in earlier years of twenty centuries dominated over traditional method of preparing citric acid by extraction from various juices; at the present time a much greater emphasis is placed on the use of submerged culture production. Batch techniques in stirred tank or airlift bioreactors are in general use. Very promising results were obtained in fed-batch process [38–40] and by continuous fermentation [126–129] where various kinds of bioreactors as stirred tank reactors [99,102,105], airlift reactors [130,131], external loop reactors [132,133], magnetic drum contactors [134], reciprocated jet reactors, biodisc reactor [135], deep jet reactors [136,137], in hollow fibre [138] or by use of fix bed reactor [133]. Several report of citric acid fermentation using immobilized A. niger cells on various kinds of carriers as glass [139], polyurethane foams [140], entrapment in calcium alginate beds [141–143] polyacrylamide gels [144,145] agar [146] agarose [147] cellulose carriers [148,149] metal screens and polyester felts [150,151]. The traditional method of preparing citric acid by extraction from the juice of lemons, limes and pineapple wastes is still in practice in the developing World, but its production is not significant, as it comprises less World production [152,153]. Various chemical syntheses of citric acid have appeared in the chemical and patent literature since the first one based on the reaction of glycerol-derived 1.3 dichloroacetone with cyanide by Grimoux and Adam [2]. However, none

319 of these has reached a commercial status competitive with fermentation processes.

Surface process on liquid substrate The surface fermentation process, using liquid substrate, is the oldest production method and accounts for 5–10 per cent of the World supply of citric acid. This process is still in use because of low investment, and energy cost for the cooling and heating system, and due to simple technology, despite to the higher labour costs as compared to submerged fermentation. The system consists of fermentation rooms in which a large number of trays are mounted one over the other in stable racks. The trays are generally made of high purity aluminium or special stainless steel. Their size varies from 2 2.5 0.15 m to 2.5 4 0.15 m, with usage liquid depths of 0.08–0.12 m. Provision is made for continuous filling and draining by appropriate overflow devices. Aeration is provided by climatized sterile air circulation, which serves the purpose of temperature regulation and only to a lesser extent that of supplying oxygen and controlling humidity. Air is introduced in to the fermentation chamber in an almost laminar flow manner [27]. Molasses substrates are generally employed as substrates containing 15–20 per cent of sucrose, added nutrients, various natural polymers [42,43], acidified with, e.g., phosphoric acid to a pH 6.0–6.5 and heated at temperature 1101C for 15–45 min. Subsequently, potassium hexacyanoferrate is added to the hot substrate, to precipitate or complex trace metals [Fe, Mn, Zn] and to act in excess as a metabolic inhibitor restricting growth and promoting acid production [43]. For some molasses combined treatment with tricalcium phosphate, hydrochloric acid and Sephadex was used [46]. Inoculation is performed in two ways, as a suspension of conidia added to the cooled medium, or as a dry conidia mixed with sterile air and spread as an aerosol over the trays [27]. The temperature is kept constant at 301C during the fermentation by means of air current. Ventilation is also important for gas exchange because the rate of citric acid production drops if carbon dioxide in the atmosphere increases over 10 per cent. Within 24 h after inoculation, the germinating spores start forming a 2–3 cm cover blanket of mycelium floating on the surface of the substrate. As a result of the uptake of ammonium ions, the pH of the substrate falls to 2.0. After 30 h the idiophase begins. If too much iron ions are present, oxalic acid is produced and a yellowish pigment is formed, which later complicates the recovery process. The fully developed mycelium floats as a thick white layer on the nutrient solution. Through evaporation, the temperature can be maintained constant, but the culture loses 30–40 per cent of its original volume. The fermentation process stops after 8–14 days.

320 For recovery, the mycelium and nutrient solution are removed from the chambers. Owing to its volume, the mycelium must be carefully washed in sections. On some cases, mechanical presses are also used to obtain more citric acid from the cells. Solid state fermentation Surface process employing solid substrate may use fibrous residues from apple [44], grape pommace [45], wheat bran or rice starch containing residual pulps from starch manufacture [46], potato [46] and sweet potato [47]. In this process, based on the traditional koji process know-how, the Aspergillus niger strains are not sensitive to trace elements as in surface fermentation with liquid substrate or in submerged process [48]. On the solid-state fermentation process, the solid substrate is soaked with water up to 65–70 per cent of water content. After the removal of excess water, the mass undergoes a steaming process. After sterilization, sterile starch paste is inoculated by spreading Aspergillus niger conidia in the form of aerosol or as a liquid conidia suspension on the substrate surface [27]. The pH of the substrate is about 5–5.5, and incubation temperature 28–301C. Growth can be accelerated by adding a-amylase, although the fungus can hydrolyse starch with its own a-amylase. During the citric acid production pH dropped to values below 2 [40]. The solid-state surface process takes 5–8 days at the end of which the entire is extracted with hot water. On other cases, mechanical passes are also used to obtain more citric acid from the cells. Using cane bagasse as the substrate by solid-state fermentation citric acid was obtained in 6 days [154]. Total World production of citric acid by solidstate fermentation was in 1990 about 350,000 tons [152,153]. Submerged fermentation An effective alternative to surface fermentation processes is the submerged process. Although taking a longer fermentation time it has several advantages: lower investment by a factor of 2.5, 25 per cent lower total investment and labour costs, more effective process control and sterility. The disadvantages are the higher energy costs and more sophisticated control, which require more highly trained personal. Three main factors especially important for high yielding citric acid production in submerged processes are [40]: – quality of the stainless steel for the construction of the bioreactor, – mycelium structure, and – oxygen transfer.

321 Batch techniques in stirred tank or airlift bioreactors are in general use. Very promising results were obtained in fed-batch process [38–40] and by continuous fermentation [126–129] where various kinds of bioreactors as stirred tank reactors [99,102,105], airlift reactors [130,131], external loop reactors [132,133], magnetic drum contactors [134], reciprocated jet reactors, biodisc reactor [135], deep jet [136,137] hollow fibre [138] or by use of fix bed reactor [133]. Several report of citric acid fermentation using immobilized A. niger cells on various kinds of carriers as glass [139], polyurethane foams [140], entrapment in calcium alginate beds [141–143] polyacrylamide gels [144,145] agar [146] agarose [147] cellulose carriers [148,149] metal screens and polyester felts [150,151]. Bioreactors for citric acid production must be either protected from acids or constructed of special stainless steel. At pH value 2, the heavy metals leached from ordinary steel fermenter walls can inhibit the formation of citric acid [40]. Various substrates as beet [43,155–158] and cane molasses [159–161], media [162–164], starch hydrolysates [165–169], C9-C23 paraffins [28,29] and consume oil [170–172] have been reported. The concentration of carbon source in fermentation substrate is of great importance. Maximum citric acid production is usually achieved at carbon concentrations as high as 14–27 per cent. Submerged fermentation using A. niger In the case of beet molasses substrate, the reducing sugar content is usually 12–15 per cent. The row molasses is previously clarified by sulphuric acid and neutralized. Potassium hexacyanoferrate is added to the preparatory substrate for the purpose of suppressing, by means of complex formation, any detrimental effect of metal ions, particularly iron and to prevent a too rapid growth of the mycelium. Nutritive salts, such as ammonium nitrate or potassium dihydrogen phosphate may be added. For substrate preparation common tap water can be used. Owing to its content of salts, it is generally more suitable than deionized water, pH of the substrate should be adjusted to 5.5–5.9, which is most suitable for the germinated conidia aggregation. Substrate is sterilized by heat, mostly by continuous sterilization [165]. In the case of the relatively pure sucrose containing substrates, fermentation is generally run at the medium sucrose concentration of 15–27 per cent. After ion exchange of the cations, the filtered solution is sterilized subsequently by heating and after cooling to 501C fed to bioreactor. The bioreactors filled up to its working volume and nutrients are added. The pH is adjusted to an initial value of 2.5–3.0 [27]. The process can usually run in one or two stages, using hydrophilic spores suspensions [40] or germinated conidia from the propagator stage [165]. The use of germinated conidia may shorten the fermentation cycle from 12 to 24 h

322 [27]. Amounts of spores are 5–25 106 per litre of substrate [27] and conidia 1010 per litre [165]. It has been proved useful to incubate the spore suspension for 6–8 h in saline solution with added surface active agents prior to inoculation, thus shortening the fermentation cycle for 12 h [162]. On the two-stage fermentation process, germinated conida are produced in the first stage at pH 5.8, with absence of phosphate, at low dissolved oxygen level and at a sugar concentration of 7–9 per cent at temperature of 321C [166]. For citric acid production, the spherical mycelia pellet growth form is widely used [170,171]. An effective pellet formation is preferably performed by a higher shear stress affected by aeration and agitation of the substrate. The development of the hyphae and the aggregation generally requires a period from 9 to 25 h at temperature of 321C. The first two or three days of fermentation, i.e., the period of initial mycelial growth and pellet formation, are decisive for the success of the fermentation. Heavy metals in the medium (Fe, Mn), exceeding concentration of iron ions higher than 1.5 mg/l and manganese 1 mg/l strongly affect cellular morphology, by promoting unproductive filamentous mycelial growth form [27]. The production of citric acid starts after 24 hours of inoculation. Mycelial aggregates and spherical pellets, the productive form, can be detected at the first and the second maximum of the redox potential curve [172,173]. The start of citric acid production is followed by an excessive foaming, therefore an effective foam control system is essentially needed [174]. The additions of silicone antifoam agents can reduce the dissolved oxygen concentration [175,176] or influence increasingly the pseudoplastic rheological behaviour of the fermentation broth [171]. On the production phase the aeration is set from 0.3 v.v.m., in germination phase to 1 v.v.m. The change of pH in this phase is from 5.5 to 3.5, for beet molasses substrate, and to 2.2 for the sucrose substrate. The pH of 4.5–3 is also characteristic for the fed batch fermentation production phase, where bioreactor is only filled to 40 per cent of its working volume with the propagation substrate containing 7 per cent of sucrose content, and fulfilled with citric acid production substrate of 17 per cent sucrose content [27]. The temperature in the production phase is from 281C to 321C. Temperature change from 321C to 281C is also the base of some industrial processes and patents [177]. Submerged citric acid fermentation using starch hydrolysates as a carbon source is also an effective alternative to standard processes on beet molasses or sucrose substrate. An advanced step in this technology is LIKO process based on various starch hydrolysates [166]. On the first step starch is treated by thermostabile a-amylase and temperature of 1031C. Often additional enzymatic treatment by combination of pullulanase and fungal a-amylase is needed. Nutrient salts were added to the starch hydrolysate and substrate is after continuous sterilization used in propagation and production stage.

323 For the separation of the product and waste biomass tangential flow filtration is used. Compared to classic Ca citrate precipitations with 6.0 per cent bases, only 28 per cent of the citric acid was lost. The benefit of this process are simpler isolation more pure product and no calcium sulphate. Submerged citric acid fermentation on wheat flavour hydrolysates designed is inexpensive technology with more pure product has a lot of perspectives in the future production Comparing to using beet molasses and sucrose substrates, this is also a cheaper process. Probably one of the most perspective submerged citric acid technology on sucrose substrate by Lesˇ niak et al. (88–90 per cent) yields on industrial 150 m3 reactors [38]. As an alternative to batch and fed-batch process is Aspergillus niger continuous citric acid fermentation developed by B. Kristiansen and co-workers [126,127]. Submerged fermentation using yeasts Candida strains are also used novel process that permits production of citric acid from C9 to C20 normal paraffins. Citric acid yields up to 95 per cent were claimed. On 1974 Pfizer patented a continuous process for fermentation by Candida lipolitica using a single bioreactor to which paraffin was continuously added and fermented broth continuously withdrawn [168]. On citric acid fermentation stirred tank bioreactors, with usual capacities from 50 to 150 m3 and airlift bioreactors up to 220 m3, are used. The fermentation is a growth-associated process which lasts from 6 to 8 days [190,191]. Submerged fermentation using immobilization of microorganisms It is worth noting that some of the problems arising in the downstream processing of citric acid produced by submerged cultivation, especially in a continuous process, might be minimized by immobilization of microorganisms in the bioreactor. The successful application of immobilized microorganisms as living biocatalysts, involving more careful handling and often having higher production rates than free microorganisms, has prompted a rapid development of this technique. Citric acid production by immobilized A. niger has been performed on a laboratory scale with the use of calcium alginate gel [143,178], polyacrylamide gel [147,179], polyurethane foam [142,180,181] and cryopolymerized acrylamide [182]. The profitable effect of the immobilization of A. niger mycelium in view of the citric acid recovery from the fermentation broth depends on the type of the support material and process conditions. Solid-state fermentation The solid culture process is completed within 96 h under optimal conditions (8). The most common organism used in solid-state fermentation is A. niger.

324 However, there have also been reports with yeasts. The strains with large requirements of nitrogen and phosphorus are not ideal microorganisms for solid culture due to lower diffusion rate of nutrients and metabolites occurring at lower water activity in solid-state process. The presence of trace elements may not affect citric acid production so harmfully as it does in submerged fermentation, thus, substrate pretreatment is not required. This is one of the important advantages of the solid culture [183,184]. Product recovery On completion of the citric acid fermentation, the obtained solution contains, besides the desirable product, mycelium and varying amounts of other impurities, e.g., mineral salts, other organic acids, proteins, etc. The method of citric acid recovery from the fermentation broth may vary depending on the technology and raw materials used for the production [185]. Separation of biomass from fermentation broth takes place in first step of the recovery process. Separated mycelia retain about 15 per cent of the citric acid formed during fermentation. The mycelia are then washed and pressed in filter presses dried and often used as a protein-rich feed for cattle. If oxalic acid is formed as a side product due to suboptimal fermentation control, it can be precipitated as calcium oxalate at pH below 3.0 [186]. Surface process In the surface process, the fermentation fluid is drained off the trays and hot water is introduced to wash out the remaining amount of citric acid from the mycelial mats. Although it is a relatively simple procedure in the case of surface fermentation, where biomass is in the form of 2–3 cm cover blanket on the substrate surface. Thorough washing at this stage is necessary, because the mycelium retains about 15 per cent of the product formed in the fermentation. In this vessel the mycelium is heated to about 1001C by steam. The solution containing 2–4 per cent of citric acid is added to the fermentation fluid, whereas the filtration cake, containing not more than 0.2 per cent of citric acid, is dried to yield a protein-rich feedstuff [186]. Submerged fermentation In the submerged fermentation the mycelium is far more difficult to separate from the fermentation broth. After the fermentation process is completed the mycelium-containing broth is heated to a temperature of 701C for about 15 min, to obtain partial coagulation of proteins, and then filtered. Rotating vacuum drum or belt discharge filters or in centrifuges are used in this case [40]. If the mycelium is to be used as a feedstuff, the filter aid must also be digestible, e.g., from cellulosic materials. If during the fermentation process

325 oxalic acid is formed, it has to be removed from the broth. This is usually achieved by increasing the pH of the fermentation fluid with the calcium hydroxide to pH ¼ 2.7–2.9 at a temperature of 70–751C. Calcium oxalate thus precipitated may be removed from the solution by nitration or centrifugation, and the citric acid remains in solution as the mono-calcium citrate. Recovery of citric acid from pretreated fermentation broth may be accomplished by several procedures: classical method of precipitation, solvent extraction, ion-exchange and some more sophisticated methods such as electrodialysis, ultra- and nanofiltration or application of liquid membranes [186]. Precipitation The standard method of citric acid recovery has involved precipitating the insoluble tri-calcium citrate by the addition of an equivalent amount of lime to the citric acid solution. Successful operation of the precipitation depends on citric acid concentration, temperature, pH and rate of lime addition. To obtain large crystals of high purity, milk of lime containing calcium oxide (180–250 kg/m3) is added gradually at a temperature of 901C or above and pH below, but close to, 7. The concentration of citric acid in the solution should be above 15 per cent. The process of neutralization usually lasts about 120–150 min. The minimum loss of citric acid due to solubility of calcium citrate is 4–5 per cent. Calcium citrate is then filtered off and subsequently treated with concentrated sulphuric acid (60–70 per cent) to obtain citric acid and the precipitate of calcium sulphate (gypsum). After filtering off the gypsum a solution of 25–30 per cent of citric acid is obtained. The filtrate is treated with activated carbon to remove residual impurities or may be purified in ion-exchange columns. The purified solution is then concentrated in vacuum evaporators at temperature below 401C (to avoid caramelization), crystallized. A conventional crystallization scheme consists of a batch vacuum-pan evaporator or a forced circulating evaporator coupled with auxiliary tankage and appropriate centrifuge equipment. Within these systems, the crystals formed are separated by centrifugation and the mother liquor is fed back to the activated carbon stage. Both batch and continuous units have been employed in this cooperation depending of process adaptability and economics [40]. The drying of citric acid monohydrate is usually performed in conventional rotary drying equipment or in fluidized bed dryers. As anhydrous citric acid is hygroscopic, care must be taken to achieve the final moisture specification during drying and to avoid storage in areas of high temperature and humidity [186]. The disadvantage of this technology is the large amount of lime required for citric acid neutralization and of sulphuric acid for calcium citrate decomposition. Moreover, it results in the formation of large amounts of liquid

326 and solid wastes (solution after calcium citrate filtration and gypsum). For 1 ton of citric acid, 579 kg of calcium hydroxide, 765 kg of sulphuric acid and 18 m3 of water are consumed and approximately 1 ton of waste gypsum is produced [186]. With the aim of decreasing the amount of lime and sulphuric acid by about one–third, [187] has proposed recovery of citric acid by precipitation of di-calcium acid citrate. An additional advantage of this method is that dicalcium acid citrate has a definite crystalline structure and washes cleaner than the amorphous tri-calcium citrate. Solvent extraction An alternative method of citrate-free recovery of citric acid from a fermentation broth is extracted by means of a selective solvent, which is insoluble or only sparingly soluble in the aqueous medium [188–190]. The solvent should be chosen so as to extract the maximum amount of citric acid and the minimum amount of impurities. The citric acid can then be recovered from the extract either by distilling off the solvent or by washing the extract with the water. From the aqueous solution purified citric acid is subsequently crystallized by concentration. Various organic solvents which are partly or wholly immiscible with water, such as certain aliphatic alcohols, ketones, ethers or esters [190,191], organophosphorus compounds, such as tri-n-butylphosphate (TBP) [192] and alkylsulphoxides [193] and water-insoluble amines or a mixture of two or more of such amines are used [194–196]. Ion exchange The efficiency of the ion-exchange separation process may be greatly enhanced by applying a simulated moving bed counter-current flow system. It consists of at least two static beds, connected with appropriate valving so that the feed mixture is passed through one adsorbent bed while the desorbent material can be passed through the other. Progressive changes in the function of each ion-exchange bed simulate the counter-current movement of the adsorbent in relation to liquid flow. In such a system, the adsorption and desorption operations are continuously taking place, which allows both continuous production of an extract and a raffinate stream and the continual use of feed and desorbent streams [197]. The disadvantage of the ion-exchange method may be seen in the fact that elution of citric acid from the adsorption bed may require a large amount of desorbent, due to the tailing effect known in chromatography, causing considerable dilution of the resulting citric acid solution. The periodical regeneration of the ion-exchange resins by inorganic bases may also be a source of unwanted effluent wastes.

327 Liquid membranes Liquid membranes containing mobile carriers consist of an inert, microporous support impregnated with a water-immiscible, mobile ion-exchange agent. The mobile carrier, which is held in the pores of the support membrane by capillarity, acts as a shuttle, picking up ions from an aqueous solution on one side of the membrane, carrying them across the membrane and releasing them to the solution on the opposite side of the membrane [198]. For citric acid separation by liquid membranes, the tertiary amines which give the best results also in solvent extraction can also be used. Recently, more sophisticated methods of citric acid separation with the application of liquid membranes are being developed [199–201]. Microporous hollow fibres Microporous hollow fibres have been employed by Basu and Sirkar [202]. In this case the permeator consists of two sets of identical hydrophobic microporous hollow fibres. One set carries the feed solution of citric acid and the other the strip solution flowing in the lumen. The organic liquid membrane is contained in the shell side between these two sets of hollow fibres. This technique has been shown to be promising for citric acid separation even in the large scale, as the extent of citric acid recovery of up to 99 per cent was linear with the membrane area, suggesting easy scale-up [186]. Electrodialysis This process enables separation of salts from a solution and their simultaneous conversion into the corresponding acids and bases using electrical potential and mono- or bipolar membranes. Bipolar membranes are special ion exchange membranes which, in an electrical field, enable the splitting of water into H+ and OH2 ions [203]. By integrating bipolar membranes with anionic and cationic exchange membranes, a three- or four-compartment cell may be arranged, in which electrodialytic separation of salt ions and their conversion into base and acid takes place [204]. Before the fermentation solution comes to the electrodialysis, some pretreatment steps are normally necessary: filtration of the broth, removal of ionogenic substances (especially Ca2+ and Mg2+ ions) and neutralization by means of sodium hydroxide. In the subsequent electrodialytic step the sodium citrate solution is converted into base and citric acid, which is simultaneously concentrated and for the most part purified. The produced NaOH may be reused for the neutralization [205]. The energy consumption (excluding pumping) for the separation of 1 kg of citric acid using bipolar membranes is in the range of 6.1 103–7.2 103 kWs

328 [206]. Owing to low mass transfer at low pH values, it is advantageous to adjust the pH of the feed acid stream to 7.5 [207,208].

Ultrafiltration Continuous separation and concentration of citric acid may be also achieved by ultra and/or nanofiltration, verified in a laboratory scale a two-stage membrane process for citric acid recovery from the broth obtained in A. niger cultivation on sucrose. Polysulphone membrane with cut-off 10,000 used in the first stage allowed the product to pass through to the permeate stream, while the retentate stream contained most of the peptides and proteins from the broth. The rejection coefficient for the product in this step was 3 per cent, for the reducing sugars 14 per cent and for the proteins 100 per cent. Tighter nanofiltration membrane with cut-off 200 in the second stage rejected approximately 90 per cent of citric acid and 60 per cent of reducing sugars (mono-saccharides). A similar two-stage membrane technique was adapted by Bohdziewicz and Bodzek [209] for simultaneous separation and concentration of pectinolytic enzymes and citric acid from a fermentation broth. Recovery of citric acid via calcium salt precipitation is a complex process. In this process calcium citrate is formed in further by adding a lime slurry at a neutral pH. After sufficient reaction time, the slurry is filtered and the precipitate washed free of soluble impurities. The resulting calcium citrate is then acidified with sulphuric acid. This reaction converts calcium citrate to calcium sulfate and citric acid in the presence of free sulphuric acid. Calcium sulphate is then filtered and washed free of citric acid solution. Both the calcium citrate and calcium sulphate reactions are generally performed in agitated reactors and filtrated commercially available filtration equipment. The aqueous citric acid solution is demineralized at this step by strong cation exchange resin in the H+ form (Dowex 50) and an anion exchange resin of medium strength. The purified citric acid solution is subsequently evaporated in a multi-stage evaporator at temperature of 401C to avoid caramelization [27]. The clear citric acid solution undergoes a series of crystallization steps to achieve the physical separation of citric acid from the remaining trace impurities. A conventional crystallization scheme consists of a batch vacuumpan evaporator or a forced circulating evaporator coupled with auxiliary tankage and appropriate centrifuge equipment. Within these systems, the crystals formed are separated by centrifugation and the mother liquor is fed back to the activated carbon stage. Both batch and continuous units have been employed in this cooperation depending of process adaptability and economics [40]. The drying of citric acid monohydrate is usually performed in conventional rotary drying equipment or in fluidized bed dryers. As anhydrous citric acid is hygroscopic, care must be taken to achieve the final

329 moisture specification during drying and to avoid storage in areas of high temperature and humidity [27]. Solvent extraction is an alternative recovery process, which involves the extraction of citric acid from fermentation brutish using hydrocarbons such as: n-octanol, C10 or C11 isoparaffin, benzene, kerosene; ethers: n or isobutyleter; esters: n-butylacetate; ketones: methyl isobutylketone [104] or various amines: trilaurylamine [105]. The recovery process by solvent extraction consists of selectively transferring citric acid via a solvent from an aqueous solution containing various by-products to another aqueous solution in which the citric acid is more concentrated and contains substantially less by-products. The final processing steps begin with a different wash of the aqueous solution by the hydrocarbon solvent, followed by the passage of the acid solution through a conventional sequence of evaporator-crystallizer steps to complete the manufacturing process. Anhydrous citric acid and its monohydrate can be stored in dry form without difficulties; however, high humidity and elevated temperatures should be avoided to prevent caking. Therefore, the use of packing materials with a desiccant is suggested [106]. The citric acid recovery process leads to considerable accumulation of waste products. More than 60 per cent of it belongs to gypsum (calcium sulphate). Which still contains potassium hexacyanoferrate, charcoal and organic compounds from molasses making it so unsuitable as a building material. The waste mycelium from submerged and surface fermentation can be dried and used as an animal protein-rich feed or alternatively as fertilizer. Economic aspects Although the surface production process is from the viewpoint of energy requirements, a less expensive, there are a lot of disadvantages in it. This involves larger space requirements for production and isolation, higher steam requirement and higher sterility requests. One of the greatest problems of this production process sterility. Main advantages of the submerged fermentation process are: shorter fermentation time (6–7 days), higher level of process sterility and control of process parameters, simpler process operations, lower space requirements, process reproducibility and higher yields. Schierholt [210] compared the economy of surface and submerged fermentation process for the citric acid production. Capacities of 300 m3 and 150 m3 in 9 days of fermentation time at productivity 72 tons and 12 tons per day were compared. On his work he concludes that the building investment costs connected with the surface fermentation process are 2.5 times higher than those connected with the submerged fermentation. Contrary to this, the expenses on equipment are considerably higher at submerged fermentation, and more than 60 per cent of those expenses consist of complicated component as are bioreactors and

330 more sophisticated instrumental control, which are subject to relatively high wear. The total investment costs for the submerged process are about 25 per cent lower for higher capacities and 15 per cent lower for smaller capacities than for surface fermentation. The more favourable total investment costs for the submerged process are in contrast to considerably higher production costs for any capacity. Especially evident is the high consumption of electric energy, which is about 30 per cent higher as much as that required at surface fermentation. The labour costs in highly developed countries are for surface fermentation considerably higher. On countries where cooling water temperature exceeds 201C, additional expenses for cooling the bioreactors are incurred by installation of cooling aggregates for submerged process. The submerged fermentation is sensitive to short interruptions or breakdowns in aeration, which results not only in loss of yield, but also in total breakdown of the respective batch. At surface fermentation, the resulting citric acid solution or fermentation broth is much more concentrated than at submerged fermentation, effected by higher evaporation rates during fermentation. Production of citric acid by surface solid state or by isolation from citrus juices does not represent a significant percentage on the World scale. Although both processes are from all aspects very cheap, they are in use mostly in the countries with old traditions (Italy, Greece, Asia). Citric acid World production Development of citric acid fermentation industry during the nearly passed century has aroused a great deal of interest. Formerly, the raw material, calcium citrate, was produced almost entirely from citrus products, Italy being by far the largest producer. The bulk of the Italian production of calcium citrate was shipped to England, France and the United States. Because of the development of the fermentation process and the increased output of citrus materials, import in the United States has practically ceased since 1927. The fermentation process has to a large extent developed also in Europe. Large quantities of fermentation-based citric acid have been produced in England, Belgium and Czechoslovakia and probably Russia. The former dominant position occupied by the Italian producers of this commodity has thus been last through new methods introduced by scientific research. The first successful commercial development of the citric acid fermentation process was achieved in the United States. Miles and later Charles Pfizer Company gradually developed in to the World’s leading companies. The United States citric acid production in 1929 was 4,900 tons per year. While in 1978 the production by Miles was 29,000 tons and by Pfizer (U.S.) 42,500 tons, and raised in 1990 to 66,000 tons by Miles and 105,000 tons by

331 Pfizer (U.S.). Pfizer overall production, including the U.S. and other countries (Irish Republic, Nigeria, Taiwan, Argentina), takes about 30 per cent of the World’s citric acid production. From 1978 until 1990 Pfizer increased its production by 34.5 per cent. Although the World’s greatest producers are in the United States, the World’s greatest production continentally belongs to Europe with 250,000 tons per year, produced in 16 countries. The yearly production in North America was in 1990 about 215,000 tons followed by Asia with 66,000 tons, Africa 14,000 tons, Australia about 8,000 tons and South America with 7,000 tons. In Asia, citric acid production is characterized also by the use of traditional solid-state production on the food industry wastes and by submerged technologies based on various yeast strains (Table 1). Citric acid is a commodity chemical produced and consumed throughout the World. It is used mainly in the food and beverage industry, primarily as an acidulant. It is estimated that over 65 per cent of the citric acid produced is consumed for food and beverages. Global production of citric acid in 2004 was about 1.4 million tons estimated by Business Communications Co. but in 2005 it was about 1,600 thousand metric tons. The majority of production capacity and consumption was in China, Western Europe and the United States. China is estimated to account for at least half of the global production capacity, while Western Europe and the United States combined account for about a third. Western Europe, the United States and China combined are estimated to account for 65–70 per cent of global citric acid consumption. The citric acid industry continues to be influenced by increased supply from China and abundant global capacity. In recent years, plant closures have occurred as a result of competition, and prices have continued to decline.

Table 1. The industrial World production of citric acid in 1990 [152]. The greatest World producers (tons) USA Belgium Austria Ireland Germ. Fed. Rep. Italy Mexico Soviet Union Great Britain Israel World production (1990) a

Surface production.

Pfizer Miles Citrique Belgea Jungbunzlauer Pfizer Biochemic Ladenburg Biacor Quimica Mexama State Authoritya John & E. Sturge Cadot Petroch

105,000 66,000 55,000 40,000 36,000 30,000 25,000 19,000 18,000 14,000 8,000 598,000

332 From the first industrial fermentations, the World’s production has increased exponential from 5,000 tons in 1929 until the present production in 2006 of about 1,600,000 tons per year (Fig 1). [152,153]. In 2005, the top six producing companies accounted for about 53 per cent of the World’s total capacity for citric acid. China’s capacity was 800 thousand metric tons (50 per cent of World capacity), most of which is unrefined citric acid. The global production capacity of the World’s six largest citric acid-producing companies, plus China (which produces mostly unrefined citric acid), is shown in Fig. 2. Figure 3 shows World’s consumption of citric acid by region in 2005. Over half of the global consumption of citric acid is used for the beverage industry. The food industry consumes about 15–20 per cent, followed by detergent and soaps (15–17 per cent), pharmaceuticals and cosmetics (7–9 per cent) and industrial uses (6–8 per cent). In the United States, the citric acid market will continue to grow mainly as a result of growth in the beverage market. New product introductions and continued use in diet colas, fruit-flavored waters, iced teas and sports drinks will lead to higher growth. Liquid detergent growth will also contribute to growing citric acid demand. New growth will also be seen in industrial applications, as renewable resources continue to grow. In Canada, citric acid use may increase significantly as a result of use in oil recovery. In Europe, the market is impacted by price, which is driven down by a combination of strong competition from Chinese product and an abundance of global capacity. European producers are contending with imports from Chinese producers. While the average prices were declining, Chinese imports of citric acid to Western Europe grew from roughly 46 thousand metric tons in 1999 to 109 thousand metric tons in 2004. Chinese competition is mainly in

Fig. 1. Citric acid World production.

333

Fig. 2. World capacity for citric acid producer in 2006 [153].

Fig. 3. World consumption of citric acid by region in 2005 [153].

citric acid monohydrate (solid form) and among citrates in sodium citrate, the most-used-form of citric acid salts. Chinese suppliers have started to adopt Western-pricing practices, which might lead to a more stabilized price. In the future, European manufacturers of citric acid and citrates might concentrate on the production of citric acid solutions (using solid form produced in-house or imported) and/or higher-value citrates. The citric acid market continues to face pressure from Asian imports and increased global supply causing selling prices to decline. However, tight supplies from Europe caused by closures, and high energy and freight costs are

334 some of the factors leading to higher citric acid prices. The overall global market for citric acid is expected to grow at an average annual rate of 3.5–4.5 per cent in the next few years. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19.

20. 21.

22. 23. 24. 25. 26. 27.

Wehmer C. Darstellung von Citronensaure mittels Ga¨rung. Chem Zentr 1893;2:457–462. Grimoux E and Adam P. Synthese del’acide citrique. CR Acad Sci Paris 1880;90:1252–1256. Wehmer C. Note sur la fermentation citrique. Bull Soc Chem Fr 1893;9:728–732. Wehmer C, U.S. Patent 515033, 1984. Maze’ P and Perrier A. The role of microbes in alcoholic fermentation attributed to zymase. Ann Inst Pasteur 1904;18:533–534. Buchner E and Wu¨stenfeld H. Citric acid fermentation by Citromycetes. Biochem Ztg 1909;17:395–442. Zahorsky B, U.S. Patent 1 065358, 1913. Thom C and Currie JN. Oxalic acid production of special Aspergillus. J Agric Res 1916;7:1–6. Currie JN. The citric acid fermentation of Aspergillus niger. J Biol Chem 1917;32:15–37. Mezzadroli G, French Patent 833 631, 1938. LHC Perquin, Dissertation, Tech. University Delft, Delft, The Netherlands, 1938. Ro¨hr M, Kubicek CP and Kominek J, Citric acid. In: Biotechnology, Vol. 3, HJ Rehm and G Reed (ed), Verlag Chem, Weinheim, 1983, p. 420. Tanaka K, Kimura K and Yamaguchi K, Japanese Patent 13677, 1968 Fukui S and Tanaka A. Production of useful compounds from alkane media in Japan. Adv Biochem Eng 1980;17:1–35. Crueger W and Crueger A, In ‘‘Biotechnology,’’ Science Tech. Inc. (1984) Sunderland, MA, USA. Buchard EF and Merrit EG, Citric acid. In: Kirk-Othmers Encyclopedia of Chemical Technology, 3rd edn, Vol. 6, Wiley, New York, 1979, p. 150. Chem. Abstr. Service, Ca File, Columbus, OH, USA, 1990. Chem-Intell, Chem. Int. Service, London, Great Britain, 1990. Moledina KH, Regenstein JM, Baker RC and Steinkraus KH. Effects of antioxidants and chelators on the stability of frozen stored mechanically deboned flounder meat from racks after filleting. J Food Sci 1977;42(3):759–764. Wells CE, Martin DC and Tichenor DA. J Am Dietetic Assoc 1972;61(6):665–668. Bhatacharya S, Chowdhury NK and Basu UP. Vitamin A in solution. VI. Stability in relation to concentration and chemical forms of vitamin A and solvents. J Indian Chem Soc 1954;31:231–240. MacDonald LH. Comparative testing of preservative systems. Am Perfum Essential Oil Rev 1961;76(7):22–23. Krummel HK and Gault TW, U.S. Patent 3, 985 669, 1976. Horner DC. Citric, tartaric, and gluconic acids in metal finishing. Electroplat Met Finish 1968;21(3):75–80. Smith JE, Nowakowska-Waszchuk A and Anderson JG. Organic acid produced by mycelial fungi. Ind Asspects Biochem 1974;30(1):284–297. French Patent 813, 548, 1937. U.S. Patent 2, 474092, 1949.

335 28. Kawasaki M and Harada K. Fixing properties of reaction products prepared from thiocyanate and salts of carboxylic acids. Nippon Shaskin Gakkaishi 1975;38(1):37–40. 29. Hushedeck HR, U.S. Patent 3, 212 928, 1965. 30. Sinha J, Bae JT and Park JP. Changes in morphology of Paecilomyces japonica and their effect on broth rheology during production of exo-biopolymers. Appl Microbiol Biot 2001;56(1–2):88–92. 31. Wolschek MF and Kubicek CP. Biochemistry of citric acid accumulation by Aspergillus. In: Citric Acid Biotechnology, Kristiansen B, Mattey M and Linden J (eds), London, Taylor & Francis Press, 1999, pp. 33–54. 32. Mattey M. Biochemistry of citric acid production by yeasts. In: Citric Acid Biotechnology, Kristiansen B, Mattey M and Linden J (eds), London, Taylor & Francis Press, 1999, pp. 11–32. 33. Whittaker A and Long PA. Fungal pelleting. Process Biochem 1973;8:27–31. 34. Martin SM. Ind Eng Chem 1957;49:1231. 35. Clark DS. Submerged citric acid fermentation of ferrocyanide-treated beet molasses: morphology of pellets of Aspergillus niger. Canad J Microbiol 1962;8:133–136. 36. Chmiel A. Kinetics of citric acidproduction by pre-cultivated mycelium of Aspergillus niger. Trans Br Mycol Soc 1977;68(3):403–406. 37. Cimerman A, Johanides V and Sˇkafar S. In: Abstracts of Fifth International Fermentation Symposium, Vol. 405, H. Dellweg (ed), Berlin, VCH-Verlag, 1976. 38. Kutermankiewicz M and Lesˇ niak W. Premysl Ferm. Owocowo-Warzywny 1984;11:26. 39. Martin JF and Demain AL. In: The Filamentous fungi, Vol. 3, Smith JE and Berry DR (eds), London, Eduard Arnold, 1977. 40. Noyes R, Citric acid production process. Chem Proc Rev 1969;37:42. 41. Berovicˇ M, PhD Thesis, University of Ljubljana, Ljubljana, 1986. 42. Berovicˇ M, MSc Thesis, University of Ljubljana, Ljubljana, 1979. 43. Waksman SA and Karow EO, U.S. Patent 2,394031, 1946. 44. Prescott SC and Dunn CG. Industrial Microbiology, New York, McGraw-Hill, 1959. 45. Perlman D and Sich CJ. Fungal synthesis of citric fumaric and itaconic acids. Prog Ind Microbiol 1960;2:168–194. 46. Mayrath J. Citric acid production. Proc Biochem 1967;2:25–27. 47. Krebs HA and Johnson WA. Citric acid in intermediate metabolism in animal tissues. Enzymologia 1937;4:148–156. 48. Lehninger AL. Biochemistry, 2nd edn, New York, North Pub. Inc., 1975. 49. Torres NV. Modelling approach to control of carbohydrate metabolism during citric acid accumulation by Aspergillus niger: I. model definition and stability of steady state. Biotechnol Bioeng 1994;44:104–111. 50. Torres NV. Modelling approach to control of carbohydrate metabolism during citric acid accumulation by Aspergillus niger: II. sensitivity analysis. Biotechnol Bioeng 1994;44:112–118. 51. Torres NV, Riol-Cimas JM, Wolschek M and Kubicek CP. Glucose transport by Aspergillus niger: the low affinity carrier is only formed during growth on high glucose concentrations. Appl Microbiol Biotechnol 1996;44:790–794. 52. Mischak M, Kubicek CP and Ro¨hr M. Citrate inhibition of glucose uptake in Aspergillus niger. Biotechnol Lett 1984;6:425–430. 53. Wayman FM and Mattey M. Simple diffusion is the primary mechanism for glucose uptake during the production phase of the Aspergillus niger citric acid process. Biotechnol Bioeng 2000;67:451–456.

336 54. Peksel A, Torres NV, Liu J and Kubicek CP. 13C-NMR analysis of glucose metabolism during citric acid production by Aspergillus niger. Appl Microbiol Biotechnol 2002;58:157–163. 55. Arisan-Atac I, Wolschek MF and Kubicek CP. Trehalose-6-phosphate synthase A affects citrate accumulation by Aspergillus niger under conditions of high glycolytic flux. FEMS Microbiol Lett 1996;140:77–83. 56. Ruijter GJG, Panneman H and Visser J. Overexpression of phosphofructokinase and pyruvate kinase in citric acid-producing Aspergillus niger. Biochim Biophys Acta 1997;1334:317–326. 57. Ro¨hr M, Zehentgruber O and Kubicek CP. Kinetics of biomass formation and citric acid production by Aspergillus niger on pilot plant scale. Biotechnol Bioeng 1981;23:2433–2445. 58. Papagianni M, Wayman F and Mattey M. Fate and role of ammonium ions during fermentation of citric acid by Aspergillus niger. Appl Environ Microbiol 2005;71:7178–7186. 59. Legisˇ a M and Mattey M. Glycerol synthesis by Aspergillus niger under citric acid accumulating conditions. Enzyme Microb Technol 1986;8:607–609. 60. Ro¨hr M, Kubicek CP, Zehentgruber O and Orthofer R. Accumulation and partial re-consumption of polyols during citric acid fermentation by Aspergillus niger. Appl Microbiol Biotechnol 1987;27:235–239. 61. Legisˇ a M and Kidricˇ J. Initiation of citric acid accumulation in the early stages of Aspergillus niger growth. Appl Microbiol Biotechnol 1989;31:453–457. 62. Karaffa L and Kubicek CP. Aspergillus niger citric acid accumulation: do we understand this well working black box? Appl Microbiol Biotechnol 2003;61: 189–196. 63. Legisˇ a M and Mattey M. Glycerol as an initiator of citric acid accumulation in Aspergillus niger. Enzyme Microb Technol 1986;8:258–259. 64. Arisan-Atac I and Kubicek CP. Glycerol is not an inhibitor of mitochondrial citrate oxidation by Aspergillus niger. Microbiology 1996;142:2937–2942. 65. Cordewener J, Busink R and Visser J. A permeabilized cell assay system for studying enzyme regulation and localisation in Aspergillus niger. J Microbiol Methods 1989;10:231–240. 66. Mattey M. Citrate regulation of citric acid production in Aspergillus niger. FEMS Microbiol Lett 1977;2:71–74. 67. Legisˇ a M and Mattey M. Citrate regulation of the change in carbohydrate degradation during the initial phase of the citric acid production by Aspergillus niger. Enzyme Microbiol Technol 1988;8:33–36. 68. Mesojednik S and Legisˇ a M. Posttranslational modification of 6-phosphofructo-1-kinase in Aspergillus niger. Appl Environ Microbiol 2005;71:1425–1432. 69. Mlakar T and Legisˇ a M. Citrate inhibition resistant from of 6-phosphofructo-1-kinase from Aspergillus niger. Appl Environ Microbiol 2006;72:4515–4521. 70. Poorman RA, Randolph A, Kemp RG, et al. Evolution of phosphofructokinase-gene duplication and creation of new effector sites. Nature 1984;309:467–469. 71. Kubicek-Pranz EM, Mozelt M, Ro¨hr M and Kubicek CP. Changes in the concentration of fructose-2,6-bisphosphate in Aspergillus niger during stimulation of acidogenesis by elevated sucrose concentration. Biochim Biophys Acta 1990;1033:250–255. 72. Legisˇ a M and Bencˇina M. Evidence for the activation of 6-phosphofructo-1-kinase by cAMP-dependent protein kinase in Aspergillus niger. FEMS Microbiol Lett 1994;118:327–333.

337 73. Gradisˇ nik-Grapulin M and Legisˇ a M. A spontaneous change in the intracellular cyclic AMP level in Aspergillus niger is influenced by the sucrose concentration in the medium and by light. Appl Environ Microbiol 1997;63:2844–2849. 74. Thevelein JM and de Winde JH. Novel sensing mechanisms and targets for the cAMPprotein kinase. A pathway in the yeast Saccharomyces cerevisiae. Mol Microbiol 1999;33:904–918. 75. Jernejc K and Legisˇ a M. A drop of intracellular pH stimulates citric acid accumulation by some strains of Aspergillus niger. J Biotechnol 2004;112:289–297. 76. Hesse SJ, Ruijter GJG, Dijkema C and Visser J. Intracellular pH homeostasis in the filamentous fungus Aspergillus niger. Eur J Biochem 2002;269:3485–3494. 77. van den Hombergh JP, van de Vondervoort PJ, Fraissinet-Tachet L and Hawkins AR. Aspergillus as a host for heterologous protein production: the problem of proteases. Trends Biotechnol 1997;15:256–263. 78. Ma H, Kubicek CP and Ro¨hr M. Metabolic effects of manganese deficiency in Aspergillus niger: evidence for increased protein degradation. Arch Microbiol 1985;141:266–268. 79. Schreferl G, Kubicek CP and Ro¨hr M. Inhibition of citric acid accumulation by manganese ions in Aspergillus niger mutants with reduced citrate control of phosphofructokinase. J Bacteriol 1986;165:1019–1022. 80. Ro¨hr M and Kubicek CP. Regulatory aspects of citric acid fermentation by Aspergillus niger. Process Biochem 1981;16:34–37. 81. Kubicek CP and Ro¨hr M. The role of the tricarboxylic acid cycle in citric acid accumulation by Aspergillus niger. Eur J Appl Microbiol Biotechnol 1978;5:263–271. 82. Schreferl-Kunar G, Grotz M, Ro¨hr M and Kubicek CP. Increased citric acid production by mutants of Aspergillus niger with increased glycolytic capacity. FEMS Microbiol Lett 1989;59:297–300. 83. Jaklitsch WM, Kubicek CP and Scrutton MC. Intracellular organisation of ctrate production in Aspergillus niger. Can J Microbiol 1991;37:823–827. 84. Ma H, Kubicek CP and Ro¨hr M. Malate dehydrogenase isoenzyme in Aspergillus niger. FEMS Microbiol Lett 1981;12:147–151. 85. Kubicek CP, Zehentgruber O and Ro¨hr M. An indirect method for studying fine control of citric acid accumulation by Aspergillus niger. Biotechnol Lett 1979;1:47–52. 86. Kubicek CP. The role of the citric acid cycle in fungal organic acid fermentation. Biochem Soc Symp 1988;54:113–126. 87. Kubicek CP, Zehentgruber O, El-Kalak H and Ro¨hr M. Regulation of citrate synthase from the citric acid producing fungus Aspergillus niger. Biochim Biophys Acta 1980;615:449–457. 88. Dawson MW, Maddox IS, Boag IF and Brooks JD. Application of fed-batch culture to citric acid production by Aspergillus niger: the effects of dilution rate and dissolved oxygen tension. Biotechnol Bioeng 1988;32:220–226. 89. Kirimura K, Hirowatari Y and Usami S. Alterations of respiratory systems in Aspergillus niger under the conditions of citric acid fermentation. Agric Biol Chem 1986;51:1299–1303. 90. Kirimura K, Yoda M, Shimizu H, Sugano S, Mizuno m, kino K and Usami S. Contribution of cyanide-insensitive respiratory pathway, catalyzed by the alternative oxidase, to citric acid production in Aspergillus niger. Biosci Biotechnol Biochem 2000;64(10):2034–2039.

338 91. Kirimura K, Matsui T, Sugano S and Usami S. Enhancement and repression of cyanide-insensitive respiration in Aspergillus niger. FEMS Microbiol Lett 1996;141:251–254. 92. Kirimura K, Yoda M and Usami S. Cloning and expression of the cDNA encoding an alternative oxydase gene from Aspergillus niger WU-2223L. Curr Genet 1999;34:472–477. 93. Netik A, Torres NV, Riol JM and Kubicek CP. Uptake and export of citric acid by Aspergillus niger is reciprocally regulated by manganese ions. Biochim Biophys Acta 1997;1326(2):287–294. 94. Burgstaller W. Thermodynamic boundary conditions suggest that a passive transport step suffices for citrate excretion in Aspergillus and Penicillium. Microbiology 2006;152(3):887–893. 95. Nierman WC, Pain A, Anderson MJ, Wortman JR, Kim HS, Arroyo J, Berriman M, Abe K, Archer DB, Bermejo C, et al. Genomic sequence of the pathogenic and allergenic filamentous fungus Aspergillus fumigatus. Nature 2005;438:1151–1156. 96. Galagan JE, Calvo SE, Cuomo C, Ma LJ, Wortman JR, Batzoglou S, Lee SI, Basturkmen M, Spevak CC, Clutterbuck J, et al. Sequencing of Aspergillus nidulans and comparative analysis with A. fumigatus and A. Oryzae. Nature 2005;438:1105–1115. 97. Machida M, Asai K, Sano M, Tanaka T, Kumagai T, Terai G, Kusumoto K, Arima T, Akita O, Kashiwagi Y, et al. Genome sequencing and analysis of Aspergillus oryzae. Nature 2005;438:1157–1161. 98. David H, Akesson M and Nielsen J. Reconstruction of the central carbon metabolism of Aspergillus niger. Eur J Biochem 2003;270:4243–4253. 99. Roehr M. A century of citric acid fermentation and research. Food Technol Biotechnol 1998;36(3):167–173. 100. Tomlinson N, Cambell JJR and Trussel PC. The influence of zinc, copper, iron and manganese on production of citric acid by A. niger. J Bacteriol 1950;59:217–224. 101. Kubicek CP and Roehr M. Citric acid fermentation. CRC Crit Rev Biotechnol 1986;3(4):331–373. 102. Shu P and Johnson MJ. The interdependence of medium constituents in citric acid fermentation by submerged fermentation. J Bacteriol 1948;56:577–586. 103. Ruijter GJG, Kubicek CP and Visser J. Production of organic acids by fungi. In: The Mycota X Industrial Applications, Osiewacz HD (ed), Heidelberg, Berlin, Springer, 2002, pp. 213–227. 104. Ruijter GJG, Kubicek CP and Visser J. Metabolic engineering of the glycolytic pathway in A. niger. Food Technol Biotechnol 1998;36(3):185–189. 105. Clark DS, Ito K and Horitsu H. Effects of manganese and other heavy metals on submerged citric acid fermentation of molasses. Biotechnol Bioeng 1966;8:465–471. 106. Berovicˇ M, Vodopivec M and Milicˇicˇ S. The influence of manganese ions on Aspergillus niger biomass and citric acid biosynthesis in repeated fed batch fermentation. Chem Biochem Eng Q 2006;20(3):281–284. 107. Zidwick MJ. Organic acids. In: Biotechnology of Filamentous Fungi, Finkelstein DB and Ball C (eds), Boston, Butterworth-Heinemann, 1992, pp. 303–334. 108. Orthofer R, Kubicek CP and Roehr M. Lipid levels and manganese deficiency in citric acid producing strains of A.niger. FEMS Microbiol Lett 1979;5:403–406. 109. Mirminachi F, Zang A and Roehr M. Citric acid fermentations and heavy metal ions I. Effects of iron, manganese and copper. Acta Biotechnol 2002;22(3–4):363–373. 110. Zang A and Roehr M. Citric acid fermentations and heavy metal ions II. The action of elevated manganese ion concentration. Acta Biotechnol 2002;22(3–4):375–382.

339 111. Haq I-U, Ali S, Quadeer MA and Iqbal J. Effect of copper ions on mould morphology and citric acid productivity by Aspergillus niger using molasses based media. Process Biochem 2001;4:28–34. 112. Gupta S and Sharma CB. Biochemical studies of citric acid production and accumulation by A. niger mutants. World J Microbiol Biotechnol 2002;18(5):379–383. 113. Lesˇ niak W, Pietkiewicz J and Podgorski W. Citric acid fermentation from starch and dextrose syrups by trace metal resistant mutant of A. niger. Biotechnol Lett 2002;24(13):1065–1067. 114. Lesˇ niak W. In: Citric Acid Biotechnology: Fermentation Substrates, Kristiansen B, Mattey M and Linden J (eds), London, Taylor Francis Ltd., 1999, pp. 149–159. 115. Lesniak W, Podgorski W and Pietkiewicz J. Possibilities of application of glucose hydrol for citric acid fermentation. Pol Przem Ferm i Owoc-Warzyw 1986;30(6):22–25. 116. Lesniak W, Studies on Submerged Citric Acid Fermentation, PhD Thesis, University of Wroclaw, Poland, 1972. 117. Kutermankiewicz M, Lesniak W and Bolach E. Use of dense sugar juice in the submerged fermentation of citric acid. Wytwornia Kwasu Cytrynowego Raciborz, Pol Przem Ferm i Owoc-Warzyw 1980;24(6):27–31. 118. Pietkiewicz J, Podgorski W and Lesniak W, Proceedings of the International Conference on Advances in Citric Acid Technology, Bratislava, Slovak Republic, 1996, p. 9. 119. Lesniak W, Podgorski W and Pietkiewicz J. Przem Ferm i Owoc-Warzyw 1986;6:22. 120. Maldonaldo P and Charpentier M. German Patent 2551469, 1975. 121. Kimura K and Nakanishi T. German Patent 2 065 206, 1985. 122. Wojtatowicz M and Sobieszczanski J. Synthetic hydrocarbons from coal processing as a raw material for the microbiological industry. II. Selection of some components of the medium for Candida scottii A-61. Pol Acta Microbiol Polonica 1981;30(1):69–77. 123. Ikeno Y, Masuda M, Tanno K, Oomori I and Takahashi N. Citric acid production from various raw materials by yeasts. Hakko Kogaku Zasshi. J Ferment Technol 1975;53(10):752–756. 124. Gutcho SJ. Chemicals by Fermentation, Park Ridge, NY, Noyes Data Corporation, 1973. 125. Elimer E, Studies on Use of Plant Fats for Citric Acid Production by Aspergillus niger, PhD Thesis, University of Wroclaw, Poland, 1994. 126. Kristiansen B and Sinclair CG. Production of citric acid in continuous culture. Biotech Bioeng 1979;21:297–303. 127. Kristiansen B and Charley R. Continuous process for citric acid production. In: Advanced Biotechnology, Vol. 1, Moo-Young M (ed), New York, Pergamon Press, 1981, pp. 69–76. 128. Saha ML and Takahashi F. Continuous citric acid fermentation by magnetic rotating biological contactors using, Aspergillus niger AJ 117173. J Ferment Bioeng 1997;84(3):244–248. 129. Sommariva C, Zilli M and Converti A. Continuous citric acid fermentation in a supported biomass reactor. Chem Eng Technol 1997;20(5):348–353. 130. Godo S, Klein J and Bales V, Mixing time in airlift reactors during citric acid fermentation. Bioprocess Eng 1999;21(3):245–248. 131. Godo S, Klein J and Polakovic M. Periodical changes of input air flowrate – a possible way of improvement of oxygen transfer and liquid circulation in airlift bioreactors. Chem Eng Sci 1999;54(21):4937–4943. 132. Zheng YG, Wang Z and Chen XL. Citric acid production from the mash of dried sweet potato with its dregs by Aspergillus niger in an external-loop airlift bioreactor. Process Biochem 1999;35(3–4):237–242.

340 133. Berovic M and Popovic M. Characterisation of gas mixed bioreactors in submerged citric add fermentation. Chem Biochem Eng Quat 2001;15(2):65–69. 134. Saha ML, Sakai Y and Takahashi F. Citric acid fermentation by magnetic drum contactor: use of methanol and ethanol for higher production. J Biosci Bioeng 1999;87(3):394–396. 135. Anderson JG, Blain JA, Divers M and Todd JR. Use of the disc fermenter to examine production of citric acid by Aspergillus niger. Biotechnol Lett 1980;2:99–104. 136. Ro¨hr M, Kubicek CP and Kominek J. Citric acid. In: Biotechnology, Vol. 2, Rehm H-J and Reed G (eds), Weinheim, VCH, 1983, pp. 420–454. 137. Kubicek CP. Organic acids. In: Basic Biotechnology, Ratledge C and Kristiansen B (eds), Cambridge, Cambridge Press, 2001, pp. 305–325. 138. Thalasso F, Naveau H and Nyns EJ. Effect of dry periods in a ‘‘mist-foam’’ bioreactor designed for gaseous substrate. Environ Technol 1996;17(8):909–913. 139. Heinich M and Rehm HJ. Formation of gluconic acid at low pH-values by free and immobilized Aspergillus niger cells during citric acid fermentation. Eur J Appl Microbiol Biotechnol 1982;28:1024–1036. 140. Lee YH, Lee CW and Chang HN. Citric acid production by Aspergillus niger, immobilized on polyutethane foam. Appl Microbiol Biotechnol 1989;30:141–143. 141. Eikmeier H and Rehm HJ. Production of citric acid with immobilized Aspergillus niger. Appl Microbiol Biotechnol 1984;20:365–370. 142. Vaija J, Linko PYY and Linko P. Citric acid production with alginate bead entrapped Aspergillus niger, ATCC 9142. Appl Microbiol Biotechnol 1982;7:51–54. 143. Cruz R, Cruz DV, Belini MZ, Belote JG and Viera CR. Production of fructooligosaccharides by mycelia of Aspergillus japonicus, immobilized in polyacrylamide gels. Bioresource Technol 1998;65:139–143. 144. Horitsu H, Adachi S, Takahashi Y, Kawai K and Kawano Y. Production of citric acid by Aspergillus niger, immobilized in polyacrylamide gels. Appl Microbiol Biotechnol 1985;22:8–12. 145. Mittal Y, Mishra IM and Varshney BS. Characterisation of metabolically active development stage of Aspergillus niger cells immobilized in polyacrylamide gels. Biotechnol Lett 1993;15:41–46. 146. Borglum GB and Marshall JJ. The potential of immobilized biocatalyst for production of industrial chemicals. Appl Microbiol Biotechnol 1984;9:117–230. 147. Khare SK, Jha K and Gandhi AP. Use of agarose-entrapped Aspergillus niger cells from production of citric acid from soy whey. Appl Microbiol Biotechnol 1994;41:571–573. 148. Fuji N, Yasuda K and Sakahibara M. Effect of volume ratio of cellulose carriers and time interval of repeated batch culture on citric acid productivity by immobilized A. niger. J Ferment Bioeng 1994;78:389–393. 149. Sankpal NV, Joshi AP and Kulkarni BD. Citric acid production by Aspergillus niger, immobilized on cellulose microfibris influence of morphology and fermenter conditions on productivity. Process Biochem 2001;36:1129–1139. 150. Liu F, Li W, Ridgway D, Gu T and Moo-Young M. Inhibition of extracellular protease secretion by Aspergillus niger using cells immobilization. Biotechnol Lett 1998;20:539–542. 151. Papiaganni M, Joshi N and Moo-Young M. Comparative studies on extracellular protease secretion and glucoamylase production by free and immobilized Aspergillus niger cultures. J Ind Microbiol Biotechnol 2002;29:259–263. 152. http://www.sriconsulting.com/CEH/Public/RE-Hosteports/636.5000/?Abstract.html 153. http://www.bccresearch.com/editors/RGA-115.html

341 154. Shu P, Funk F and Neish AC. Mechanism of citric acid formation from glucose by Aspergillus niger. Canad J Biochem Physiol 1954;32:68–80. 155. Bomstein R and Johnson MJ. The mechanism of formation of citrate and oxalate by Aspergillus niger. Biol Chem 1952;198:143–152. 156. Cleland WW and Johnson MJ. Tracer experiments on the mechanism of citric acid formation by Aspergillus niger. J Biol Chem 1954;208:679–689. 157. Ramakrishnan CV, Steel R and Lenz CP. Mechanism of citric acid formation and acccumulation in Aspergillus niger. Arch Biochem Biophys 1955;55:270. 158. La Nauze JM. Aconitase and isocitric dehydrogenase od Aspergillus niger in relation to citric acid production. J Gen Microbiol 1966;44:73–81. 159. Mattey M. Citrate regulation of citric acid production in Aspergillus niger. FEMS Microbiol Lett 1977;2:71. 160. Kubicek CP and Ro¨hr M. The influence of manganese on enzyme synthesis and citric acid accumulation in Aspergillus niger. Eur J Appl Microbiol 1977;4:167–175. 161. Habison A, Kubicek CP and Ro¨hr M. Phosphofructokinase as a regulatory enzyme in citric acid producing Aspergillus niger. Biochem J 1983;209:669–676. 162. Legisˇ a M, Doctoral Thesis, University of Ljubljana, Ljubljana, 1987. 163. Leopold H and Valtr Z. Zur Wirkung des Kaliumferrocyanids bei der Herstellung von Melasselo¨sungen fu¨r die Citronensaürega¨rung. Nahrung 1964;8(1):37–48. 164. Noguchi Y and Johnson MJ. Citric acid fermentation on sugars purified with chelatin resin. J Bacteriol 1961;82:538–541. 165. Bowes I and Mattey M. The effect of manganese and magnesium ions on mitochondrial NADP+-dependent isocitrate dehydrogenase from Aspergillus niger. FEMS Microbiol Lett 1979;6(4):219–222. 166. Schweiger LB, U.S. Patent 2,970084, 1961. 167. Ahmed SA, Smith JE and Anderson JG. Mitochondrial activity during citric acid production by Aspergillus niger. Trans Br Mycol Soc 1972;59(Pt. 1):51–61. 168. Wold WS and Suzuki I. Regulation by zinc and adenosine 30 ,50 -cyclic monophosphate of growth and citric acid accumulation in Aspergillus niger. Canad J Microbiol 1976;22(8):1093–1101. 169. Weinberg ED. Biosynthesis of secondary metabolites: roles of trace metals. Adv Microbiol Physiol 1970;4:1–41. 170. Perman D, Kita DA and Peterson WH. Arch Biochem 1946;11(1):123. 171. Schlegel HA, ‘‘Allgemeine microbiologie’’ Geor Thieme, Vol. 273, Stuttgart, Verlag, 1974. 172. Adiga PR, Sastri KS, Vankatasubramanyan V and Sarma PS. Biochem J 1947;81:545–557. 173. Trentjeva OF, Trudy Leningradskogo Nauchno-Issledovatelskogo Instituta Phischevoy Promyshlenosti 1971;1:52. 174. Lesˇ niak W, Symposium Intensification of Citric Acid Production in Poland, Lodz, 1974. 175. Zhuravsky GI, Trentova OF, Aglish IB and Fisakova ES, Trudy Leningradskogo Nauchne : Issledovatelskoge Instituta Pishchevoy Promyshlennosti 27 (1971). 176. Roukas T and Harvey L. The effect of pH on production of citric and gluconic acid from beet molasses using continuous culture. Biotechnol Lett 1988;10(4): 289–294. 177. Honecker S, Bisping B, Zhu Y and Rehm HJ. Influence of sucrose concentration and phosphate limitation on citric acid production by immobilized cells of Aspergillus niger. Appl Microbiol Biotechnol 1989;31(1):17–24.

342 178. Tsay SS and To KY. Citric acid production using immobilized conidia of Aspergillus niger TMB 2022. Biotechnol Bioeng 1987;29:297–304. 179. Gary K and Sharma CB. Continuous production of citric acid by immobilized whole cells of Aspergillus niger. Journal of General and Applied Microbiology 1992;38:605–615. 180. Sanroman CA, Pintado J and Lema JM. A comparison of two techniques for the immobilisation of Aspergillus niger in polyurethane foam. Biotechnol Tech 1994;8:389–394. 181. Pallares J, Rodriguez S and Sanroman A. Citric acid production by immobilised Aspergillus niger in a fluidised bed reactor. Biotechnol Tech 1996;10:53–57. 182. Wang JL, Wen XH and Zhou D. Production of citric acid from molasses integrated with in-situ product separation by ion-exchange resin adsorption. Biores Technol 2000;75(3):231–234. 183. Roukas T. Citric acid production from carob pod by solid state fermentation. Enz Microbiol Technol 1999;24:54–59. 184. Lengon S, Brooks JD and Maddox IS. Influence of the glycolitic rate on production of citric acid and oxalic acid by Aspergillus niger in solid state fermentation. World J Microbiol Biotechnol 1999;15:493–495. 185. Grewal HS and Kalra KL. Fungal production of citric acid. Biotechnol Adv 1995;13:209–234. 186. Glusz P and Ledakowicz S. In: Citric Acid Biotechnology: Down Stream Processing in Citric Acid Production, Kristiansen B, Mattey M and Linden J (eds), London, Taylor Francis Ltd., 1999, pp. 135–148. 187. Ayers R, Jr., U.S. Patent 2,810,755. Baker, R W, Tuttle, M E, Kelly, 1957. 188. Kertes AS and King CJ. Extraction chemistry of fermentation product carboxylic acids. Biotechnol Bioeng 1986;28:269–282. 189. Hartl J and Mare J. Extraction processes for bioproduct separation. Separation Sci Technol 1993;28:805–819. 190. Schugerl K. Solvent Extraction in Biotechnology, Springer, 1994. 191. Kasprzycka Guttman T, Jarosz K, Semeniuk B, Myslinsky A, Wilcura H and Kurcinska H, Polish Patent 160 397, 1989. 192. Pagel HA and Schwab KD. Effect of temperature on tributyl phosphate as extracting agent for organic acids. Anal Chem 1950;22:1207–1212. 193. Grinstead RR, U.S. Patents 3 980 701–704, 1976 194. Bauer U, Marr R, Rueckl W and Siebenhofer M. Extraction of citric acid from aqueous solutions. Chem Biochem Eng 1988;2:230–232. 195. Prochazka J, Heyberger A, Bizek V, Kousova M and Volaufova E. Amine extraction of hydroxy-carboxylic acids. 2. Comparison of equilibria for lactic, malic and citric acids. Ind Eng Chem Res 1994;33:1565–1573. 196. Juang RS and Huang WT. Kinetics studies of extraction of citric acid from aqueous solution with tri n-octylamine. J Chem Eng Japan 1995;28:274–281. 197. Edlauer R, Kirkovits AE, Westermayer R and Stojan O, European Patent 377 430, 1990. 198. Baker RW, Tuttle ME, Kelly DJ and Lonsdale HK. Coupled-transport membranes. J Membr Sci 1977;2:213–221. 199. Basu R and Sirkar KK. Citric acid extraction with microporous hollow fibres. Solvent Extraction and Ion Exchange 1992;10:119–144. 200. Friesen DT, Babcock WC, Brose DJ and Chambers AR. Recovery of citric acid from fermentation beer using supported-liquid membranes. J Membr Sci 1991;56:127–141.

343 201. Juang RS, Recovery of citric acid from aqueous streams by supported liquid membranes containing various salts of tri-n-octylamine, presented at the AIChE Annual Meeting, Miami, paper 28f, 1995. 202. Basu R and Sirkar KK. Hollow fibre contained liquid membrane separation of citric acid. AIChE J 1991;37:383–393. 203. Strathmann H, Rapp HJ, Bauer B and Bell CH. Theoretical and practical aspects of preparing bipolar membranes. Desalination 1993;90(1–3):303–323. 204. Sappino F, Mancini M and Moresi M. Recovery of sodium citrate from aqueous solutions by electrodialysis. Italian J Food Sci 1996;8:239. 205. Novalic S and Kulbe KD, Separation and concentration of citric acid by means of electrodialytic bipolar membrane technology, Proceedings of International Conference Advances in Citric Acid Technology, Bratislava, October, 1996, pp. 41–44. 206. Novalic S, Jagschits F, Okwor J and Kulbe KD. Behaviour of citric acid during electrodialysis. J Membr Sci 1995;108:201–205. 207. Moresi M and Sappino F, Effect of temperature and pH on sodium citrate recovery from aqueous solutions by electrodialysis, Proceedings of International Conference Advances in Citric Acid Technology, Bratislava, October, 1996, p. 29. 208. Novalic S, Okwor J and Kulbe KD. The characteristics of citric acid separation using electrodialysis with bipolar membranes. Desalination 1996;105:277–282. 209. Bohdziewicz J and Bodzek M. Ultrafiltration preparation of pectinolytic enzymes from citric acid fermentation broth. Process Biochem 1994;29:99–107. 210. Schierholt J. Fermentation processes of production of Citric-Acid. Process Biochem 1977;12(9):20–21.


345

Using nanotechnology to improve the characteristics of antineoplastic drugs: Improved characteristics of nab-paclitaxel compared with solvent-based paclitaxel MaryAnn Foote MA Foote Associates, 4284 Par Five Court, Westlake Village, CA 91362, USA Abstract. Nanotechnology refers to the use of very small pieces of matter, typically p200 nm in diameter. Nanoparticle albumin-bound (nab) paclitaxel, a soluble form of the cytotoxin paclitaxel that has demonstrated utility in the setting of cancer chemotherapy, is produced by nab technology using the protein albumin. nab-Paclitaxel targets tumors, enhances tumor penetration by the novel mechanism of albumin receptor-mediated (gp60) endothelial transcytosis, and avoids the use of surfactants and solvents such as Cremophor and Tween. nab-Paclitaxel minimizes the toxicities associated with Cremophor and eliminates the need for premedication for hypersensitivity reactions caused by Cremophor. The albumin coating that surrounds the active drug assists in the transport of the nanoparticles to the interior of the tumor cell that preferentially takes in albumin as a nutrient through the gp60 pathway. In nonclinical studies, nab-paclitaxel achieved higher intratumoral concentrations compared with solvent-based paclitaxel and increased the bioavailability of paclitaxel by eliminating the entrapment of paclitaxel in the plasma. Compared with solvent-based paclitaxel, at equitoxic doses, the nab-paclitaxel produced more complete regressions, longer time to recurrence, longer doubling times, and prolonged survival. nab-Paclitaxel has been shown to have superior efficacy compared with solvent-based paclitaxel without the need for premedication in clinical trials of patients with advanced solid tumors. nab-Paclitaxel has been effective in patients for whom previous chemotherapy has not been helpful. nab Technology has the potential to be applied to other insoluble drugs. Keywords: albumin, cancer, chemotherapy, gp60, solvent

Introduction Traditionally, drugs have been delivered by topical, oral, intravenous, or intramuscular routes, and new delivery systems have added nasal, intrathecal, and respiratory routes. When developing a new therapeutic, several issues of drug delivery must be resolved, including patient compliance. The issue of patient compliance is important: large difficult-to-swallow pills, multiple injections, and long intravenous infusion sessions may discourage patients from starting, continuing, or maintaining a therapy. Inadequate delivery of chemotherapeutic drugs, for example, can limit the therapeutic response [1]. Another issue that may effect patient compliance is the nature and severity of adverse events from the therapy and the need for premedication or observation after medication. The taxanes, paclitaxel, and docetaxel, are agents with significant antitumor activity that are used alone or with other agents in Corresponding author. Tel: 805-370-6360.

E-mail: [email protected] (M. Foote). BIOTECHNOLOGY ANNUAL REVIEW VOLUME 13 ISSN 1387-2656 DOI: 10.1016/S1387-2656(07)13012-X


346 the treatment of patients with advanced metastatic breast cancer [2]. The taxanes, however, are very hydrophobic and must be formulated in solvents that make them more hydrophilic to increase their solubility and bioavailability. Traditionally, paclitaxel is formulated in a castor oil-derived solvent (Cremophor EL) (CrEL) and docetaxel is formulated in polysorbate-80 (Tween 80) (P-80). Both solvents, CrEL and P-80, are biologically and pharmacologically active [3]. CrEL can leach plasticizers from polyvinyl chloride bags and polyethylene tubing sets that can cause some hepatic toxicity [4]. Both CrEL and P-80 can form micelles in aqueous solutions, which may reduce the availability of paclitaxel to tumors [3]. Because of these solvent-mediated problems, patients who receive solventbased paclitaxel must be premedicated with steroids and antihistamines to reduce the risk of hypersensitivity reactions; administration of the drug requires special diethylhexylphthalate-free infusion sets and in-line filters. Nanoparticle albumin-bound (nab) technology uses the protein albumin to produce a therapeutic product that targets tumors, enhances tumor penetration by the novel mechanism of albumin receptor-mediated (gp60) endothelial transcytosis, and avoids the use of surfactants and solvents such as CrEL and P-80. nab-Paclitaxel was developed to minimize the toxicities associated with CrEL and to eliminate the need for premedication for hypersensitivity reactions caused by CrEL. Removing the need for premedication, special intravenous bags, and tubing can increase the ease with which the product can be administered and may reduce the cost associated with administration. Most importantly, elimination of CrEL-associated toxicities should increase patient compliance with the chemotherapy regimen.

Paclitaxel Paclitaxel, an antineoplastic agent extracted from various species of the yew tree (Taxus spp.), is an antimicrotubule agent that promotes the assembly of microtubules from tubulin dimers and stabilizes microtubules by preventing depolymerization (Fig. 1) [5]. Stabilization of the microtubular network compromises interphase and mitosis by inducing abnormal arrays of microtubules and multiple asters of microtubules [6]. Paclitaxel (solvent-based) has remarkable activity against metastatic breast cancer and is approved for multiple indications, including the treatment of breast cancer in patients for whom a previous regimen of doxorubicincontaining combination chemotherapy has failed to cure their disease [7]. Solvent-based paclitaxel also is indicated as first-line and subsequent therapy for the treatment of patients with advanced ovarian carcinoma. The combination of paclitaxel and cisplatin is indicated as first-line treatment of both ovarian cancer and nonsmall-cell lung cancer (NSCLC) in patients who are not candidates for potentially curative surgery or radiation therapy, or both [7].

347

AcO O

O

O

OH O

N H

OH HO BzO AcO

O

Fig. 1. Structural formula of paclitaxel. Courtesy of Abraxis BioScience.

Chemical name: 5b,20-epoxy-1,2a,4,7b,10b,13a-hexahydroxytax-11-en-9-one 4,10diacetate 2-benzoate 13-ester with (2R,3S)-N-benzoyl-3-phenylisoserine.

Because paclitaxel is water insoluble, CrEL is needed to solubilize the drug, and CrEL has important limitations and disadvantages. CrEL leaches plasticizers and causes severe toxicities, including hypersensitivity reactions (some fatal), and prolonged, sometimes irreversible sensory neuropathy associated with demyelination and axonal degeneration [8–10]. In addition, CrEL has been shown to adversely affect drug efficacy because it entraps the active drug in micelles in the plasma compartment [11,12]. This entrapment that affects not only the taxanes but also co-administered drugs (e.g., anthracyclines), leads to increased systemic drug exposure, decreased drug clearance, nonlinear pharmacokinetics, and lack of dose-dependent antitumor activity [11–14]. In nonclinical studies, CrEL produced axonal swelling, vesicular degeneration, and demyelinization [15,16]. In a rat model, weekly injections of CrEL and ethanol produced a delayed sensory neuropathy, which had not improved by the end of the experiment [17]. It is thought that the neurotoxic effects of CrEL may exacerbate taxane-induced neuropathy. nab-Paclitaxel nab nanotechnology allows the delivery and tumor targeting of insoluble drugs in the form of nanoparticles typically o200 nm in diameter (Fig. 2). nab-Paclitaxel, albumin-bound nanoparticle paclitaxel suspension, is the first product to combine biologically interactive human albumin with an active pharmaceutical agent (paclitaxel) in the nanoparticle state. Albumin is a natural carrier of water-insoluble molecules in the human circulatory system. Albumin complexes are carried from the systemic circulation into tissues through an albumin receptor (gp60; albondin)-mediated pathway [18,19]. By developing an albuminbased nanoparticle that can carry a chemotherapeutic agent, such as paclitaxel, it is possible to use the biologic pathway and the gp60 receptor to provide higher concentrations of the drug in the tumor. nab-Paclitaxel has the same mechanism of action as paclitaxel, i.e., inhibition of the microtubular networks needed for mitosis [20–22]. nab Technology allows administration of a higher dose of paclitaxel (almost 2 times greater) compared with solvent-based paclitaxel [23].

348

Drug Drug Size = 50-200 nm

Albumin

Fig. 2. nab Technology. The figure shows the active drug surrounded by albumin.

Courtesy of Abraxis BioScience.

Nonclinical studies with nab-paclitaxel CrEL, the solvent used in the formulation of solvent-based paclitaxel, has been shown to inhibit the binding of paclitaxel, the active drug, to endothelial cells and to albumin. In nonclinical studies in nude mice with human tumor xenoplants, nab-paclitaxel achieved a 33% higher intratumoral concentration compared with solvent-based paclitaxel [24]. nab-Paclitaxel increases the bioavailability of paclitaxel by eliminating the entrapment of paclitaxel in the plasma compartment caused by micelles, resulting in enhanced tumor penetration and higher intratumor concentrations. Compared with animals treated with solvent-based paclitaxel, at equitoxic doses, the nab-paclitaxeltreated mice had more complete regressions, longer time to recurrence, longer doubling times, and prolonged survival. Clinical studies with nab-paclitaxel Mechanism of action studies in the oncology setting The antineoplastic mechanism of action of paclitaxel is well characterized [5,6]. A transcellular pathway for the transport of albumin across the endothelial cell wall to the underlying interstitium may account for the higher

349 antitumor activity and clinical efficacy noted with nab-paclitaxel compared with solvent-based paclitaxel in patients with breast cancer [19,25,26]. Binding of albumin to this receptor on endothelial cells activates caveolin-1, which leads to the formation of vesicles (caveolae) that transport the albumin–drug complex and other fluid-phase components across endothelial cells and into the tumor interstitium [18,19,27,28]. Pharmacokinetics CrEL micelles form in the plasma compartment and entrap paclitaxel, resulting in nonlinear pharmacokinetics [11,29]. Several studies have suggested significant pharmacokinetic interactions between CrEL (with or without paclitaxel) and doxorubicin [30–33]. Safety – hypersensitivity reactions The most extensively described biologic effect attributed to CrEL is an acute hypersensitivity reaction characterized by dyspnea, flushing, rash, chest pain, tachycardia, hypotension, angioedema, and generated urticaria [9,10,34,35]. Despite premedication with corticosteroids and histamine H1 and H2 blockers, minor reactions to CrEL-based paclitaxel (e.g., flushing and rash) still occur in approximately 40% of treated patients [36,37]; major potentially life-threatening hypersensitivity reactions are observed in 1.5–3% of treated patients [9,38,39]. In clinical trials, nab-paclitaxel-treated patients rarely were premedicated with corticosteroids and antihistamines, but medication was necessary with solvent-based paclitaxel [23,40–44]. Because nab-paclitaxel was engineered as albumin-bound particles that are water soluble, no toxic solvent is required. Hence, hypersensitivity reactions common to solvent-based paclitaxel would not be expected in patients treated with nab-paclitaxel. Safety – neuropathy In the clinical setting, sensory and motor neuropathy are known adverse events attributable to solvent-based formulations of taxanes [10,45,46]. In a study with solvent-based paclitaxel given at 250 mg/m2 (standard dose is 175 mg/m2), grade 3 sensory neuropathy was reported at 32% [14]. Because nab-paclitaxel is not formulated with solvents, clinical studies have often used a higher dose of nab-paclitaxel (260 mg/m2) compared with solvent-based paclitaxel (175 mg/m2). Peripheral neuropathy would be expected in nab-paclitaxel-treated patients, as this event appears to be a class effect [47]. Patients who were treated with nab-paclitaxel did not appear to sustain axonal degeneration and demyelination, recognized by prolonged and often irreversible sensory neuropathy, common in patients treated with solvent-based paclitaxel [2,48].

350 General overview of studies The results from the nonclinical studies allowed nab-paclitaxel to move into the clinical stages of development. Several key clinical studies have been published (Table 1). Phase 1 studies Three phase 1 studies established the preliminary safety risks, antitumor effects, and pharmacokinetics of nab-paclitaxel [40,44,49]. All patients enrolled in these studies were adults aged X18 years with advanced malignancy or solid tumors, refractory to standard therapy (including previous taxane therapy). nab-Paclitaxel 80–375 mg/m2 was administered as a 30-min intravenous infusion. The first patients studied did receive premedication for hypersensitivity reactions, but as data were collected, the premedication was eliminated without adverse effects. No hypersensitivity reactions were reported. The dose-limiting toxicities were grade 4 neutropenia and grade 3 sensory neuropathy. Other adverse events were of the type expected in this patient population treated with a taxane: fatigue, nausea/vomiting, mild anemia, and alopecia. Complete and partial responses were seen, including a partial response 412 months in a woman with ovarian cancer. The pharmacokinetic profile of nab-paclitaxel emerged over the dose range of 80–150 mg/m2; the range of mean values for plasma clearance was 22–31 L/h/m2, the means for plasma half-life were 15–18 h, and the means for apparent volume of distribution were 549–772 L/m2. Pharmacokinetic analyses of nab-paclitaxel exhibited linear increases for Cmax and area under the curve (AUC). nab-Paclitaxel exhibited multiphasic pharmacokinetics with a rapid distribution phase, a rapid elimination phase having a T1/2 414 h, a large volume of distribution (i.e., 4 500 L/m2), and a linear relationship between dose and AUC. In all phase 1 studies, nab-paclitaxel was well tolerated when administered without pretreatment or recombinant human granulocyte colony-stimulating factor (rHuG-CSF) support and nab-paclitaxel demonstrated significant antitumor activity, even in patients previously treated with taxanes. The results from the phase 1 studies suggested that nab-paclitaxel could be safely administered to patients with cancer without the need for premedication, and that nab-paclitaxel had antitumor activity. Phase 2 studies Building on the results of phase 1 studies, nab-paclitaxel proceeded to phase 2 studies, 3 of which have been published [41–43]. In these three phase 2

351 Table 1. Key published clinical studies. References

Indication

Study phase

Key finding

[23]

Breast cancer

3

[43]

Nonsmall-cell lung cancer

2

[42]

Breast cancer

2

[41]

Tongue cancer

2

[44]

Advanced nonhematologic malignancies

1

Showed greater efficacy of nabpaclitaxel compared with solvent-based paclitaxel First report of efficacy of nabpaclitaxel in the treatment of patients with NSCLC. No premedications required and both significant tumor response and prolonged disease control were documented in patients treated with 260 mg/m2 nabpaclitaxel Results suggested that nabpaclitaxel was effective in women with metastatic breast cancer as both first-line and greater-than-first-line treatment. High doses of nabpaclitaxel could be administered without premedication for hypersensitivity reactions to Cremophor found in solventbased paclitaxel Report of the intraarterial administration of nabpaclitaxel as induction therapy to treat patients with advanced cancer of the tongue. Results documented a 78% response rate (complete, 26%; partial, 52%), with an additional 13% of patients experiencing stable disease Results showed that nabpaclitaxel could be administered at higher doses than solvent-based paclitaxel using a weekly administration schedule. Pharmacokinetics of nab-paclitaxel was linear over the doses studied and antitumor responses were documented in patients who

352 Table 1 (Continued ) References

Indication

Study phase

[40]

Advanced solid tumors

1

[49]

Head-and-neck cancer or anal cancer

1

Key finding had been treated previously with solvent-based paclitaxel Results of a phase 1 trial showed that nab-paclitaxel can be infused more quickly than solvent-based paclitaxel without the need for premedication for hypersensitivity reactions Intraarterial administration of nab-paclitaxel treated patients with advanced cancer. Most dose levels showed considerable antitumor activity with documented complete or partial response in 80% of evaluable patients

studies, nab-paclitaxel was administered at 260 and 300 mg/m2 for the treatment of metastatic breast cancer, NSCLC, and tongue cancer. The results of these studies suggested that nab-paclitaxel was safe and had significant antitumor effect. The most frequently reported adverse events were toxicities expected with treatment with a taxane in this patient population, and treatment-related toxicities reported included alopecia, anemia, neutropenia, leukopenia, sensory neuropathy, fatigue, nausea, myalgia, infection, vomiting, and stomatitis/pharyngitis. Antitumor activity was noted: in the Green et al. [43] study in patients with NSCLC, a disease with generally poor prognosis, the overall response rate was 16.3% (95% CI: 5.24–27.31) (all responses were partial). The median time to progression was 6 months (95% CI: 3.9–6.5) and median survival was 11 months (95% CI: 9.5–16.2). The probability of not having progressed was 50% at 6 months and 13% at 1 year, with the probability of surviving 1 year 45%. Phase 3 study The pivotal phase 3 study was done by Gradishar et al. [23]. This randomized phase 3 trial evaluated the antitumor activity and the safety of nab-paclitaxel given intravenously at 260 mg/m2 compared with solvent-based paclitaxel given at 175 mg/m2 for the treatment of metastatic breast cancer. Women aged X18 years who had measurable metastatic breast cancer were assigned randomly to receive nab-paclitaxel 260 mg/m2 every 3 weeks without

353 premedication, or solvent-based paclitaxel 175 mg/m2 every 3 weeks with standard steroid and antihistamine premedication. Of the 454 patients who received at least 1 dose of the drug, a significantly higher overall response rate was noted in patients who received nab-paclitaxel compared with patients who received solvent-based paclitaxel. A significantly higher tumor response was noted in patients who received chemotherapy for the first time, patients who received second- or greater-line therapy, and patients who had received previous anthracycline therapy in either the adjuvant/metastatic setting or the metastatic setting only. A longer time to tumor progression was noted in patients who received nab-paclitaxel compared with patients who received solvent-based paclitaxel (median of 23.0 weeks vs. 16.9 weeks, respectively). The median survival for those patients who received nab-paclitaxel was 10 weeks longer (65.0 weeks vs. 55.7 weeks) than for patients who received solvent-based paclitaxel, but this difference was not statistically significant; however, there was a significant increase in survival in patients who received nab-paclitaxel in the second-line or greater setting compared with patients who received solvent-based paclitaxel (56.4 weeks vs. 46.7 weeks). Within the nab-paclitaxel group, 98% of treatment cycles were administered without steroid premedication, whereas in 95% of treatment cycles patients receiving solvent-based paclitaxel received premedication with steroids and antihistamines. Although the patients in the nab-paclitaxel group received an average paclitaxel dose intensity 49% greater than that received by patients in the solvent-based paclitaxel group, treatment compliance was high, with 96% and 94% of patients in the nab-paclitaxel group and solvent-based paclitaxel group receiving X90% of the protocol-specified dose. Consistent with this higher dose of paclitaxel, the incidence of grade 3 sensory neuropathy was higher in the nab-paclitaxel cohort compared with the solvent-based cohort (10% vs. 2%, respectively). The grade 3 sensory neuropathy resolved more rapidly in the nab-paclitaxel group (median: 22 days) and was easily managed with treatment interruptions and dose reductions. Neither grade 4 sensory neuropathy nor severe motor neuropathy was reported for either treatment group. Despite the higher dose of solvent-based paclitaxel delivered, there was significantly less grade 4 neutropenia with nab-paclitaxel compared with solvent-based paclitaxel (9% vs. 22%), providing evidence that CrEL may contribute such toxicity as bone marrow damage and loss of white blood cells. nab-Paclitaxel (ABRAXANE) received marketing approval from the United States Food and Drug Administration (FDA) in January 2005 for the treatment of patients with breast cancer after failure of combination chemotherapy for metastatic disease or relapse within 6 months of adjuvant chemotherapy and from the Therapeutic Products Directorate of Health Canada in June 2006 for the treatment of patients with metastatic breast cancer, including use as first-line treatment.

354 Discussion nab-Paclitaxel uses nanotechnology to bind paclitaxel to human albumin, in the same concentrations as found in human blood. Albumin is a natural solubilizer and eliminates the need for traditional chemical solvents for paclitaxel. The particles formed by the proprietary nanotechnology are smaller than a red blood cell. Without solvents, tumors may preferentially capture the albumin-bound drug using gp60 receptors, which actively transport paclitaxel across cell membranes. The result is more drug is delivered to the tumor. Elimination of the solvents also eliminates the need for premedication and allows nab-paclitaxel to be delivered more rapidly than a course of solvent-based paclitaxel, enhancing patient compliance. nab-Paclitaxel takes advantage of a biologic pathway, using the natural tumor biology (increased blood supply and nutrient extraction for rapid growth) to deliver chemotherapeutic agents to malignant cells. This ability to target the vascular endothelium and to increase intratumor concentration of the drug by taking advantage of the amplification effect imparted by angiogenesis, while minimizing exposure to normal tissue, has been associated with increased efficacy of nab-paclitaxel in the treatment of patients with metastatic breast cancer. nab Technology has the potential to improve the characteristics of other antineoplastic drugs, particularly ones that require the use of toxic solvents to permit their administration.

References 1. 2. 3. 4.

5. 6. 7. 8.

Norton L. Evolving concepts in the systemic drug therapy of breast cancer. Semin Oncol 1997;24:S3–S10. Rowinsky EK, Chaudhry V, Cornblath DR and Donehower RC. Neurotoxicity of Taxol. J Natl Cancer Inst Monogr 1993;15:107–115. van Zuylen I, Verweij J and Sparreboom A. Role of formulation vehicles in taxane pharmacology. Invest New Drugs 2001;19:125–141. Goldspeil BR. Guidelines for administration. In: Paclitaxel in Cancer Treatment, McGuire WP and Rowinsky EK (eds), New York, NY, Marcel Dekker Inc, 1995, pp. 175–186. Schiff PB and Horwitz SB. Taxol stabilizes microtubules in mouse fibroblast cells. Proc Natl Acad Sci USA 1980;77:1561–1565. Schiff PB, Fant J and Horowitz SB. Promotion of microtubule assembly in vitro by taxol. Nature 1979;227:665–667. Taxol (paclitaxel) Injection [prescribing information]. Available at www.bms.com Accessed 21 March 2006. Lorenz W, Reimann HJ, Schmal A, et al. Histamine release in dogs by Cremophor EL and its derivatives: oxethylated oleic acid is the most effective constituent. Agents Actions 1977;7:63–67.

355 9. 10. 11.

12. 13.

14.

15.

16. 17. 18.

19.

20.

21. 22.

23.

24.

25.

26.

Weiss RB, Donehower RC, Wiernik PH, et al. Hypersensitivity reactions from taxol. J Clin Oncol 1990;8:1263–1268. Gelderblom H, Verweij J, Nooter K and Sparreboom A. Cremophor EL: the drawbacks and advantages of vehicle selection for drug formulation. Eur J Cancer 2001;37:1590–1598. Sparreboom A, van Zuylen L, Brouwer E, et al. Cremophor EL-mediated alteration of paclitaxel distribution in human blood: clinical pharmacokinetic implications. Cancer Res 1999;59:1454–1457. ten Tije AJ, Verweij J, Loos WJ, et al. Pharmacological effects of formulation vehicles: implications for cancer chemotherapy. Clin Pharmacokinet 2003;42:665–685. van Tellingen O, Huizing MT, Nannan Panday VR, et al. Cremophor EL causes (pseudo–) non-linear pharmacokinetics of paclitaxel in patients. Br J Cancer 1999;81:330–335. Winer EP, Berry DA, Woolf S, et al. Failure of higher-dose paclitaxel to improve outcome in patients with metastatic breast cancer: cancer and leukemia group B trial 9342. J Clin Oncol 2004;22:2061–2068. Boer HH, Moorer-van Delft CM, Muller LJ, et al. Ultrastructural neuropathologic effects of taxol on neurons of the freshwater snail Lymnaea stagnalis. J Neurooncol 1995;25:49–57. Windebank AJ. The vehicle for cyclosporine is neurotoxic in vitro. Ann Neurol 1997;41:563–564. Authier N, Gillet J-P, Fialip J, Eschalier A and Coudore F. Description of a short-term taxol-induced nociceptive neuropathy in rats. Brain Res 2000;887:239–249. Schubert W, Frank PG, Razani B, Park DS, Chow C-W and Lisanti MP. Caveolaedeficient endothelial cells show defects in the uptake and transport of albumin in vivo. J Biol Chem 2001;276:48619–48622. John TA, Vogel SM, Tiruppathi C, Malik AB and Minshall RD. Quantitative analysis of albumin uptake and transport in the rat microvessel endothelial monolayer. Am J Physiol Lung Cell Mol Physiol 2003;284:L187–L196. Riondel J, Jacrot M, Picot F, Beriel H, Mouriquand C and Potier P. Therapeutic response to taxol of six human tumors xenografted into nude mice. Cancer Chemother Pharmacol 1986;17:137–142. Rowinsky EK, Cazenave LA and Donehower RC. Taxol: a novel investigational antimicrotubule agent. J Natl Cancer Inst 1990;82:1247–1259. Adams JD, Flora KP, Goldspiel BR, Wilson JW, Arbuck SG and Finley R. Taxol: a history of pharmaceutical development and current pharmaceutical concerns. J Natl Cancer Inst Monogr 1993;15:141–147. Gradishar WJ, Tjulandin S, Davidson N, et al. Phase III trial of nanoparticle albuminbound paclitaxel compared with polyethylated castor oil-based paclitaxel in women with breast cancer. J Clin Oncol 2005;23:7768–7771. Desai N, Trieu V, Yao Z, et al. Increased antitumor activity, intratumor paclitaxel concentrations, and endothelial cell transport of Cremophor-free, albumin-bound paclitaxel, ABI-007, compared with Cremophor-based paclitaxel. Clin Cancer Res 2006;12:1317–1324. John TA, Vogel SM, Mishall RD, Ridge K, Tiruppathi C and Malik AB. Evidence for the role of alveolar epithelial gp60 in active transalveolar albumin transport in the rat lung. J Physiol 2001;533:547–549. Vogel SM, Minshall RD, Pilipovic´ M, Tiruppathi C and Malik AB. Albumin uptake and transcytosis in endothelial cells in vivo induced by albumin-biding protein. Am J Physiol Lung cell Mol Physiol 2001;281:L1512–L1522.

356 27. Simionescu M, Gafencu A and Antohe F. Transcytosis of plasma macromolecules in endothelial cells: a cell biological survey. Microsc Res Tech 2002;57:269–288. 28. Tahir SA, Ren C, Timme TL, et al. Development of an immunoassay for serum caveolin1: a novel biomarker for prostate cancer. Clin Cancer Res 2003;9:3653–3659. 29. Gianni L, Kearns CM, Giani A, et al. Nonlinear pharmacokinetics and metabolism of paclitaxel and its pharmacokinetic/pharmacodynamic relationships in humans. J Clin Oncol 1995;13:180–190. 30. Holmes FA, Madden T, Newman RA, et al. Sequence-dependent alteration of doxorubicin pharmacokinetics by paclitaxel in a phase I study of paclitaxel and doxorubicin in patients with metastatic breast cancer. J Clin Oncol 1996;14:2713–2721. 31. Webster LK, Cosson EJ, Stokes KH, et al. Effect of the paclitaxel vehicle, Cremophor EL, on the pharmacokinetics of doxorubicin and doxorubicinol in mice. Br J Cancer 1996;73:522–524. 32. Gianni L, Vigano L, Locatelli A, et al. Human pharmacokinetic characterization and in vitro study of the interaction between doxorubicin and paclitaxel in patients with breast cancer. J Clin Oncol 1997;15:1906–1915. 33. Millward MJ, Webster LK, Rischin D, et al. Phase I trial of Cremophor EL with bolus doxorubicin. Clin Cancer Res 1998;4:2321–2329. 34. Dye D and Watkins J. Suspected anaphylactic reaction to Cremophor EL. Br Med J 1980;280:1353. 35. Hayes FA, Abromowitch M and Green AA. Allergic reactions to teniposide in patients with neuroblastoma and lymphoid malignancies. Cancer Treat Rep 1985;69:439–441. 36. Michaud LB. Methods for preventing reactions secondary to Cremophor-EL. Ann Phamacother 1997;31:1402–1404. 37. Price KS and Castells MC. Taxol reactions. Allergy Asthma Proc 2002;23:205–208. 38. Essayan DM, Kagey-Sobotka A, Colarusso PJ, et al. Successful parenteral desensitization to paclitaxel. J Allergy Clin Immunol 1996;97:42–46. 39. Kloover JS, den Bakker MA, Gelderblom H, et al. Fatal outcome of a hypersensitivity reaction to paclitaxel: a critical review of premedication regimens. Br J Cancer 2004;90:304–305. 40. Ibrahim NK, Desai N, Legha S, et al. Phase I and pharmacokinetic study of ABI 007, a Cremophor-free, protein-stabilized, nanoparticle formulation of paclitaxel. Clin Cancer Res 2002;8:1038–1044. 41. Damascelli B, Patelli GL, Lanocita R, et al. A novel intraarterial chemotherapy using paclitaxel in albumin nanoparticles to treat advanced squamous cell carcinoma of the tongue: preliminary findings. Am J Roentgenol 2003;181:253–260. 42. Ibrahim NK, Samuels B, Page R, et al. Multicenter phase II trial of ABI-007, an albumin-bound paclitaxel, in women with metastatic breast cancer. J Clin Oncol 2005;23:6019–6026. 43. Green MR, Manikhas GM, Orlov S, et al. Abraxane, a novel Cremophor-free, albuminbound particle form of paclitaxel for the treatment of advanced non-small-cell lung cancer. Ann Oncol 2006;17:1263–1268. 44. Nyman DW, Campbell KJ, Hersh E, et al. Phase I and pharmacokinetics trial of ABI-007, a novel nanoparticle formulation of paclitaxel in patients with advanced nonhematologic malignancies. J Clin Oncol 2005;23:7785–7793. 45. Postma TJ, Vermorken JB, Liefting AJM, et al. Paclitaxel-induced neuropathy. Ann Oncol 1995;6:489–494.

357 46. Tabernero J, Climent MA, Lluch A, et al. A multicentre, randomized phase II study of weekly or 3-weekly docetaxel in patients with metastatic breast cancer. Ann Oncol 2004;15:1358–1365. 47. Gligorov J and Lotz JP. Preclinical pharmacology of the taxanes: implications of the differences. The Oncol 2004;9:3–8. 48. Sahenk Z, Barohn R, New P, et al. Taxol neuropathy. Arch Neurol 1994;51:72–729. 49. Damascelli B, Cantu G, Mattavlli F, et al. Intraarterial chemotherapy with polyoxyethylated castor oil free paclitaxel incorporated in albumin nanoparticles (ABI-007): phase II study of patients with squamous cell carcinoma of the head and neck and anal canal – preliminary evidence of clinical activity. Cancer 2001;92:25-92-2602.


359

Index of authors

Ather, A. 223 Berovic, M. 265, 303 Boh, B. 265 Droege, M. 1 Faye, L. 115 Foote, M.A. 345 Gomord, V. 115 Hashida, Y. 43 Hatfield, G.W. 27 Houdebine, L.-M. 65 Inouye, K. 43 Khan, M.T.H. 223 Kusano, M. 43 Laurell, T. 149

Legisa, M. 303 Lieńard, D. 115 Marko-Varga, G. 149 Minoda, M. 43 Morrow, K.J. Jr. 95 Ressine, A. 149 Roth, D.A. 27 Soler, E. 65 Sourrouille, C. 115 VoX, C. 201 Yasukawa, K. 43 Zhang, J. 265 Zhi-Bin, L. 265 Ziebolz, B. 1


361

Keyword index

Acacia mellifera 228 Adjuvants of throat 65, 67, 77–82, 84–85 affinity chromatography 43, 54–55, 57, 59–60, 101, 210, 274 affinity ligands 95, 111, 211, 213, 218 affinity separation 1, 201, 210 AIDS 17, 110, 204, 223, 225, 234, 245, 255–256 albumin 185, 345–349, 354 alkaloids 227, 232–234, 267, 289 anti-cancer effects 265 anti-HIV 223, 225–238, 240–241, 244–246, 252, 254–256, 282 anti-HIV drugs 223, 225–226, 231, 252, 255–256 anti-HIV-1 223, 226–227, 230, 238, 241, 244, 246, 282 antibody microarrays 149, 151, 157 Bacillus stearothermophilus 43 Bacillus subtilis 43 Bacillus thermoproteolyticus 43–44 b-D-glucans 265, 274 Bergenin 229 bioassay guided-fractionation 228 Biochemistry 291, 303, 306 biomarker discovery 149 bioreactor design 95 Camellia japonica 229 cancer 2, 18–21, 36, 79–80, 96–97, 161, 201, 265, 275, 281, 284, 287–288, 290–291, 345–346, 349–354 capillary gel electrophoresis 1, 201, 207, 214, 218 carbohydrates 69, 71, 227, 315 catechin 223–224, 229, 238–239

cell lysis 1, 201 chalcone 223–224, 247, 251–252 chemogenomics 223, 252 chemotherapy 19, 245, 275, 291, 345–346, 353 chromatography 1, 43, 54–55, 57, 59–60, 101, 108, 110, 201, 208–211, 213–214, 274, 289, 326 citric acid biosynthesis 303, 314, 316 codon context 27–28, 30–31 codon pair utilization bias 28 Combretaceae 229 Combretum molle 229 coumarin 223, 227, 233–234, 237, 238, 245, 247, 251–252 cultivation 1, 98, 100, 201–203, 265–276, 281, 286, 292, 314–316, 323, 328 digallic acid 229 disposable bioreactor components 95 disposable devices for clarification and purification 95 disposable protein purification devices 95 DNA molecule 8, 10, 203, 225 docking 223, 228, 230, 238, 242 dTTP 224, 228, 247 economic aspects 303, 329 engineering 27–28, 39, 43–44, 46, 60, 66, 98, 111, 115–116, 201, 204, 265, 303 enzyme 4, 14, 23, 43–46, 48–53, 57, 59, 77, 151–152, 183–184, 213–214, 218, 225, 227–228, 241–242, 245, 247, 281, 289, 306, 308–313 epitope carrier 65 Escherichia coli 27–28, 43, 81, 110, 201, 207

362 expression 2, 20, 27–28, 31–32, 35–40, 43, 46, 48–52, 60, 72, 75, 77, 95, 98–101, 106, 110–111, 115–117, 201–202, 216–217, 238, 240–241, 247, 252, 284–286, 288, 290

immunomodulation 265, 277, 283–284 in silico 223, 228, 230–231, 313 inclusion body 43 integrase 223–224, 241–245, 254 integration 73–74, 98, 182, 223, 225, 233–234, 241

Fabaceae 229 flavonoids 223, 227, 230, 233–236, 238, 240, 255 functional genomics 223 fusion inhibitor 224–226, 238 fusion proteins 28, 36, 67

kaempferol 229, 233, 243, 247, 249–250 Kenyan medicinal plant 228 Korean medicinal plants 229

gallate 223–224, 238, 240 gallic 229 gallotannin 229 Ganoderma lucidum 265–266, 268, 277 gene therapy 1, 100, 201, 205, 207, 216 genetic vaccination 1, 201 glucopyranose 229 glycosylation 68, 70–71, 76–77, 86, 95, 99, 102–104, 115–117, 233–234 glycyl-D-phenylalanine 43, 54 gp60 345–347, 354 gp120-CD4 binding 225 gp120-coreceptor 225 HAART 224, 226, 231, 255 high-speed biomarker identification 149 HIV 2, 17, 33–35, 79, 84, 201, 223–256, 267, 277, 282 HIV promoter 223, 247, 250–254 HIV RT inhibitors 227 HIV-1 33–35, 223–233, 235, 238–250, 254, 256, 282 HIV-1 life cycle 223, 225, 232–233, 240 HIV-1 treatment 225, 231 HIV-2 225 Hot Rod genes 27–28, 37, 39 human astrocytes 229 human immunodeficiency virus type 223–224 hydrophobic-interaction chromatography 43, 55, 57, 59–60

Leguminosae 228 lentivirus 225 lysate microarrays 149, 161 MALDI-TOF MS 149, 151, 180–181, 183, 185–186 manufacturing 1, 29, 35, 79, 81, 95, 107, 109–111, 158, 161, 189, 201, 207, 213, 218, 271, 275, 329 Maytenus buchanani 228 Maytenus senegalensis 228 medicinal plants 115, 227–230, 232 Melia azedarach 228 metalloproteinase 43–44, 47, 54–55 microbial strains 303 molecular farming 115 monoclonal antibodies 28, 86, 105 natural products’ databases 223 neolignans 227 neural network 223–224, 228 Nevirapine 226, 230, 246 NNRTIs 225 norlignans 227 NprM 43, 45–46, 48–51, 55 NRTIs 225 Peltophorum africanum 229 peptides 67, 69, 182, 185, 188, 227, 287, 328 pharmacological effects 223, 265, 276–277 phenolic compounds 223, 233–234, 237, 240

363 phenolics 223, 227, 232–237 phosphonoformic acid 228 Phyllanthus emblica 228 Planned Pause genes 28, 33 plant-made pharmaceutical 115, 122 plasmid 1, 10, 12, 51–52, 100, 201–205, 207–218, 247 polyphenol 71, 227, 238, 240, 255 polysaccharides 227, 232–234, 265, 267, 272–277, 283–290, 292 porous silicon 149, 152–158, 161–163, 165, 169–173, 176–180, 182–183, 185–187 product recovery 303, 318, 324 production 23, 27, 29, 32, 37, 39, 43, 45, 48–52, 59–60, 68, 70–72, 75–79, 81–82, 84–87, 95, 97–99, 101, 103–107, 109–111, 115–117, 161, 173, 201–204, 208, 210, 218, 238–240, 267–276, 284–287, 291–292, 303–308, 313–324, 326, 329–333 production processes 105, 202, 303, 318 propeptide 43, 49–52 proprietary cell lines 95 protease 43, 45–46, 52, 60, 81, 223–226, 233–234, 238–239, 245–247, 249, 254–256, 282, 312 protein databank 223–224, 241, 246 protein domains 28 protein engineering 43–44, 60 protein expression 27–28, 32, 35–38, 101, 111 protein microarray technology 149, 151–152, 155 protein subunit vaccines 28 protein therapeutics 28 protein-free and serum-free culture media 95 proteins 21, 23, 28, 30–32, 35–37, 39–40, 45, 49–51, 55, 57, 59, 65–73, 75–78, 80, 85–86, 95, 99–103, 107–111, 115–118, 151, 157, 166, 170, 180, 182, 185, 202–203, 207, 210, 212,

214, 224–227, 234, 238, 252–254, 267, 288, 324, 328 proteins A G and L 95, 110 protocatechuic acids 229 Prunus africana 228 Prunus sargentii 229 pseudotyped HIV-1 229 purification 12, 43, 52, 54–55, 57, 59–60, 68–69, 73, 76, 95, 98, 101, 107–111, 116, 201–203, 205, 207–211, 213, 218, 288–289 quercetin 229, 233, 237, 242–243 Quercus pedunculata 228 recombinant antibodies 95, 117 recombinant protein 43, 49, 51, 57, 65–71, 73, 75–78, 86, 98, 101, 107, 109, 115–117, 123, 126–128, 132–135, 202, 210 recombinant RNase 1, 213, 215, 218 Retrovir 226 reverse transcriptase 223–226, 242, 245, 254 reverse transcription 223, 225, 233–234, 245 Rhus natalensis 228 Ritonavir 226 RNase 1, 201, 207–208, 213–215, 218, 224, 229, 247 Rosa rugosa 229 Rumex cyprius 228 Sageretia theezans 229 Sargassum fusiforme 229 sawdust 265, 267–272 solid-state cultivation 265, 272 solvent 44, 166–168, 171, 184, 325–327, 329, 345–349, 351–354 Sophora flavescens 229 South African medicinal plants 229 speed plots 28, 33–35, 37–39 stability 1, 23, 44, 46, 48–49, 59–60, 68, 76, 83, 99, 117, 149, 180, 201–203, 211, 213–214, 216–218, 289, 305

364 submerged cultivation 265, 268, 272, 275–276, 281, 286, 323 substrates 59, 149, 156–157, 163, 165, 167, 170, 173, 176, 188, 265, 267–268, 270–272, 275, 303–304, 314, 317–319, 321, 323 superhydrophobic 149–152, 172–173, 176–177, 179, 185–186 superhydrophobic target anchor chips 149 synthetic biology 27–29 tenofovir 226 Terminalia bellerica 228 Terminalia chebula 228 Terminalia horrida 228 terpenoids 227 therapeutic protein 77, 115–118, 122–123, 126–129, 131–135, 202 thermolysin 43–57, 59–60 TLP-ste 43, 45–51, 54–55, 57, 59 transgenic animals 65, 67–68, 73–76, 86 transgenic plant 65, 67–68, 70–72, 86, 97, 115, 117–118, 128–129, 133

translation engineering 27–29, 39 translational kinetics 27–28, 31, 33 translational step-times 28, 31–33, 39 triterpenoids 265, 267, 276–277, 281–282, 292 ultrahydrophobic 149, 172, 180, 185–187 UNAIDS 225 vaccine 39, 65–67, 69–70, 72, 75, 77–84, 86–87, 201 Vernonia jugalis 228 viral transcription 225 virtual screening 223, 225, 228, 230 VLP 65–66, 82, 85–86 VSV/NL4-3 229 water-repellency 149, 173, 175, 177, 185 wood logs 265, 267–269 xanthones 227, 233–234 Zidovudine 223, 226

Biotechnology Annual Review, Vol 13

Biotechnology Annual Review, Volume 9 (Biotechnology Annual Review)

Biotechnology Annual Review, Volume 10 (Biotechnology Annual Review)

Biotechnology Annual Review, Volume 12 (Biotechnology Annual Review)

Biotechnology Annual Review, Volume 14

Biotechnology Annual Review, Volume 2

Biotechnology Annual Review, Volume 3

Biotechnology Annual Review, Volume 1

Annual Review of Psychology Vol 58 2007

Annual Review of Psychology, vol 52 2001

Annual Review of Psychology, vol 53 2002

Annual Review of Psychology, vol 55 2004

Annual Review of Psychology, Vol 56 2005

Annual Review of Psychology, vol 54 2003

Annual Review of Psychology, vol 50 1999

Annual Review of Psychology Vol 57, 2006

Annual Review of Psychology, vol 51 2000

Annual Review of Psychology vol 60, 2009

Annual Review of Genomics, 2003

Annual Review of Scalable Computing

Detail Annual (vol. 4)

Detail Annual (vol. 3)

Oceanography and Marine Biology: An Annual Review, Vol. 36

Annual review of nano research

Annual Review of Law and Social Science (2007 Vol 3)

Social Policy Review 13 (No.13)

Biotechnology, 2E, Vol. 12, Modern Biotechnology

Annual Review of Political Science (2003, Vol 6)

Annual Review of Law and Social Science, Vol. 1, 2005

Annual Review of Political Science (2004, Vol 7)

Annual Review of Political Science (2005, Vol 8)

Biotechnology Annual Review, Vol 13

Biotechnology Annual Review, Volume 9 (Biotechnology Annual Review)

Biotechnology Annual Review, Volume 10 (Biotechnology Annual Review)

Biotechnology Annual Review, Volume 12 (Biotechnology Annual Review)

Biotechnology Annual Review, Volume 14

Biotechnology Annual Review, Volume 2

Biotechnology Annual Review, Volume 3

Biotechnology Annual Review, Volume 1

Annual Review of Psychology Vol 58 2007

Annual Review of Psychology, vol 52 2001

Annual Review of Psychology, vol 53 2002

Annual Review of Psychology, vol 55 2004

Annual Review of Psychology, Vol 56 2005

Annual Review of Psychology, vol 54 2003

Annual Review of Psychology, vol 50 1999

Annual Review of Psychology Vol 57, 2006

Annual Review of Psychology, vol 51 2000

Annual Review of Psychology vol 60, 2009

Annual Review of Genomics, 2003

Annual Review of Scalable Computing

Detail Annual (vol. 4)

Detail Annual (vol. 3)

Oceanography and Marine Biology: An Annual Review, Vol. 36

Annual review of nano research

Annual Review of Law and Social Science (2007 Vol 3)

Social Policy Review 13 (No.13)

Biotechnology, 2E, Vol. 12, Modern Biotechnology

Annual Review of Political Science (2003, Vol 6)

Annual Review of Law and Social Science, Vol. 1, 2005

Annual Review of Political Science (2004, Vol 7)

Annual Review of Political Science (2005, Vol 8)

Recommend Documents