Pharming: Promises and risks of Biopharmaceuticals derived from genetically modified plants and animals (Ethics of Science and Technology Assessment)

Ethics of Science and Technology Assessment Volume 35 Book Series of the Europäische Akademie zur Erforschung von Folge...

Author: Eckard Rehbinder | Margret Engelhard | Kristin Hagen | R. B. Jørgensen | Rafael Pardo Avellaneda | Angelika Schnieke | Felix Thiele

106 downloads 642 Views 12MB 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

Ethics of Science and Technology Assessment Volume 35 Book Series of the Europäische Akademie zur Erforschung von Folgen wissenschaftlich-technischer Entwicklungen Bad Neuenahr-Ahrweiler GmbH edited by Carl Friedrich Gethmann

.

.

.

K. Hagen R. B. Jørgensen M. Engelhard E. Rehbinder A. Schnieke R. Pardo-Avellaneda F. Thiele

.

.

P harming Promises and risks of biopharmaceuticals derived from genetically modified plants and animals

123

.

Series Editor Professor Dr. Dr. h.c. Carl Friedrich Gethmann Europäische Akademie GmbH Wilhelmstraße 56, 53474 Bad Neuenahr-Ahrweiler Germany On Behalf of the Authors Professor Dr. Eckard Rehbinder Johann Wolfgang Goethe-Universität Senckenberganlage 31, 60054 Frankfurt am Main Germany Desk Editors Irene Rochlitz Herts Great Britain Katharina Mader, M.A. Friederike Wütscher Europäische Akademie GmbH Wilhelmstraße 56, 53474 Bad Neuenahr-Ahrweiler Germany

ISBN: 978-3-540-85792-1

e-ISBN: 978-3-540-85793-8

Ethics of Science and Technology Assessment ISSN: 1860-4803 e-ISSN: 1860-4811 Library of Congress Control Number: 2008935322 c Springer-Verlag Berlin Heidelberg 2009 This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilm or in any other way, and storage in data banks. Duplication of this publication or parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer. Violations are liable to prosecution under the German Copyright Law. The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. Cover design: eStudio Calamar S.L. Typesetting: Lambertz Druck, Köln, Germany Printed on acid-free paper 9 8 7 6 5 4 3 2 1 springer.com

The Europäische Akademie The Europ¨ aische Akademie zur Erforschung von Folgen wissenschaftlich-technischer Entwicklungen GmbH is concerned with the scientific study of consequences of scientific and technological advance for the individual and social life and for the natural environment. The Europäische Akademie intends to contribute to a rational way of society of dealing with the consequences of scientific and technological developments. This aim is mainly realised in the development of recommendations for options to act, from the point of view of long-term societal acceptance. The work of the Europäische Akademie mostly takes place in temporary interdisciplinary project groups, whose members are recognised scientists from European universities. Overarching issues, e.g. from the fields of Technology Assessment or Ethic of Science, are dealt with by the staff of the Europäische Akademie. The Series The series Ethics of Science and Technology Assessment (Wissenschaftsethik und Technikfolgenbeurteilung) serves to publish the results of the work of the Europäische Akademie. It is published by the academy’s director. Besides the final results of the project groups the series includes volumes on general questions of ethics of science and technology assessment as well as other monographic studies. Acknowledgement

The project “Pharming. Genetically Modified Plants and Animals as Future Production Site of Pharmaceuticals?” (“Pharming. Gentechnisch veränderte Pflanzen und Tiere als Arzneimittel-Produktionsstätten der Zukunft? Vergleich von Innovationshemmnissen und Durchsetzungschancen”) was supported by the Federal Ministry of Education and Research (Bundesministerium für Bildung und Forschung, Förderungskennzeichen 16|1547). In addition, the support of the Banco Bilbao Vizcaya Argentaria (BBVA) Foundation in Spain made the fieldwork of chapter 5 possible. The authors of this study are responsible for the content.

Preface

The Europäische Akademie deals with the scientific study of the consequences of scientific and technological advances for individuals and society, as well as for the natural environment with the lifesciences being an important focus of its work. The application of bio- and genetechnology for medical purposes has been a hot spot of research in the lifesciences for several decades. One major field is the development and production of biopharmaceuticals, with therapeutic hormones and antibodies as prominent examples. They are pharmaceutical proteins that have to be isolated from biological material or be produced by genetically modified organisms. Besides the use of fermenter grown recombinant cell cultures for their production, it is now also possible to use higher organisms (plants and animals) for this purpose. This new application of genetechnology – called “pharming” – seems to be a promising strategy to produce a broad variety of biopharmaceuticals in large quantities at comparatively low costs. It attracted special attention due to its potential for profitable investments by the pharmaceutical industry. However, taking into account the generally cautious attitudes of at least the European public towards gene- and biotechnology it is obvious that pharming should undergo a thorough evaluation of its ethical, legal, and social aspects and implications. For this task the Europäische Akademie set up an interdisciplinary and international project group that produced the report at hand. Besides it should be noted that the group consisted of senior and junior scientists contributing to the joint project on an absolutely equal footing. This show that intergenerational scientific collaboration can well transcend the often denounced state of dependence of younger scientists – given an adequate institutional framework is provided. I would like to thank the authors Dr. Margret Engelhard; Kristin Hagen, Ph.D.; Rikke Bagger Jørgensen, Ph.D.; Professor Dr. Rafael Pardo-Avellaneda; Professor Angelika Schnieke, Ph.D.; Dr. Felix Thiele, and in particular the chair Professor Dr. Eckard Rehbinder, for their dedication to this project. The Federal Ministry of Education and Research (Bundesministerium für Bildung und Forschung, BMBF) is hereby acknowledged for the funding of the project. In addition, the Banco Bilbao Vizcaya Argentaria (BBVA) foundation in Spain is thanked for their support that made the fieldwork of chapter 5 possible. Bad Neuenahr-Ahrweiler, July 2008

Carl Friedrich Gethmann

Foreword

Among the many products of modern scientific and technological innovation, gene technology has from the very beginning been highly controversial, especially for moral and environmental reasons. In the public debate, stem cell research, cloning of animals and cultivation of genetically modified plants are dominant themes. However, it is now also possible to produce biopharmaceuticals in genetically modified plants and animals. This new biotechnological method, which is called “pharming”, has a great potential on the growing market for biopharmaceuticals. It has important technical advantages over existing production methods and offers the prospect of much lower prices for pharmaceuticals, although its economic competitiveness remains to be seen. Besides benefits for producers, patients and health care systems, pharming also raises a number of complex environmental, health-related, moral, legal and social questions that have as yet not been thoroughly discussed. The degree of public awareness of the problems associated with pharming has been low. Now that the first biopharmaceuticals produced in transgenic animals have been authorized or are close to authorization in Europe and the United States, it is time to enter into an open fundamental debate about the issues raised by pharming. To evaluate the potentials and risks associated with pharming and to determine the need for, and means of, legal regulation and policy action for the responsible further development of pharming, the Europäische Akademie zur Erforschung von Folgen wissenschaftlich-technischer Entwicklungen established an international, interdisciplinary project group in 2006. Disciplines represented in and members of the group were plant biotechnology (Dr. M. Engelhard, Bad Neuenahr-Ahrweiler, project coordinator), livestock biotechnology (Professor A. Schnieke, Ph.D., Freising), ecology (R. B. Jørgensen, Ph.D., Roskilde), animal welfare (K. Hagen, Ph.D., Bad Neuenahr-Ahrweiler), social science (Professor R. Pardo-Avellaneda, Ph.D., Madrid), ethics (Dr. F. Thiele, Bad Neuenahr-Ahrweiler) and environmental law (Professor Dr. E. Rehbinder, Frankfurt a. M., chair). Over a period of two and a half years the project group held 13 internal meetings. In addition two workshops with external experts took place in Berlin in September 2006 and September 2007. The contributions of the colleagues involved profoundly enriched the study and in this respect the authors’ special thanks go to: N. S. Andersen (Roskilde), Professor Dr. D. Birnbacher (Düsseldorf), Privatdozent Dr. B. Breckling (Bremen), Profes-

X

Foreword

sor M. Eaton, Pharm.D., J.D. (Stanford), Dr. T. Fahrendorf (Langförden), A. Kind, Ph.D. (Freising) (who has in addition made specific contributions to chapter 2.3), Professor Dr. J. Luy (Berlin), Professor Dr. P. Sandøe (Kopenhagen), Dr. J. Schiemann (Braunschweig), Dr. S. Schillberg (Aachen), Dr. E. Schmitt (Darmstadt), Professor Dr. R. Müller-Terpitz (Passau), Professor B. Whitelaw, Ph.D. (Roslin), and Professor Dr. G. Winter (Bremen). For a fruitful discussion in the course of the symposium “New applications of genetic engineering in livestock”, that took place in September 2007 in Berlin, we also thank Professor Dr. L.-M. Houdebine (Jouy en Josas), Professor Dr. M. Kaiser (Oslo), Professor Dr. H. Niemann (Neustadt), Dr. C. Van Reenen (Lelystad), Professor G. Walsh, Ph.D. (Limerick), and Professor Dr. A. Zanella (Oslo) as invited speakers. Most papers of the workshops were published in 2007 in the Graue Reihe of the Europäische Akademie. Contributions of the symposium are being published parallel to this project in the same book series. For the editing I express my gratitude to I. Rochlitz, Ph.D. (Herts), K. Mader, M.A., and F. Wütscher from the Europäische Akademie. I also thank the numerous people who helped the group in organising the various meetings that took place outside the academy’s seat in Berlin, Bilbao, Bonn, Frankfurt, Freising, Madrid and Roskilde. Frankfurt am Main, July 2008

Eckard Rehbinder

Short table of contents

List of abbreviations ....................................................................................... XIX 1

Introduction .................................................................................................. 1

2

The technology of pharming ....................................................................... 9

3

Risk assessment of plant pharming and animal pharming .................. 73

4

The welfare of pharming animals ........................................................... 101

5

Public views and attitudes to pharming ................................................ 121

6

The ethical evaluation of pharming ....................................................... 179

7

The role of patents in the development of pharming .......................... 201

8

Legal problems of pharming ................................................................... 213

9

Conclusions and recommendations ...................................................... 291

Glossary ............................................................................................................. 303 Appendix: Examples of GM pharmaceutical crops and animals .............. 315 List of authors ................................................................................................... 323 Index .................................................................................................................. 329

Comprehensive table of contents

Preface ...............................................................................................................VII Foreword............................................................................................................. IX List of abbreviations ..................................................................................... XIX 1

Introduction ................................................................................................ 1 References ...................................................................................................... 7

2

The technology of pharming ......................................................................9 2.1 Recombinant pharmaceutical proteins – the advent of biotechnology ....................................................................................9 2.2 Plants as a production platform for recombinant biopharmaceuticals ............................................................................ 11 2.2.1 Genetic engineering of the host plant .................................. 13 2.2.1.1 Gene constructs ......................................................... 13 2.2.1.2 Post-translational modifications ............................. 14 2.2.1.3 Plant transformation method ................................. 15 2.2.2 Transient expression using viral vectors .............................. 20 2.2.3 Choice of species and site of production ............................. 21 2.2.3.1 Leaves ......................................................................... 21 2.2.3.2 Cereals, legume seeds and oilseeds ........................ 22 2.2.3.3 Fruits and vegetables ................................................ 22 2.2.3.4 Plant cell cultures and hairy root systems ............. 22 2.2.4 Cultivation................................................................................ 24 2.2.5 Purification of biopharmaceuticals from transgenic plants ...................................................................... 24 2.2.5.1 Purification of biopharmaceuticals from whole plants ............................................................... 24 2.2.5.2 Purification of biopharmaceuticals from plant cell cultures and hairy root cultures ............. 26 2.3 Animals as a production platform for recombinant biopharmaceuticals ............................................................................ 27 2.3.1 Transgene constructs used for animal pharming .............. 27

XIV


2.3.2 Methods of producing transgenic livestock ........................ 29 2.3.2.1 Pronuclear DNA microinjection ............................ 30 2.3.2.2 Viral gene transfer..................................................... 34 2.3.2.3 Sperm-mediated gene transfer ................................ 38 2.3.2.4 Embryonic stem cells ................................................ 39 2.3.2.5 Embryonic germ cells ............................................... 42 2.3.2.6 Nuclear transfer ......................................................... 43 2.3.2.7 Spermatogonial stem cells ....................................... 47 2.3.2.8 Adult stem cells ......................................................... 48 2.3.2.9 Overview .................................................................... 48 2.3.3 Choice of species and site of production ............................. 48 2.3.3.1 Milk ............................................................................ 53 2.3.3.2 Urine ........................................................................... 56 2.3.3.3 Seminal fluid .............................................................. 57 2.3.3.4 Blood .......................................................................... 58 2.3.3.5 Bird eggs ..................................................................... 58 2.3.4 Production of proteins from transgenic animals ............... 59 2.3.4.1 Analysis of transgenic animals ............................... 59 2.4 Quality and safety of the product ..................................................... 63 2.5 Choice of expression systems ........................................................... 64 2.6 References ............................................................................................ 66 3

Risk assessment of plant pharming and animal pharming............... 73 3.1 Environmental risks and co-existence of plants genetically modified for production of pharmaceuticals ................................. 73 3.1.1 Legal framework and basic principles of risk assessment of GM plants ............................................................................ 74 3.1.2 Risks of pharming plants ....................................................... 78 3.1.2.1 Risks of unintended exposure ................................. 78 3.1.2.2 Transgene dispersal .................................................. 80 3.1.2.3 Horizontal gene flow................................................. 91 3.1.3 The environmental risks – will pharming plants differ from the current GM plants? ...................................... 92 3.1.4 Concluding remarks ............................................................... 93 3.2 Environmental risks of animal pharming ....................................... 93 3.3 References ............................................................................................ 95

4

The welfare of pharming animals ......................................................... 101 4.1 Introduction ...................................................................................... 101 4.2 Animal welfare risks ........................................................................ 102


XV

4.3 The concept and assessment of animal welfare ............................ 104 4.4 Animal welfare considerations in the animal pharming production phase .............................................................................. 105 4.4.1 Housing and management ................................................... 106 4.4.2 Protein collection and excess offspring .............................. 108 4.4.3 Reproduction ......................................................................... 108 4.4.4 Effects of genotype ................................................................ 109 4.5 Animal welfare considerations in the development phase ......... 109 4.5.1 Transgenesis, expression of medicinal protein, and transgene evaluation ..................................................... 110 4.5.2 Reproductive technologies ................................................... 112 4.5.2.1 Developmental problems in somatic cell nuclear transfer (cloning) ...................................... 112 4.5.2.2 Donor animals and foster mothers ...................... 114 4.5.3 Excess offspring ..................................................................... 115 4.6 Conclusions ....................................................................................... 115 4.7 References ......................................................................................... 117 5

Public views and attitudes to pharming ............................................. 121 5.1 Introduction ...................................................................................... 121 5.2 Methodological considerations ...................................................... 126 5.3 Attitudes to pharming in advanced societies: awareness and evaluative perspectives ............................................................. 131 5.3.1 Awareness about pharming ................................................. 132 5.3.2 Evaluative perspectives ........................................................ 132 5.4 A differentiated landscape of perceptions of pharming ............. 136 5.4.1 Ranking of biomedical and socio-economic goals and acceptance of plant pharming ..................................... 137 5.4.2 The specifics of the means in the acceptance of plant pharming ...................................................................... 138 5.4.3 Ranking of biomedical and social goals and acceptance of animal pharming .......................................... 140 5.4.4 The specifics of the means in the acceptance of animal pharming ................................................................... 140 5.5 Preferences for methods of production of pharmaceuticals....... 142 5.6 Awareness and acceptance of plant and animal pharming ........ 143 5.7 Elements of an explanatory model ................................................ 144 5.8 Conclusions ....................................................................................... 152 5.9 Tables ................................................................................................. 153 5.10 References ......................................................................................... 176

XVI


6

The ethical evaluation of pharming ..................................................... 179 6.1 Introduction ...................................................................................... 179 6.2 Foundations of moral reasoning .................................................... 180 6.3 Common moral concerns regarding pharming ........................... 184 6.3.1 The moral status of plants and animals ............................. 185 6.3.2 Naturalness ............................................................................ 189 6.3.3 Integrity .................................................................................. 191 6.3.4 Aims and means of using and manipulating animals and plants for pharming ...................................................... 192 6.4 Risk assessment and risk-benefit analysis ..................................... 193 6.5 References .......................................................................................... 198

7

The role of patents in the development of pharming ...................... 201 7.1 The general justification of patents ................................................ 201 7.2 The existing regulatory framework ................................................ 202 7.3 Basic rules on patentability of biological products, biological material and biological and microbiological processes .............. 203 7.4 Extent of protection ......................................................................... 205 7.5 Mandatory licenses .......................................................................... 206 7.6 Patents as obstacles to innovation in pharming? ......................... 207 7.7 References .......................................................................................... 211

8

Legal problems of pharming ................................................................. 213 8.1 Introduction ...................................................................................... 213 8.2 Development phase I: Protection from risks to the environment caused by the use and release of GMOs ................ 213 8.2.1 Sources of legal regulation and their scope of application ......................................................................... 213 8.2.2 Development of recombinant medicinal products with containment .................................................................. 218 8.2.3 Development of recombinant medicinal products without containment ............................................................ 222 8.2.3.1 Scope of application and regulatory principles of Directive 2001/18................................................ 222 8.2.3.2 Information requirements and risk assessment.. 223 8.2.3.3 Authorization prerequisites ................................... 226 8.2.3.4 Special issues relating to animal pharming ......... 235 8.2.3.5 Institutional arrangements .................................... 235 8.2.4 Coexistence between experimental cultivation of GMOs and organic and conventional agriculture ....... 237


8.3

8.4

8.5 8.6

8.7

8.8

XVII

8.2.5 Waste disposal ....................................................................... 238 8.2.5.1 GMO-specific regulation ....................................... 238 8.2.5.2 Regulation under general waste law ..................... 240 8.2.5.3 Disposal of excess animals and animal parts ...... 240 Development phase II: Animal protection ................................... 241 8.3.1 Sources of regulation ............................................................ 242 8.3.2 Animal trials: Scope of application of the relevant laws .. 242 8.3.3 The European Convention and Directive 86/609 ............. 244 8.3.4 National law ........................................................................... 245 Development phase III: Protection of occupational safety and health in the development of recombinant medicinal products . 255 8.4.1 Contained use ........................................................................ 255 8.4.2 Release without containment .............................................. 256 8.4.3 General regulation of occupational safety and health ...... 257 Development phase IV: Regulation of development medicinal products .......................................................................... 257 Market authorization phase ............................................................ 258 8.6.1 Regulation 726/2004: Objectives and scope of application ..........................................................................258 8.6.2 Special regime for recombinant pharmaceuticals? ........... 259 8.6.3 Authorization prerequisites and procedure ...................... 261 8.6.4 Labelling ................................................................................. 266 8.6.5 Institutional design ............................................................... 267 Production phase ............................................................................. 267 8.7.1 Protection against risks to the environment by use and release of GMOs ............................................................ 268 8.7.2 Coexistence between pharming and conventional and organic agriculture ........................................................ 272 8.7.2.1 The problem ............................................................ 272 8.7.2.2 Sources of regulation .............................................. 273 8.7.2.3 Confinement and protection measures ............... 274 8.7.2.4 Labelling requirements........................................... 276 8.7.2.5 Liability ..................................................................... 278 8.7.2.6 Special issues in animal pharming ....................... 281 8.7.3 Animal protection ................................................................. 281 8.7.4 Production-related requirements under pharmaceuticals regulation .................................................. 282 References .......................................................................................... 283

XVIII 9


Conclusions and recommendations .................................................... 291 9.1 Pharming technology and its market ............................................ 291 9.2 Public attitudes and moral evaluation ........................................... 292 9.2.1 Attitudes ................................................................................. 292 9.2.2 Moral evaluation ................................................................... 293 9.3 The assessment and management of risks associated with pharming .................................................................................. 294 9.3.1 Principles ................................................................................ 294 9.3.1.1 Case by case ............................................................. 294 9.3.1.2 Risk-benefit evaluation .......................................... 295 9.3.1.3 Independent risk assessment research ................ 295 9.3.1.4 Transparent procedures and independence of risk assessment bodies ....................................... 295 9.3.2 Product safety and information .......................................... 296 9.3.2.1 Measures to prevent contamination and ensure product quality .................................... 296 9.3.2.2 New guidelines on pharming medicinal products and European Medicines Agency (EMEA) committee on pharming products ....... 297 9.3.2.3 Labelling and consumer information ...................297 9.3.3 Risks to the environment and food and feed chains ........ 298 9.3.3.1 Experiments and cultivation with containment, and deliberate releases ............................................ 298 9.3.3.2 Coexistence .............................................................. 300 9.3.4 Risks to animals in pharming .............................................. 300

Glossary ............................................................................................................ 303 Appendix: Examples of GM pharmaceutical crops and animals .......... 315 I. Production of molecular farmed human intrinsic factor (rhIF) in potato (Solanum tuberosum) ...................................................... 315 II. Production of Molecular Farmed human lactoferrin (rhLf) in rice (Oryza sativa) ....................................................................... 316 III. Production of antithrombin in goats’ (Capra hircus) milk......... 317 References .................................................................................................. 321 List of authors.................................................................................................. 323 Index.................................................................................................................. 329

List of abbreviations

APHIS Bt cDNA CFIA C.F.R. CHO CHO cells DEFRA DNA EC EFSA EMEA ES F1 FDA GAP GM GMM GMO GMP ICSI IV IVF IVM IVP kb

Animal and Plant Health Inspection Service (under USDA) toxin producing transgene from Bacillus thuringiensis complementary DNA Canadian Food Inspection Agency (Canadian government) Code of Federal Regulations (USA) Chinese hamster ovary Chinese hamster ovary cells Department for Environment, Food and Rural Affairs (UK government) deoxyribonucleic acid European Community European Food Safety Authority (EU) European Agency for the Evaluation of Medicinal Products embryonic stem first filial generation The Food and Drug Administration (USA) good agricultural practices genetically modified genetically modified microorganism genetically modified organism good manufacturing practices (also used elsewhere for genetically modified plants) intra-cytoplasmic sperm injection in vitro in vitro fertilization in vitro maturation in vitro production kilobase

XX LOS mb mRNA NGO OECD PCR PMI PMP PTM RNA SOPs SSCs TRIPS TSE U.S.C. USDA

List of abbreviations

large offspring syndrome megabase messenger RNA non governmental organisation Organisation for Economic Cooperation and Development polymerase chain reaction plant-made industrial product plant-made pharmaceutical post-translational modification ribonucleic acid standard operation procedures spermatogonial stem cells Agreement on Trade Related Aspects of Intellectual Property Rights Transmissible Spongiform Encephalopathy United States code United States Department of Agriculture (USA)

1 Introduction

Proteins are an important subclass of pharmaceuticals in medicine. Most pharmaceutical proteins, however, cannot easily be synthesized chemically and are called biopharmaceuticals. Until recently these were either derived from biological material such as donated blood, or produced in genetically engineered bacteria, yeast or animal cell lines1. It is now also possible to produce biopharmaceuticals in genetically modified plants and animals: recombinant human proteins have, for example, been expressed in maize kernels, tobacco leaves, goats’ milk and chickens’ eggs. Throughout the book this new technology is termed ‘pharming’ 2, composed of pharmaceutical and farming. Other terms sometimes used include ‘biopharming’, ‘gene farming’, and for plants only: ‘pharm crops’, ‘molecular farming’. Reflecting common practice, the term ‘plant pharming’ will be used here to refer to the use of whole plants, plant cell cultures, hairy root cultures, and algae, while ‘animal pharming’ refers to whole animals, but not animal cell cultures. The market for biopharmaceuticals is large and growing, with an estimated global value of $33 billion in 2004, $40 billion in 2007 and forecast to reach $70 billion by the end of the decade. There is huge potential for novel medications, significant research and development spending, and a rising number of biopharmaceutical products on the market for human use: 170 in 2007, with more than 2,000 in clinical trials3. The growing market of biopharmaceuticals includes many products, including antibodies, that could be produced by pharming. Factors contributing to the growth of the biopharmaceutical market include increases in the number of medical indications for protein therapies, and the number of patients with illnesses usually treated with pharmaceutical proteins. The provision of insulin for diabetes is one example. It has been estimated that in the year 2000, about 171 million people worldwide suffered from diabetes types I and II, and this number is projected 1 2

3

See chapter 2.1 for a brief introduction to biopharmaceuticals and biotechnology. The term ‘pharming’ is also used on the internet to denote hackers’ attacks that redirect a website’s traffic to a bogus website in order to steal identity information. Lawrence 2005; Pavlou and Reichert 2004; Pavlou and Belsey 2005; Walsh 2006; Ernst & Young 2007.

2

1 Introduction

to increase to 366 million by 20304. Although only a fraction of diabetics require insulin, the need for this protein will certainly increase. Changes in practice, such as the adoption of oral delivery rather than injection, may contribute to increased demand because substantially larger doses of insulin are required. Plant pharming has been explored as a means of supplying insulin at a reasonable price. The concept was realized in Arabidopsis thaliana in 20065 and since extended to commercial production in safflower by SemBioSys (Calgary, Canada). The company is currently planning to start clinical trials, with a projected US launch of the product in 2011. By using safflower as production platform, SemBioSys hopes to reduce insulin unit costs by 40 % or more, and capital costs by up to 70 % compared with the production in traditional expression systems, and to provide a production system that allows easy and cheap scale-ups to meet growing demands6. Another example is monoclonal antibodies, currently mainly produced in cell culture. Production costs per gram have been estimated as $300–3,000 in mammalian cell culture, $105 in transgenic goats and $50 in transgenic corn7. Monoclonals with applications in cancer are currently some of the most promising new drugs, with a large potential market in which pharming may become important. Pharming also offers savings in capital investment, because of the ease with which production can be scaled up. Growing more transgenic crops or breeding more transgenic animals is simpler and cheaper than constructing additional culture facilities for bacteria, yeast or animal cell culture. A recent study estimated the capital investment for bulk antibody production in mammalian cell culture to be at least double than that required for transgenic goats8. Pharming, particularly animal pharming, also provides a means of producing proteins which are difficult to make by other means. Many of the more complex human proteins require post-translational modifications for their assembly and bioactivity, which most microorganisms are unable to carry out. Cultured mammalian cells are able to fulfill many but not all of these functions. For example, several blood clotting factors require γ-carboxylation of glutamate residues; this is carried out poorly by chinese hamster ovary (CHO) cells, but successfully by the lactating mammary gland. The addition of sugars to proteins, termed glycosylation, is another important type of post-translational modification. Glycosylation patterns are quite different between bacteria, yeast, plants and mammals and this can have important pharmacological consequences for proteins produced 4 5 6 7 8

Wild et al. 2004. Nykiforuk et al. 2006. http://www.sembiosys.com/Main.aspx?id=14 (July 2008). Farid 2007; costs per gram estimated at a production rate of 100kg/year. At higher production rates, the savings estimates become even more pronounced. Lawrence 2007.

1 Introduction

3

in each, for example affecting bioactivity, clearing rate and immunogenicity. Post-translational modifications are therefore an important determinant in choosing particular species and cell types as an expression system (see chapter 2). In August 2006 the industry achieved a significant breakthrough when the European Commission authorized Genzyme Europe9 to market human antithrombin III (ATryn) produced in the milk of transgenic goats. Antithrombin is used for the prophylaxis of venous thromboembolism for patients with congenital antithrombin deficiency undergoing surgery. This advance however came very late compared with early industry expectations. The companies that pioneered pharming and developed the technology (PPL Therapeutics, GTC Biotherapeutics, Genfarm/Pharming) were founded in the late 1980s and had expected to bring the first product to market after five to seven years. Nevertheless, it has with hindsight been a big achievement to proceed from a theoretical possibility to market approval in less than two decades. Despite the market authorization, proof of principle and an increasing number of pharming field and clinical trials in progress (see tables 2.1, 2.7), there is still a long way to go before pharming products are accepted and used. Although pharming might offer a method of producing valuable proteins, and might realize important advantages over existing methods, its economic competitiveness remains to be proven, not least because competing technologies are also developing, for example the increasing availability of mammalian cell cultures and addition of post-translational modifications to yeast-produced proteins. Furthermore, pharming also raises a number of ecological, moral, legal and social questions. After an introductory technology chapter and an overview of potential applications, this book will assess risks and hurdles, and recommend appropriate measures for safe, acceptable and useful development of the technology. After a brief introduction to biopharmaceutical biotechnology (section 2.1), the technology of plant pharming is described in section 2.2: the methods of generating plants for the production of recombinant biopharmaceuticals, cultivation strategies and purification methods of biopharmaceuticals from transgenic plants. Section 2.3 provides an outline of basic recombinant DNA technology used in animal pharming, reviews methods of generating transgenic animals, describes technical issues affecting the choice of species and tissue used for production and briefly describes the purification of protein products from transgenic animals. Risks to humans consuming pharmaceuticals are discussed in chapter 2. Biological risks to the environment are described in chapter 3, as are the control measures necessary for agricultural coexistence between pharming and non-pharming crops. Pharming often makes use of conventional crop 9

Current authorization holder: Leo Pharma, Ballerup, Denmark.

4

1 Introduction

plants and animals that are normally used as food or feed. How (and to what extent) pharming crops and animals can be kept separate from food and feed- organisms at all stages of the production chains will be crucial. The plausibility of a scenario of food chain contamination is demonstrated by the first documented pharming accident in 2002 in the USA, when 13,000 tonnes of soy beans were contaminated with vaccine from co-mingled genetically modified maize volunteers10. It presently remains unclear whether confinement strategies are suitable to avoid plant pharming having consequences for nearby flora, fauna and soil microbiology. Often the knowledge of the potential interactions between the pharming crop and the environment is limited. This lack of knowledge leads to a risk assessment with a large degree of uncertainty (dealt with in chapter 3 on risk assessment and chapters 6 and 8, on ethics and law, respectively). With animal pharming, confinement is not considered to be a major problem. Animal pharming and plant pharming differ in a number of ways that will be touched on throughout the book. One important difference is that the animal species that are considered for pharming are generally considered to be sentient, and their potential suffering thus has to be taken into account. Chapter 4 reports on current knowledge and management suggestions with regard to adverse effects on animals in the experimental phase (making and evaluating transgenic founder animals) and in the production phase (husbandry and protein collection). The contrasting views on animal and plant pharming are an important aspect of chapter 5, in which the profile of public attitudes to pharming is presented, relying on new data from a major multi-country comparative survey of public perceptions of biotechnology. The commonalities, the national differences and also the singularities in views and acceptance of pharming are offered, illustrating the areas of consensus and disagreement across a number of European societies that may have regulatory implications as barriers and also as facilitating components for future harmonized regulation. The aim of the analysis in this chapter is to offer, for the first time, a map of attitudes to pharming in the context of general perceptions of biotechnology and science and technology at large. The explanatory role of general and highly specific variables will also be explored. Among the large set of variables for characterizing public views of pharming, a few are of prime interest: knowledge of and proximity to science, world views (particularly, views of the promise of and reservations about science, images of nature and its transformation by humans, views on animals), risk perceptions, evaluation of the genetic modification of the plants, animals and humans, the hierarchy of acceptability of different medical and socioeconomic goals potentially reachable through pharming, and views on the use 10

Fox 2003; Sauter 2005; Spök 2007.

1 Introduction

5

of different types of plants or animals. Finally, the current predisposition to take medicines produced by pharming will be charted. Chapter 6 addresses moral conflicts about pharming caused by discrepancies between the far-reaching medical and economic hopes connected to pharming, public attitudes to it, and moral concerns regarding amongst other things the moral status of animals and plants, the naturalness or unnaturalness of pharming, and the aims and means of using animals and plants for pharming. In addition, the difficulties of performing a systematic risk-benefit assessment of both animal and plant pharming are considered. The goal of the chapter is first to clarify how certain moral standpoints on pharming are structured. Given this map of moral arguments for and against pharming, a second goal chapter 6 will be to develop recommendations for mastering moral controversies on pharming. Intellectual property rights have a major impact on the development of biotechnology: Many biotechnological procedures relevant for developing marketable biotechnological products, including biopharmaceuticals, are protected by patents. In chapter 7 the framework of intellectual property rights relevant to pharming is introduced. Furthermore it is assessed whether – and if so why – patents are morally, legally, or economically questionable with respect to pharming. Chapter 8 analyses the legal situation with regard to pharming, with a focus on the situation in Europe. The development and manufacture of pharmaceuticals derived by recombinant DNA technology is regulated by different and highly complex European regulations and directives as well as by member state laws. Consequently, the relevant activities lie within the responsibility of a number of political and administrative institutions. For example, plant pharming represents for the first time a merger of green and red biotechnology, with the consequence that different regulatory regimes are applicable and different authorities are responsible. The contained use and the deliberate release, through cultivation of genetically modified organisms from which the recombinant pharmaceuticals are derived, is regulated by two EC directives and member state law on gene technology law that implements the directives. The same is true for animal pharming. The production of developmental recombinant pharmaceuticals and the placing on the market of the final preparation are covered by an EC regulation and supplementary national law. The placing on the market also requires an authorization from the European Medicines Agency. Furthermore, although pharming products are not intended to be used as food or feed, due to the risk of contamination of the food and feed chain pointed out above, there may be a need for preventive regulation under an EC regulation relating to genetically modified food and feed. Animal welfare law must be considered with respect to the use of transgenic animals for the development and – to a certain extent – the production of recombinant pharmaceuticals.

6

1 Introduction

Chapter 8 in addition analyses the relevant regulatory texts and administrative practice from the perspective of their adequacy for tackling the risks and considering the potential benefits of pharming. In particular, what steps regulatory institutions are presently taking in order to reduce the risks associated with pharming and whether this action is sufficiently protective of human health, the environment and animal welfare, will be discussed. In the final chapter of the book the implications of the analyses are presented, and recommendations for policy action are derived with a view to the responsible further development of pharming.

1 Introduction

7

References Ernst & Young (2007) Beyond Borders – Global Biotechnology Report 2007. EYGM Limited Farid SS (2007) Process economics of industrial monoclonal antibody manufacture. J Chromatogr B 848:8–18 Fox JL (2003) Puzzling industry response to ProdiGene fiasco. Nature Biotechnology 21:3–4 Knäblein J (2005) Plant-based expression of biopharmaceuticals. In: Meyers R (ed) Encyclopedia of molecular cell biology and molecular medicine. Wiley-VCH, Weinheim, pp 385–410 Lawrence S (2005) Biotech drug market steadily expands. Nat Biotechnol 23:1466 Lawrence S (2007) Billion dollar babies – biotech drugs as blockbusters. Nat Biotechnol 25:380–382 Nykiforuk CL, Boothe JG, Markley NA, Moloney MM (2006) Transgenic expression and recovery of biologically active recombinant human insulin from Arabidopsis thaliana seeds. Plant Biotechnol J 4:77–85 Pavlou AK, Reichert JM (2004) Recombinant protein therapeutics – success rates, market trends and values to 2010. Nature Biotechnology 22:1513–1519 Pavlou AK, Belsey MJ (2005) The therapeutic antibodies market to 2008. European Journal of Pharmaceutics and Biopharmaceutics 59:389–396 Sauter A (2005) Grüne Gentechnik – transgene Pflanzen der 2. und 3. Generation. Arbeitsbericht des Büros für Technikfolgen-Abschätzung beim Deutschen Bundestag, No. 104 Spök A (2007) Molecular farming on the rise – GMO regulators still walking a tightrope. Trends in Biotechnology 25:75–82 Walsh G (2006) Biopharmaceutical benchmarks 2006. Nature Biotechnology 24:769–776 http://www.who.int/diabetes/facts/en/diabcare0504.pdf (July 2008) Wild S, Roglic G, Green A, Sicree R, King H (2004) Global prevalence of diabetes. Diabetes Care 27:1047–1053

2 The technology of pharming

2.1 Recombinant pharmaceutical proteins – the advent of biotechnology In 1977 scientists succeeded in introducing the first human gene into a microorganism in order to produce a genetically engineered human protein. This “advent of biotechnology” took place only one decade after the discovery of the genetic code, which describes the connection between genes and the formation of proteins. The production of proteins by genetic engineering involves the incorporation of a foreign (often human) gene into an organism’s or a cell’s own genome. The genetically modified living system can then express so-called ‘recombinant’ proteins – it becomes a living “protein expression system”. Well-established expression systems for pharmaceutical proteins include fermenter-grown genetically modified Escherichia coli, baker’s yeasts (Saccharomyces cerevisiae) or Chinese hamster ovary (CHO) cell cultures. Proteins are linear chains of amino acids. Information about the sequence of each protein is encoded by the DNA of a gene. Expression of a protein commences when a gene is copied (transcribed) from a point termed the promoter, producing a messenger molecule or RNA. RNA is then processed to remove the non-coding regions or introns, and transported to the protein synthesis machinery. Here it is read (translated) and determines the amino acids added to a new protein molecule. Different cells require different proteins and therefore express different genes. The expression of a particular gene is controlled by the interaction of factors within the cell and DNA sequences associated with the gene. These regulatory elements include the promoter and others termed enhancers (see figure 2.1). enhancers

promoter

transcribed region

exons

introns

Figure 2.1: The basic structure of a eukaryotic gene

10


The basics of protein synthesis are highly conserved between living organisms, and this has allowed the successful transfer and expression of genes between widely different species. However many proteins require additional steps, termed post-translational modification, before they are fully functional. The types of post translational modification vary considerably between species and between cell types. The addition of sugar side chains to amino acids, called glycosylation, provides an example. The addition of complex chains of sugar molecules linked to asparagine residues in the protein chain (N-linked glycosylation) is important for the correct folding and stability of many mammalian proteins. Most bacteria are unable to glycosylate asparagine and this is a primary reason for choosing eukaryotic expression systems, including transgenic animals and mammalian cell cultures. However, the range of possible post-translational protein modifications required for protein function is very large and includes: propeptide cleavage, multichain assembly, disulphide bonding, phosphorylation, hydroxylation, amidation, methylation, hydroxylation, γ-carboxylation, acylation and lipid attachment. The repertoire of modification enzymes varies considerably between mammalian and plant tissue types. Ideally, the processing capability of the producing cells should match the requirements of the desired protein, or be readily modifiable to carry out the appropriate processing. Humulin®, the first recombinant protein for pharmaceutical use, received marketing authorization in the USA in 19821. Humulin® is recombinant human insulin produced by the bacterium Escherichia coli. Because of the complexity of the structure of insulin, it is not possible to synthesize it chemically. Therefore, before the development of Humulin® diabetes mellitus patients were treated with bovine or porcine insulin that was extracted from the pancreatic tissue of slaughtered cattle and pigs. Pharmaceuticals that cannot be synthesized chemically, but rather have to be produced by transgenic living cells or isolated from biological material (for example blood donations, animal tissues) are called biopharmaceuticals. They are therapeutic proteins, with hormones and monoclonal antibodies as the most important examples, or nucleic acidbased drugs. Most biopharmaceuticals today are modern biotechnological medicines, many of which are based on proteins produced by genetic engineering 2.

1 2

FDA 1982. Walsh 2003; Walsh 2006.

2.2 Plants as a production platform for recombinant biopharmaceuticals

11

2.2 Plants as a production platform for recombinant biopharmaceuticals In the past decade, plant-based expression systems have emerged as a possible alternative for the large-scale production of recombinant proteins. The major reason for the development of transgenic plants for the production of biopharmaceuticals was the expectation that costs of large-scale production would be comparatively low. This has turned out to be true for proteins that can be produced at high yields. For example, recombinant avidin was produced in maize at 20 % of total soluble seed protein3. Then the yield of one bushel of maize was equivalent to the total yield from one tonne of chicken eggs – the natural source of avidin. This case demonstrates the potential for cost reduction. Another important reason for the development of alternative platforms for the production of biopharmaceuticals was the hope of producing biopharmaceuticals that, so far, had to be isolated from biological material since they were too complex to be produced by recombinant microorganisms or cell culture. In addition, the risk of transmission of human pathogens via the product is minimized since plants are, in contrast to for example donated blood, not a source of human pathogens. The first pharmaceutically relevant protein made in plants was human growth hormone, expressed in transgenic tobacco in 19864. In this study the hormone was expressed as a fusion with the Agrobacterium nopaline synthase enzyme. Since then, many other human proteins have been produced in an increasingly diverse range of crops. In 2006 Dow AgroSciences received the world’s first regulatory approval for a plant-made vaccine for animals by the United States Department of Agriculture (USDA)5. It is a vaccine against Newcastle disease, which infects poultry. The vaccine is produced in genetically engineered cells from non-nicotine-producing tobacco plants and has to be administered by injection. Currently the commercialization of the chicken vaccine is not planned, since the market is already crowded. Instead, the company sought USDA approval to prime the regulatory process for other animal drugs produced in the same way6. Table 2.1 provides examples of biopharmaceuticals produced in transgenic plants. In broad terms the development and production of biopharmaceuticals in transgenic plants comprises the following steps: genetic engineering of the gene construct, transformation of the host plant (section 2.2.1), cultivation (section 2.2.4) and purification of plant-derived recombinant biopharmaceuticals (section 2.2.5). For the overall process see figure 2.2. 3 4 5 6

Hoot et al. 1997; Masarik et al. 2003. Barta et al. 1986. Katsnelson et al. 2006. www.dow.com (October 2006).

12


Table 2.1:

Some examples of biopharmaceuticals produced in transgenic plants7

Protein

Host plant

Company/ Organization

Indication/ application

Development stage

Animal vaccine

tobacco cells

USA, Dow AgroSciences

Newcastle disease in chicken

Enzyme, Glucocerebrosidase Monoclonal antibody Enzyme, gastric lipase Antibody, cancer vaccine AlphaInterferon Antigene

carrot cells

Israel, Protalix Biotherapeutics

Gaucher disease

approved by USDA 2/2006 phase 3

tobacco

USA, Planet Biotechnology, France, Meristem Therapeutics USA, Large Scale Biology

prophylaxis of caries Cystic Fibrosis

phase 2

non-Hodgkin Lymphoma

phase 2

USA, Biolex

hepatitis C

phase 2

hepatitis B

phase 2

vitamin B12 deficiency cold caused by Rhinoviruses diabetes

phase 2 phase 2

hepatitis B

phase 1

Norwalk virus

phase 1

rabies

phase 1

dry eye syndrome, gastro-intestinal infection diarrhoea

phase 1

diarrhoea

phase 1

diarrhoea hepatitis B

phase 1 phase 1

reducing adverse effects of chemotherapy

phase 1

Human intrinsic factor Antibody Insuline Vaccine Vaccine Vaccine Lactoferrin

7

maize tobacco duckweed potato

USA, Arizona State University Arabidopsis Denmark, Cobento Biotech tobacco USA, Planet Biotechnology safflower Canada, SemBioSys Genetics Inc. lettuce Poland, Polish academy of science potato USA, Arizona State University spinach USA, Thomas Jefferson University, Philadelphia maize France, Meristem Therapeutics

Vaccine

potato

Vaccine

maize

Vaccine AlphaInterferon Monoclonal antibody 7

maize duckweed

USA, Arizona State University USA, Arizona State University USA, ProdiGene USA, Biolex

not announced

USA, Planet Biotechnology

Data based on Marschall 2007, Fox 2006 and Sauter 2005.

phase 2

phase1

phase 1


13

Figure 2.2: An outline of the overall process of plant pharming

2.2.1 Genetic engineering of the host plant 2.2.1.1 Gene constructs

The first step in the construction of a transgene plant for the production of biopharmaceuticals is the identification of the gene that codes for the desired protein, and subsequently the sequencing and isolation of that gene. The advances of genomics (the study of an organism’s entire genome) and proteomics (the large-scale study of proteins, their structures and functions) have accelerated these steps greatly and in addition have lead to the development of new pharmaceutical applications. Once the sequence of the gene (or genes) coding for the protein is at hand an appropriate expression construct has to be developed. The expression construct needs to serve different tasks. One main aim of plant pharming is the production of recombinant proteins at high yields. To achieve this, expression construct design seeks to optimize all stages of gene expression, from transcription to protein stability. Expression constructs are chimeric structures, in which the transgene is flanked by various regulatory elements known to be active in plants. Only by the addition of these regulatory elements is the recombinant gene recognized by the molecular machinery of the host plant and subsequently synthesized. For high expression levels, the two most important elements are

14


the promoter (a sequence needed to “switch” the expression of a gene on) and the polyadenylation sites which are often derived from the 19S and 35S transcripts of the cauliflower mosaic virus (CaMV)8. The CaMV 35S promoter is now the most popular choice in dicotyledonous plants (dicots). However, this promoter has lower activity in monocotyledonous plants (monocots), so alternatives such as the maize ubiquitine promoter are preferred9. One of the most important factors in governing the yields of recombinant proteins is subcellular targeting, which affects the interlinked processes of folding, assembly and post-translational modification of the protein. It has, for example, been shown in comparative experiments with recombinant antibodies that the secretory pathway is a more suitable environment for folding and assembly than the cytosol10. Proteins are targeted to the secretory pathway through the inclusion of an N-terminal signal peptide in the expression construct. In the absence of targeting information, proteins in the endomembrane system are secreted into the apoplast. The apoplast is the extracellular space, which is a large and continuous network of cavities under the cell wall. Proteins secreted from the cell often remain trapped here. However, yields are generally higher compared with secretion11. Even when carefully designed, transgene expression is influenced by several factors that cannot be controlled precisely through construct design. This leads to variable transgene expression and, in some cases, to its complete inactivation. Such factors include the position of the transgene integration, the structure of the transgenic locus, gene-copy number and the presence of truncated or rearranged transgene copies. Several strategies have been adopted in an attempt to minimize variation in transgene expression, including the use of viral silencing suppressors12. Currently the minimization of positioning effects remains an active field of research. Researchers are trying to establish methods by which a single-copy transgene can be integrated into a precise location in the plant nucleus13. In practice, however, commercially developed transgenic plants undergo an enormous amount of screening to identify phenotypic, yield and agronomic variations. 2.2.1.2 Post-translational modifications

The possibility of plant-specific glycans inducing allergic responses in humans has been considered and the finding that human serum contains antibodies that are reactive against these residues has been interpreted as evidence14. In addition it has been shown that the ß(1,3)fucose and ß(1,2) 8 9 10 11 12 13 14

Irniger et al. 1992. Christensen and Quai 1996. Trombetta and Parodi 2003. Twyman et al. 2003. Brignetil et al. 1998. Butaye et al. 2005. Gomorda et al. 2005.


15

xylose residues lead to adverse reactions15. However, in general, carbohydrate epitopes are rarely allergenic. Currently it is too early to generalize about how crucial humanized glycolysation of plant-derived pharmaceuticals is, and whether it might be more important to some classes of proteins than to others. Similarly, the method of administration (oral versus injection) could make a difference in terms of the immune response that might occur. Plants are the production platform of choice in cases where innate mammalian molecules could interfere with the drug. For example, plants are being used as a production platform for the vitamin-binding recombinant human intrinsic factor (rhIF)16. Since plants do not use vitamin B12, contain vitamin B12, or have any proteins with affinity for vitamin B12, they can serve as a source for the vitamin-binding recombinant human intrinsic factor that is free from any vitamin B12 binding interferences. Porcine gastric-derived intrinsic factor preparations, in contrast, are often contaminated with haptocorrin, a vitamin B12 binding interference17. Another important reason to utilize plants as a production platform is that they are free of pathogens or prions that might be harmful to humans18. In addition, in cases where biopharmaceutical has to be stored or serve as an edible drug plants are the appropriate production platform19. 2.2.1.3 Plant transformation method

Two general methods are used to generate transgenic plant lines for pharming: Agrobacterium-mediated transformation20 and particle bombardment, in which DNA-coated microprojectiles are shot into plant tissue21. Each method has advantages and disadvantages, and the choice depends on a combination of factors, including selected host species, local expertise and intellectual property issues. Figure 2.3 illustrates the steps to be taken to achieve plant transformation with each method. Agrobacterium mediated transformation (illustrated in table 2.2) makes use of a naturally occurring pathogenic soil bacterium, which has the ability to transfer parts of its own DNA into plant cells. In the wild, transfer of a portion of the bacterial DNA (called T-DNA, for “transfer DNA”) causes rapid plant cell division leading to the formation of a tumour. Scientists have taken advantage of this naturally occurring transfer mechanism, and designed DNA vectors from the tumour-inducing plasmid DNA (ti-plasmid) found in the bacteria that is capable of carrying desired genes 15 16 17 18 19 20 21

Bardor et al. 2003. Pujol et al. 2007. www.cobento.dk (July 2008). Giddings et al. 2000. Mason et al. 1992. Schlappi and Hohn 1992. Klein et al. 1992.

16


A Agrobacterium

B particle bombardment

gene identified and isolated

gene inserted into ti-plasmid and transferred into Agrobacterium

Ti-plasmid moves into plant cell and inserts DNA into plant genome

gene replication

e.g. gold particles coated with DNA

cells shot with gene gun and DNA incorporated into the plant cell genome

transformed cell selected with selectable marker and transgenic plant regeneration from single transformed cell

Figure 2.3: Strategies for genetic engineering of plants

into the plant. The engineered or constructed genes are inserted into the Agrobacterium vectors and enter the plant by the bacteria’s own internal transfer mechanisms. For Agrobacterium to transfer part of its DNA into plants, living, wounded plant tissue is usually inoculated with the bacterium. After culturing the bacteria with the plant tissues, antibiotics are supplied, eliminating the bacterium from the plant tissue. The transformed plant tissue is then regenerated into a mature plant through tissue culture techniques22.

22

Kumria et al. 2001.


17

Table 2.2: Agrobacterium mediated transformation of plants

For the transfer of foreign genes, leaf pieces are incubated with the transformed Agrobacterium tumefasciens strain.



Leaf discs after transformation with Agrobacterium. For re-growth they are then cultivated on selective medium giving rise only to those cells which have been genetically transformed by the Agrobacteria.

 The transformed plant tissues can be regenerated to intact plants. Regenerated transformed tobacco plants grown in-vitro.

Particle bombardment, also referred to as the biolistic system, is a physical method for DNA delivery (illustrated in table 2.3). For this method, DNA is coated onto small ( zygote

No

No

Mosaic possible, multiple integrations rare

Yes, reduced by mosaicism

Very rare, insertional mutagenesis possible

Mouse, cattle, pig, sheep, goat, rabbit

0.5–1 %

Retroviral Transduction

Viral vector => packaging cells => infectious virus => zygote or oocyte

Retroviral sequences

No

Mosaic possible, multiple integrations common

Yes, reduced by mosaicism

Very rare, insertional mutagenesis or gene activation possible

Mouse, cattle, pig

50–80 %

SpermMediated Transfer

DNA => sperm => natural fertilisation or ICSI

No

No

Fragmented transgenes common by fertilisation

??

Insufficient data

Mouse, pig

34 %3

Embryonic Stem Cells (ES)

DNA=> ES cells => incorporate into diploid embryo

Selectable marker

Yes

Chimera

Yes, reduced by chimerism


Mouse,

Not done


DNA=> ES cells => incorporate into tetraploid embryo

Selectable marker

Yes

Completely ESderived

Yes


Mouse

Not done

Continued on next page


Method

Method

Transfer route

Additional DNA

Gene targeting

Founder animal


DNA ➞ ES cells ➞ form male or female gametes ➞ fertilisation

Selectable marker

Yes

Nuclear Transfer

DNA ➞ cells ➞ nuclear transfer

Selectable marker

Spermatogonial Stem Cells (SS)

DNA ➞ SS cells ➞ incorporate into embryo

Spermatogonial Stem Cells (SS)

DNA ➞ SS cells ➞ transplant into testis ➞ fertilisation

Embryonic Germ Cells (EG)

Germline transmission

Inadvertent health effects

Demonstrated in mammalian species

Efficiency in livestock (oocyte/live offspring4)

Heterozygous for Yes ES genotype

Imprinting problems, poor development, low viability

Mouse

Not done

Yes

Nuclear genome completely cellderived, mitochondria from oocyte

Yes

High founder mortality, morbidity. Subsequent generations OK

Mouse, sheep, 0.5–5% pig, cattle, goat1

Selectable marker

Yes

Chimera

Unknown

Unknown

Mouse

Not done

Selectable marker

Yes

Heterozygous for Yes SS genotype

Insufficient data

Mouse, goat

?%3

DNA ➞ EG cells Selectable marker ➞ incorporate into embryo

Yes

Chimera

Imprinting problems

Mouse

Not done

Not demonstrated in mammals2

51

Notes: 1. Use of F1 hybrid ES cells as nuclear donors shows significantly reduced mortality and morbidity in mice. 2. Germline transmission of EG genotype demonstrated in chickens. 3. Data from one laboratory. 4. In large animals, oocytes and zygotes are now almost always derived from slaughterhouse material. The efficiency of the procedure, therefore, primarily affects the number of animals required as recipients to gestate manipulated embryos.

2.3 Animals as a production platform for recombinant biopharmaceuticals

Table 2.5: Methods of producing transgenic animals

52


Table 2.6: Basic reproductive data for livestock mammals Species

Gestation period (months)

Age at sexual maturity (months)

Number of offspring

Age at first lactation (months)

Recombinant protein production (kg/individual/yr)

Cattle

9

16

1

33

40–80

Goat

5

8

1–2

18

4

Pig

4

6

~10

16

1.5

Sheep

5

8

1–3

18

2.5

Rabbit

1

5

~8

7

0.02

In birds, both research and production have focused on the domestic chicken. A modern hen produces more than 300 eggs per year, and the relatively short time to sexual maturity (circa 5 months) allows for rapid expansion of transgenic flocks. As outlined in the introduction, the main reason to choose to express a particular protein in animals or cultured animal cells, rather than bacteria, plants or yeast, is because functional and/or immunological properties require addition of appropriate sugar chains (oligosaccharides) to the amino acid chain. This process is termed glycosylation. The pattern and type of protein glycosylation vary widely between microorganisms, plants and mammals. Oligosaccharide groups exist as either short chains linked to either serine or threonine amino acid residues in the protein chain (O-linked glycosylation), or as longer complex branching chains linked to asparagine (N-linked glycosylation). Many mammalian proteins require N-linked glycosylation for correct folding and stability; most bacteria are unable to do this. Plants can carry out N-linked glycosylation, but the sugars added are frequently very different to those present on mammalian proteins. Notably, plants do not add a group of sugars termed sialic acids, which frequently terminate oligosaccharide chains on glycoproteins. Glycosylation patterns also vary between mammalian species, tissue types and even between metabolic states of the same tissue. Importantly, humans differ from the majority of other mammals in the type of sialic acids present. The two principle forms are N-acetylneuraminic acid (Neu5Ac) and N-glycolylneuraminic acid (Neu5Gc). Humans have a mutation in the enzyme responsible for producing Neu5Gc and therefore lack this form. Oligosaccharide analysis of immunoglobulins from different species illustrates these differences. Only Neu5Ac: Mainly Neu5Ac: Mainly Neu5Gc: Only Neu5Gc:

Human, chicken Guinea pig Rat, rabbit, dog, cat Cattle, sheep, goat, horse, mouse, rhesus monkey


53

Expressing a protein in a species or tissue different from its normal location may therefore result in altered glycosylation. The importance and possible pharmacological consequences of differences in glycosylation must be thoroughly investigated by molecular, biochemical and physiological analysis, and ultimately by clinical trial. Two important considerations are the effects on pharmokinetic properties, i.e. activity, rate of clearance from the body and immunogenicity. It is known that humans carry antibodies that recognize Neu5Gc. Glycosylation, while important, is however only one of many possible modifications that may be required for correct protein function. These include: propeptide cleavage, multichain assembly, disulphide bonding, phosphorylation, hydroxylation, amidation, methylation, hydroxylation, γ-carboxylation, acylation and lipid attachment. The repertoire of enzymes that carry out these functions varies considerably between mammalian tissue types. Ideally, the protein processing capability of the producing cells should match the requirements of the desired protein, or be readily modifiable to carry out the appropriate processing. The current state of the art, however, offers only a limited choice regarding site of production in transgenic animals, and the ability of different tissues to express and process exogenous proteins has not been comprehensively studied. In birds, animal pharming has focused exclusively on protein production in the white of eggs. In mammals, four sources of exogenous proteins have been studied to date, each of which are body fluids. These are: milk, urine, seminal fluid and blood. Fluids are more suitable than solid tissues for this purpose because they are renewable and can be obtained without harm or excessive invasion. Furthermore, many biomedically important proteins are themselves secreted into body fluids. Milk is by far the best studied of these production systems and the only method that has been examined on a large-scale. 2.3.3.1 Milk

Milk was a natural focus for the development of animal pharming, see review89. The lactating mammary gland has a huge capacity to synthesize proteins and other biochemicals for infant nutrition. The dairy industry is well established and scientifically advanced, not only in cattle but also sheep and goats. The necessary equipment and expertise required for the collection, processing and early stage purification of transgenic milk are therefore readily available. The major milk proteins are caseins, with five types in mice, four in sheep and cows and two in human. Caseins are hydrophobic and associate into spherical complexes that form a fatty suspension. The soluble, or whey, fraction of milk contains hydrophilic proteins that differ between mam89

Clark 1998.

54


malian species. Whey acidic protein (WAP) is the major whey protein in rodent milk and is also present in pigs, while β-lactoglobulin (BLG) is the major whey protein in sheep, goats and cows; both are absent from human milk. α-lactalbumin has a role in lactose synthesis and is present in the whey of all milks that contain lactose. In 1989 John Clark of the Roslin Institute, Edinburgh, demonstrated that the BLG promoter could be used to direct the expression of the human blood clotting Factor IX gene in sheep and that the product was secreted during lactation90. Since then, the promoters and regulatory sequences of almost all major milk genes have been utilized and investigated for their suitability in driving the expression of potentially useful proteins. A large number and wide variety of foreign proteins have been expressed in the milk of transgenic animals and expression levels as high as 35g/L have been achieved91. In each case the foreign protein is secreted as part of the whey fraction. This work has included: complex multichain proteins, for example fibrinogen; combinations of transgenes designed to supplement the natural protein processing abilities of the lactating mammary gland, for example prolyl hydroxlyase co-expressed with type 1 procollagen; and coexpression of transgenes to improve the stability of secreted protein in milk, for example α1-antitrypsin protease inhibitor with fibrinogen. Hundreds of transgenes have been “trialled” in the lactating mammary gland, with the great majority in mice. Pilot studies in mice provide an indication of the feasibility of a large animal study and indicate whether any adverse effects to animal health can be expected, an important issue in the expression of highly bio-active recombinant proteins such as erythropoietin. It is difficult to provide a definitive list of proteins expressed in milk, because some of this work has been carried out by companies and not made public. Those published, or otherwise known to the authors, are listed below: – Anti-microbial proteins: Lysozyme, lactoferrin, tissue non-specific alkaline phosphatase, lysostaphin, antimicrobial peptides for example β-defensins. – Blood clotting and anti-clotting factors: Antithrombin III, protein C, factor VII, factor VIII, factor IX, fibrinogen, tissue plasminogen activator, hementin, urokinase, thrombin activated plasminogen. – Cell surface proteins expressed in soluble form: CD4 (HIV receptor), transferrin receptor, cystic fibrosis transmembrane conductance regulator, intercellular adhesion molecule 1 (human rhinovirus receptor), pentraxins for example serum amyloid P and C-reactive protein. – Cytokines and growth factors: Erythropoietin, Interleukin 2, Interleukin 10, thrombopoietin, insulin-like growth factor 1, nerve growth factor b, granulocyte colony stimulating factor, Interferonγ. 90 91

Clark et al. 1989. Wright et al. 1991.


55

– Detoxifying enzymes: Butyrylcholinesterase. – Digestive enzymes: Bile salt stimulated lipase. – Hormones: Follicle stimulating hormone, lutenising hormone, parathyroid hormone, growth hormone, leptin. – Milk proteins: α-lactalbumin, αs1-casein, β-casein, κ-casein, β-lactoglobulin, whey acid protein. – Immunoglobulins: Many types of single chain antibodies and also monoclonal antibodies composed of both light and heavy chains. – Protease inhibitors: α1-antitrypsin, lung elastase inhibitor, C1 inhibitor (complement system inhibitor). – Protein modification enzymes: These are usually co-expressed with other products. Furin, prolyl hydroxylase, glycosyltransferases. – Peptides: These are usually expressed as a fusion with another carrier protein. Calcitonin, amylin, anti-microbial peptides. – Structural proteins: Type 1 collagen, type 2 collagen, spider dragline silk. – Viral and microbial proteins for vaccine production: Rotavirus Vp2 and Vp6 antigens, malaria parasite surface antigens, hepatitis B virus surface antigen. – Others: Transferrin, serum albumin, alpha-fetoprotein, hepatocarcinoma-intestine-pancreas/pancreatic-associated protein, n-3 fatty acid desaturase, stat5 transcription factor, endostatin, zona pellucida glycoprotein 3, acid alpha-glucosidase, extracellular superoxide dismutase, pulmonary surfactants B and C, n-3 fatty acid desaturase. Of the subset of these proteins investigated in livestock, only a very small number have continued to preclinical and clinical trials and only one product has so far gained regulatory approval. Those currently in commercial development are summarized in table 2.7. A significant drawback to milk is that it is a complex and rich mixture of proteins, lipids and carbohydrates. Therefore, purification of the desired protein requires multiple steps which can be costly. Protein purity is of paramount importance where the protein product is to be administered to patients on a long-term basis, especially intravenously, because even minute quantities of contaminating milk components could be immunogenic. Purification can be circumvented in those special applications where milk is to be ingested as a nutraceutical. For example, transgenic goats have been produced that secrete enhanced amounts of the anti-microbial protein lysozyme into their milk. The intention is to mimic human breast milk, which is also rich in lysozyme, to alter gut flora and combat gastrointestinal microbial infections92. 92

Maga et al. 2006.

56


Table 2.7: Proteins produced in milk currently in commercial development93 Product

Animal

Company

Indication/ application

Development stage

Antithrombin III (ATryn®)

Goat

GTC, US

Antithrombin deficiency

Approved in EU. Phase 3 clinical trials in US

C1 inhibitor

Rabbit

Pharming, Holland

Hereditary angiodema

Phase 3 clinical trials

Alpha fetoprotein

Goat

Merrimack and GTC, US

Rheumatoid arthritis

Phase 2 clinical trials

Alpha glucosidase

Rabbit

Pharming, Holland

Pompe’s disease

Phase 2 clinical trials, on hold

Growth hormone

Cow

Biosidus, Argentina

Dwarfism

Preclinical

Lactoferrin

Rabbit

Pharming, Holland

Infection inflammation

Preclinical

Collagen

Rabbit, Cow

Pharming, Holland

Various biomaterials

Preclinical

Fibrinogen

Rabbit, Cow

Pharming, Holland, GTC, US

Tissue sealant

Preclinical

Albumin

Cow

GTC, US

Excipient, blood expander

Preclinical

Alpha 1 antitrypsin

Goat

GTC, US

Hereditary AAT deficiency

Preclinical

Malaria vaccine

Goat

GTC, US

Malaria

Preclinical

CD137 antibody

Goat

GTC, US

Solid tumours

Preclinical

Rotavirus pseudoviral particles

Rabbit

BioProtein, France

Antigen carriers for vaccine

Preclinical

Butyrylcholinesterase

Goat

Pharmathene, US

Organophosphate poisoning

Preclinical

2.3.3.2 Urine

The mammary gland has proved to be an unsuitable site for the production of some highly bioactive proteins, such as growth factors or cytokines, because they can enter the general circulation and affect the physiology of 93

Data from Biopharm International 1st August 2006 and Nature Biotechnology (2006) 24:877.


57

the animal. In contrast, the contents of the bladder, being potentially noxious, are sequestered from the body. A system to express foreign proteins in urine was developed in the late 1990s using the uroplakin genes94. Uroplakins are membrane-associated proteins expressed specifically in the differentiated uroepithelium of the bladder and urethra. The mouse uroplakin II gene promoter has been used to direct expression of human growth hormone (hGH) and also human granulocyte macrophage-colony stimulating factor (hG-CSF) in mice. Production in the kidney has also been investigated using the gene promoter of Tamm Horsfall protein, also called uromodulin, which is expressed and secreted from the epithelium of the ascending loop of Henle. Mice expressing and secreting hGH into urine have been produced. It is not yet clear how appropriate urine is as a source of bioactive proteins. Although the body may not be exposed to the natural contents of the bladder, segregation of transgenic proteins secreted into this compartment still depends on the tissue specificity of the gene promoters. Ectopic transgene expression may lead to circulating proteins. Notably, both the uroplakin II hG-CSF mice and the Tamm Horsfall hGH mice showed evidence of transgene protein in peripheral blood. The greatest problem with this method of production is, however, the low synthetic capacity of bladder and kidney, which is far less than the mammary gland. The yield of protein per ml is therefore very low, in the order of ng/ml. While this may be suitable for certain high value proteins, the practical usefulness of the system remains to be demonstrated. Proponents do, however, point out that low yield is partially compensated by the large volume of urine obtained from animals such as cattle. Unlike milk, urine is produced during the lifetime of the animal independent of age, sex and lactation. Furthermore, because urine contains little protein and lipid, product purification should, in theory, be simpler than from milk. The stability of recombinant protein in urine is, however, a potential problem that remains to be fully explored. Work published so far has been restricted to mice, although there have been preliminary reports of transgenic pigs expressing hGH in urine. 2.3.3.3 Seminal fluid

Porcine seminal fluid has been suggested as a suitable source for bioactive proteins95. Proponents state that the accessory male sex glands have a substantial protein synthetic capacity, semen is available in reasonably large volumes (200–300 ml per ejaculate) and, because protein secretion is strictly exocrine, bioactive proteins could be produced without adversely affecting the animal. This work is at an early stage and determination of its useful94 95

Kerr et al. 1998. Dyck et al. 1999.

58


ness will require further investigation. This will include identifying appropriate genes and sequences to drive secretion into seminal fluid. Possible candidates are the spermadhesins, the major protein component in porcine semen. Other important factors are the protein processing capacity of the producing tissue, stability of foreign proteins in semen and ease of product purification. 2.3.3.4 Blood

The physiology and development of the animal are highly exposed to any adverse effects of bioactive proteins circulating in the blood, therefore the range of suitable products is very restricted. Human haemoglobin for use in synthetic blood substitutes has been produced in pigs by the company DNX of Princeton, New Jersey 96, but this was discontinued because of difficulties purifying the human protein away from the very similar porcine protein. Production in the blood of transgenic livestock will likely gain prominence as a source of human polyclonal antibodies. Progress is being made towards the production of animals with humanized immune systems97. Such animals could, in principle, be immunized against a wide range of antigens to provide an abundant source of human polyclonal antibodies. These are likely to play an important role in passive immunotherapy in the future and offer considerable advantages over monoclonal antibodies. For example, they are more effective than monoclonals in immune complex formation and better mimic the natural immune response; they can also disable pathogens which require neutralization of multiple epitopes, pathogens with diverse strains and venoms with multiple toxic components. Importantly, polyclonal antibodies can only be produced in people or transgenic animals. Most applications would require large animals for production of adequate quantities of serum. 2.3.3.5 Bird eggs

Chicken eggs have several advantages that make them attractive for the production of foreign proteins. The poultry industry is well developed and modern breeds of chickens are highly productive, laying about one egg per day. Collection of eggs is very simple and can be scaled up easily. Production is also very flexible, large flocks of birds can be rapidly produced from a single transgenic male. Furthermore, the use of eggs for pharmaceutical purposes is already established for the production of vaccines, providing a framework of regulatory guidelines for good manufacturing practice. Production of therapeutic proteins in eggs is less advanced than production in milk, because of the technical problems of avian transgenesis. No products are as yet in the commercial pipeline, but several companies 96 97

Swanson et al. 1992. Kuroiwa et al. 2002; Jakobovits et al. 2007.


59

are actively pursuing product development. Proponents point out that the chicken may be more suitable than mammalian systems for certain proteins. For example, some bioactive proteins with toxic effects in mammals may not affect birds. There is also evidence that chickens and human proteins have similar glycosylation patterns, as discussed in earlier, however current data are restricted to a few proteins and considerably more information will be required to assess the system properly. The secretory cells of the chicken oviduct certainly have a high protein synthetic capacity. Each egg contains approximately 4g protein in the white, of which more than 54 % is ovalbumin. Other major protein constituents are ovotransferrin (12 %), ovomucoid (12 %) and lysozyme (3.4 %). This low protein complexity should simplify purification, while natural protease inhibitors present in albumin may also help stabilize foreign proteins. Chicken ES-like cells transfected with an ovalbumin gene construct, containing 7.5–15kb of the ovalbumin 5’ regulatory sequences that direct expression of human immunoglobulin heavy and light chains, have been used to generate somatic chimeric hens that secreted biologically active antibody into the egg white98. However, there was evidence of ectopic transgene expression. This was not observed in more recent experiments99 which used lentiviral vectors containing only about 3kb of regulatory sequences for the expression of an interferon or a miniantibody for cancer treatment.

2.3.4 Production of proteins from transgenic animals 2.3.4.1 Analysis of transgenic animals

Analysis of integrated transgenes. In 1995, the United States Food and Drug Administration (FDA) produced guidelines for pre-market data submission for potential products from transgenic sources. Amongst other specifications, these require that the structure and expression pattern of the integrated transgene construct be characterized in the founder animal and demonstrated as reliable through subsequent generations. Each animal line destined for commercial production should be analysed to determine the structure, integrity, copy number and integration site of each integrated transgene. This analysis will include Southern hybridization of the genomic DNA to identify the lengths of various restriction fragments predicted from the construct structure. Fluorescent in situ hybridization of metaphase chromosome spreads can also be employed to identify the chromosomal location(s) of integrated transgenes. Molecular cloning of the integrated transgene and its proximal flanking regions may be required to determine the DNA sequence of the integrated transgene locus. 98 99

Zhu et al. 2005. Lillico et al. 2007.

60


DNA introduced into mammalian embryos by microinjection tends to integrate as tandem repeats generally oriented head to tail and usually, but not always, at a single locus randomly located in the host genome. Introduction of transgenes by cell transfection has broadly similar results, but often leads to a more complex transgene array at the integration site. Lentiviral vectors also integrate randomly but as single copies at each integration site. Multiple copies of a viral transgene in a founder animal will therefore segregate in subsequent generations in Mendelian fashion. Transgene loci produced by random cell transfection differ from those produced by DNA microinjection because selectable marker genes are necessarily introduced with the transgene; these will typically encode resistance to a commonly used antibiotic, for example G418, blasticidin or puromycin. To avoid possible gene flow from the transgenic animal to prokaryotes, bacterial gene promoters are excluded from the selectable marker genes. Antibiotic resistance genes can also be flanked by site-specific recombination elements, such as loxP substrate sites for Cre recombinase, allowing their removal. However, in multiple arrays this may result in large deletions. Transgene loci produced by gene targeting are quite distinct from random events, in that a correctly targeted locus will carry a single copy of the predetermined engineered change and the rest of the genome is left unaltered. An antibiotic or other selectable marker gene is necessarily included at the target site, but again can be removed by site-specific recombination if necessary. Multiple transgenes that are co-injected or co-transfected generally co-integrate at the same locus. Founder animals carrying high transgene copy numbers are frequently chosen to establish transgenic lines because they often produce the most abundant levels of expression, but it has been observed that such lines can undergo transgene silencing or recombination and copy loss over generations. Transgene copy loss occurs most frequently where elements in a tandem array are in inverse relative orientation. Such configurations tend to be unstable and can lead to deletions, duplications and incomplete genes. Incomplete genes are particularly undesirable because, where breaks occur within coding sequences, shifts in the translational reading frame can lead to the expression of truncated and/or aberrant protein species. Arrays of multiple transgenes can be complex to analyse. Analysis of transgene mRNA expression. The pattern of transgene expression should be characterized to determine its tissue specificity. This is primarily for the benefit of the producing animals, to assess whether any undue deleterious effects are likely to arise as a consequence of inappropriate transgene expression. Samples of a wide variety of tissue types obtained by necropsy of transgenic animals are analysed by reverse transcriptase PCR (RT-PCR),


61

or Northern hybridization to detect spatial, or temporal ectopic expression of the transgene. The significance of any ectopic transgene expression will depend on the level and site of expression and the nature of the encoded protein. Transgene mRNA expressed by the appropriate tissue should be rigorously characterized to identify the full range of mRNA species present. This is necessary to determine the integrity of the mRNA and whether it is correctly spliced. Aberrant mRNAs, even if present as only minority species, can encode aberrant proteins with possibly significant clinical consequences. Analysis of transgene protein expression. The aims of transgenic protein analysis are to determine: whether a protein is fully functional, if degradation occurs for example in milk, and whether expression levels are sufficient for commercially viability. One then has to investigate to what extent the transgenic recombinant protein product resembles the native form, and whether any differences affect function, stability, half-life and immunogenicity. To this end considerable efforts will be made analysing protein products by functional assay, mass spectroscopy, peptide mapping, protein sequencing and glycoprotein analysis. Clearly, any human pharmaceutical product should be of consistent quality. Variations in expression level can affect protein structure. For example, if the post-translational modification capacity of the producing tissue is limited, high levels of expression may exceed that limit and result in partially or unmodified protein and altered bioactivity. The amount of protein produced by individuals in a transgenic herd or flock should, therefore, vary as little as possible. Acceptable upper and lower limits should be set to allow standardization and quality assurance of the purification process. The purity of the protein preparation will clearly be an important factor in the assessment of any transgenic product. This is especially important where it is to be administered intravenously. Producers must ensure removal of host animal proteins and DNA, chemical reagents and ensure exclusion of potential pathogens such as microorganisms, viruses and prions. Collection, processing and protein purification. Basic collection and processing methods for large quantities of milk and eggs are well established. Collection procedures suitable for bulk collection of other fluids such as urine or semen have yet to be devised. Large-scale recombinant protein purification has so far only been developed for milk. This is a multi-step process that combines standard methods developed for the dairy industry with procedures developed for purification of recombinant proteins produced in cell culture. The level of purity required of a particular product is determined by its application. If

62


the protein is to be ingested as a nutraceutical, then skimmed milk could be suitable. If, however, the product is to be injected intravenously on a regular basis over long periods, then very high levels of purity would be required. The nature of the protein will determine the specifics of its purification. As most recombinant proteins are present in the whey fraction, the first steps are removal of fat and suspended caseins by procedures that may include: centrifugation, acid or PEG precipitation or chymosin treatment and/or microfiltration. This would then be followed by a series of chromatography steps to isolate the recombinant protein away from the whey, remaining milk proteins and other contaminants. Final clean-up steps might include ultrafiltration and possibly heat treatment to prepare a pharmaceutical-grade therapeutic product. Current experience indicates a final yield of purified product of between 40–60 % of the amount in milk, depending on the nature of the protein and the required purification procedure. The greatest loss tends to be during casein removal. This may be reduced by treatment with chelating agents that deform casein micelles and release associated recombinant protein. Standards for processing plants are equivalent to those already established for the purification of recombinant proteins from cell culture, or native proteins derived from human sources such as blood. Requirements for the process included validation for product safety and pathogen removal. All procedures have to be carried out according to good manufacturing practice (GMP) and using standard operating procedures (SOPs). Where cell culture manufacturers are required to maintain duplicated banks of cells to ensure product continuity, transgenic manufacturers would maintain sperm banks. Animal husbandry. Regulations for the housing of transgenic animals will vary between different countries (see chapter 8). Veterinary health monitoring is required and, in addition, transgenic animals should be observed for any effects arising from recombinant protein expression. Generally, all animals will be kept under some type of containment regime, for example in double-fenced fields with each animal marked by identification tags and subject to strict accounting. Procedures for disposal of waste matter and cadavers to ensure suitable containment should be observed. EU rules for general animal husbandry (see section 8.2) also apply to transgenic herds or flocks. Animals are generally raised according to their species’ needs and requirements. Some restrictions to their freedom of movement may apply due to laws regarding genetically modified organisms (GMO) or the need for safekeeping from, for example, damage by animal rights activists. As a precaution, human access might be restricted.

2.4 Quality and safety of the product

63

2.4 Quality and safety of the product In addition to the characterization steps detailed in section 2.2.5.1 for plant pharming and in 2.3.4.1 for animal pharming, several other factors must be addressed to ensure product quality and safety. The health of production animals is important, not only to protect their own well-being but also to avoid possible transmission of zoonotic disease. As described above, regular monitoring by veterinarians is required. Strict precautions should also be taken to prevent contact with other farm animals or wild animals, or people or equipment that have been in recent contact with either. Concerns regarding transmission of prion diseases (BSE, scrapie) also mean that land used as animal pasture should not have had contact with other farm animals for several years. For this reason, some companies have chosen to raise animals in countries free of prion diseases, for example New Zealand. Alternatively, animals such as pigs and rabbits can be raised indoors in specific pathogen free facilities to minimize the risk of infectious diseases. Other precautions include exclusion of noxious agents, such as plant toxins or synthetic chemicals, for example pesticides. In the case of plant pharming, great care has to be taken to avoid contamination with toxic or noxious soil constituents, chemicals present in the environment, and on the harvesting machinery, and chemicals like fertilizers or pesticides applied to crops and soil. Also soil bacteria, parasites, animal excreta and other unwanted substances preferably should be removed from the harvest before further processing can begin. As with all biopharmaceuticals, production from transgenic animals and plants must comply with current FDA or EMEA guidelines and GMP. GMP compliance is a legal requirement (see chapter 8) and includes training of personnel, validation of procedures, equipment, materials and facilities, as well as SOPs. Production criteria must be defined at the outset, such as acceptance criteria for source material, product pooling, batch size and the product quality and purity required at various stages during purification to ensure product consistency. Throughout, documentation is essential and meticulous records must be kept of all activities, from the production of the DNA construct all the way to the final product. The purified product should be characterized prior to final formulation in a manner similar to other biopharmaceuticals. In this regard the concept of the “well-characterized biologic” has been very important. This was defined in the US Federal register in 1996 as “a chemical entity whose identity, purity, impurities, potency, and quantity can be determined and controlled”. Biopharmaceuticals of all types have sometimes encountered problems meeting this strict definition, and it is considerably more difficult than for chemically produced products. It is conceivable that this will be revised in the future. With regard to purity, acceptable low levels can be set for such contaminants as pathogens, host proteins, DNA and reagents

64


used in the purification process. However, it is more difficult to set acceptable standards for “identity” because very minor differences in protein glycosylation, folding and other post-translational modifications can alter bioactivity, efficacy and immunogenicity, possibly resulting in allergic or other adverse reactions in the patient. Some early biological products are starting to come off patent, providing opportunities for cost-effective production in transgenic animals. These follow up products – so called “similar biological medicinal products” or “biosimilars” – will have to adhere to equally high quality and safety standards. The necessary legal framework has been established by regulatory authorities in the European Union. In the United States, the FDA are initiating discussions on this topic.

2.5 Choice of expression systems Pharming has significantly extended the range of possible expression systems for biopharmaceutical proteins. Producers can choose between fermenter-grown transgenic mammalian cell cultures, transgenic bacteria or yeasts, transgenic plants or plant cell cultures and transgenic animals. The choice of production method is determined by several factors: the folding complexity of the protein, the nature and extent of post-translational processing required for protein activity, the quantity required and the value and the physiological function of the protein. Mammals are more appropriate than plants or microorganisms for the expression of proteins requir-

Yes







Transgenic animals

Small



Mammalian cell cultur

Large



Transgenic plants

Small



Yeast and bacteria

Quantity required

Mammalian pattern glycosylation, or complex protein processing required No

Large

Quantity required

Figure 2.12: Choice of production platform for the manufacture of recombinant

proteins

2.5 Choice of expression systems

65

ing mammalian patterns of glycosylation, or complex post-translational processing for bioactivity. Because of the ease of scale-up, it is often argued that transgenic animals and plants are more suitable than cultured cells for proteins that are required in large volumes. This is summarized in figure 2.12 above.

66


2.6 References Alan H, Christensen1and PH (1996) Quail1 Ubiquitin promoter-based vectors for high-level expression of selectable and/or screenable marker genes in monocotyledonous plants. Transgenic Research 5(3):213–218 Azzoni AR, Kusnadi AR, Miranda EA, Nikolov ZL (2002) Recombinant aprotinin produced in transgenic corn seed: extraction and purification studies.Biotechnol Bioeng 5;80(3):268–276 Bardor M, Faveeuw C, Fitchette A-C, Gilbert D, Galas L, Trottein F, Faye L, Lerouge P (2003) Immunoreactivity in mammals of two typical plant glyco-epitopes, core (1,3)-fucose and core xylose. Glycobiology 13(6):427–434 Barta A, Sommergruber K, Thompson D, Hartmuth K, Matzke MA, Matzke AJM (1986) The expression of nopaline synthase – human growth hormone chimaeric gene in transformed tobacco and sunflower callus tissue. Plant Mol Biol 347–357 Beals TP, Goldberg RB (1997) A novel cell ablation strategy blocks tobacco anther dehiscence. Plant Cell 9:1527–1545 Bosch P, Pratt SL, Stice SL (2006) Isolation, characterization, gene modification, and nuclear reprogramming of porcine mesenchymal stem cells. Biol Reprod 74:46–57 Brigneti G, Voinnet O, Li W-X, Ji L-H, Ding S-W, Baulcombe DC (1998) Viral pathogenicity determinants are suppressors of transgene silencing in Nicotiana benthamiana. The EMBO Journal 17:6739–6746 Buchschacher GL Jr (2001) Introduction to retroviruses and retroviral vectors. Somat Cell Mol Genet 26:1–11 Burkhardt PK, Beyer P, Wünn J, Klöti A, Armstrong GA, Schledz M, von Lintig J, Potrykus I (1997) Transgenic rice (Oryza sativa) endosperm expressing daffodil (Narcissus pseudonarcissus) phytoene synthase accumulates phytoene, a key intermediate of provitamin A biosynthesis. Plant J 11(5):1071–1078 Butaye KMJ, Cammue BPA, Delauré SL, De Bolle MFC (2005) Approaches to Minimize Variation of Transgene Expression in Plants. Molecular Breeding 16(1):79–91 Campbell KH, McWhir J, Ritchie WA, Wilmut I (1996) Sheep cloned by nuclear transfer from a cultured cell line. Nature 380:64–66 Campbell KH, Alberio R, Choi I, Fisher P, Kelly RD, Lee JH, Maalouf W (2005) Cloning: eight years after Dolly. Reprod Domest Anim 40:256–268 Capecchi MR (2005) Gene targeting in mice: functional analysis of the mammalian genome for the twenty-first century. Nat Rev Genet 6:507–512 Clark AJ, Bessos H, Bishop JO, Brown P, Harris S, Lathe R, McClenaghan M, Prowse C, Simons JP, Whitelaw CBA, Wilmut I (1989) Bio/Technology 7:487–492 Clark AJ (1998) The mammary gland as a bioreactor: expression, processing, and production of recombinant proteins. J Mammary Gland Biol Neoplasia 3:337–350 Decker EL, Reski R (2007) Moss bioreactors producing improved biopharmaceuticals. Curr Opin Biotechnol 18(5):393–398 Dobrinski I (2005) Germ cell transplantation and testis tissue xenografting in domestic animals. Anim Reprod Sci 89:137–145 Drossart J (2004) Downstream processing of plant derived recombinant therapeutic proteins. In: Fischer R, Schillberg S (eds) Molecular farming. Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Dyck MK, Gagne D, Ouellet M, Senechal JF, Belanger E, Lacroix D, Sirard MA, Pothier F (1999) Seminal vesicle production and secretion of growth hormone into seminal fluid. Nature Biotechnol 17:1087–1090

2.6 References

67

EMEA (2002) CPMP. Points to consider on quality aspects of medical producs containing active substances produced by stable transgene expression in higher plants. CPMP/BWP/764/02 (draft). The European Agency for the Evaluation of Medical Products (EMEA) Ernst&Young (2003) Zeit der Bewährung – Deutscher Biotechnologiereport 2003 Fahrner RL, Knudsen CD, Basey CD (2001) Industrial purification of pharmaceutical antibodies: development, operation, and validation of chromatography processes. Biotechnol Genet Eng Rev 18:301–327 FDA (2002) Guidance for industry: Drugs, biologics, and medical devices derived from bioengineered plants for use in humans and animals (draft guidance). United States Food and Drug Administration FDA (1982) Drug Bull 12(3):18–19 Gelvin SB (2003) Agrobacterium-Mediated Plant Transformation: the Biology behind the “Gene-Jockeying” Tool. Microbiol Mol Biol Rev 67(1):16–37 Giddings G, Allison A, Brooks D, Carter A (2000) Transgenic plants as factories for biopharmaceuticals. Nature Biotechnology 18:1151–1155 Gomord V, Chamberlain P, Jefferis R, Faye L (2005) Biopharmaceutical production in plants: problems, solutions and opportunities. Trends in Biotechnology 23(11):559–565 Gordon JW, Scangos GA, Plotkin DJ, Barbosa JA, Ruddle FH (1980) Genetic transformation of mouse embryos by microinjection of purified DNA. Proc Natl Acad Sci U S A 77:7380–7384 Grill LK, Palmer KE, Pogue GP (2005) Use of Plant Viruses for Production of PlantDerived Vaccines. Critical Reviews in Plant Sciences 24(4):309–323 Guan K, Nayernia K, Maier LS, Wagner S, Dressel R, Lee JH, Nolte J, Wolf F, Li M, Engel W, Hasenfuss G (2006) Pluripotency of spermatogonial stem cells from adult mouse testis. Nature 440:1199–1203 Hammer RE, Pursel VG, Rexroad CE Jr, Wall RJ, Bolt DJ, Ebert KM, Palmiter RD, Brinster RL (1985) Production of transgenic rabbits, sheep and pigs by microinjection. Nature 315:680–683 Hitz C, Wurst W, Kühn R (2007) Conditional brain-specific knockdown of MAPK using Cre/loxP regulated RNA interference. Nucleic Acids Res 35:e90 Hofmann A, Kessler B, Ewerling S, Weppert M, Vogg B, Ludwig H, Stojkovic M, Boelhauve M, Brem G, Wolf E, Pfeifer A (2003) Efficient transgenesis in farm animals by lentiviral vectors. EMBO Rep 4:1054–1060 Hood EE, Witcher DR, Maddock S, Meyer T, Baszczynski C, Bailey M, Flynn P, Register J, Marshall L, Bond D, Kulisek E, Kusnadi A, Evangelista R, Nikolov Z, Wooge C, Mehigh RJ, Herman R, Kappel WK, Ritland D, Li CP, Howard JA (1997) Commercial production of avidin from transgenic maize: Characterization of transformant, production, processing, extraction and purification. Mol Breed 3:291–306 Hubner K, Fuhrmann G, Christenson LK, Kehler J, Reinbold R, De La Fuente R, Wood J, Strauss JF 3rd, Boiani M and Schöler HR (2003) Derivation of oocytes from mouse embryonic stem cells. Science 300:1251–1256 Irion S, Luche H, Gadue P, Fehling HJ, Kennedy M, Keller G (2007) Identification and targeting of the ROSA26 locus in human embryonic stem cells. Nat Biotechnol 25:1477–1482 Irniger S, Sanfaçon H, Egli CM, Braus GH (1992) Different sequence elements are required for function of the cauliflower mosaic virus polyadenylation site in Saccharomyces cerevisiae compared with in plants. Mol Cell Biol 12(5):2322–2330 Jakobovits A, Amado RG, Yang X, Roskos L, Schwab G (2007) From XenoMouse technology to panitumumab, the first fully human antibody product from transgenic mice. Nat Biotechnol 25:1134–1143

68


Jenke AC, Stehle IM, Herrmann F, Eisenberger T, Baiker A, Bode J, Fackelmayer FO, Lipps HJ (2004) Nuclear scaffold/matrix attached region modules linked to a transcription unit are sufficient for replication and maintenance of a mammalian episome. Proc Natl Acad Sci U S A 101:11322–1137 Jung HS, Koo JK, Lee SJ, Park CI, Shin JY, Kim MH, Tan HK, Lim SM, Kim DI (2006) Characterization of human cytotoxic T lymphocyte-associated antigen 4-immunoglobulin (hCTLA4Ig) expressed in transgenic rice cell suspension cultures. Biotechnol Lett 28(24):2039–2048 Kanatsu-Shinohara M, Inoue K, Lee J, Yoshimoto M, Ogonuki N, Miki H, Baba S, Kato T, Kazuki Y, Toyokuni S, Toyoshima M, Niwa O, Oshimura M, Heike T, Nakahata T, Ishino F, Ogura A, Shinohara T (2004) Generation of pluripotent stem cells from neonatal mouse testis. Cell 119:1001–1012 Kanatsu-Shinohara M, Ikawa M, Takehashi M, Ogonuki N, Miki H, Inoue K, Kazuki Y, Lee J, Toyokuni S, Oshimura M, Ogura A, Shinohara T (2006) Production of knockout mice by random or targeted mutagenesis in spermatogonial stem cells. Proc Natl Acad Sci U S A 103:8018–8023 Kato Y, Imabayashi H, Mori T, Tani T, Taniguchi M, Higashi M, Matsumoto M, Umezawa A, Tsunoda Y (2004) Nuclear transfer of adult bone marrow mesenchymal stem cells: developmental totipotency of tissue-specific stem cells from an adult mammal. Biol Reprod 70:415–8 Katsnelson A, Ransom J, Vermij P, Waltz E (2006) News In Brief, Nature Biotechnology 24:233–234 Kerr DE, Liang F, Bondioli KR, Zhao H, Kreibich G, Wall RJ, Sun TT (1998) The bladder as a bioreactor: Urothelium production and secretion of growth hormone into urine. Nature Biotechnol 16:75–79 Kim TG, Baek MY, Lee EK, Kwon TH, Yang MS (2008) Expression of human growth hormone in transgenic rice cell suspension culture. Plant Cell Rep 27(5):885–891 Kioussis D, Festenstein R (1997) Locus control regions: overcoming heterochromatin-induced gene inactivation in mammals. Curr Opin Genet Dev 7:614–619 Klein TM, Arentzen R, Lewis PA, Fitzpatrick-McElligott S (1992) Transformation of microbes, plants and animals by particle bombardment. Biotechnology 3:286–291 Knäblein J (2005) Plant-based expression of biopharmaceuticals. In: Meyers R (ed) Encyclopidia of molecular cell biology and molecular medicine. 2nd Edition. Volume 10. Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Kumria R, Waie B, Rajam MV (2001)Plant regeneration from transformed embryogenic callus of an elite indica rice via Agrobacterium. Plant Cell, Tissue and Organ Culture 67(1):63–71 Kuroiwa Y, Kasinathan P, Choi YJ, Naeem R, Tomizuka K, Sullivan EJ, Knott JG, Duteau A, Goldsby RA, Osborne BA, Ishida I, Robl JM (2002) Cloned transchromosomic calves producing human immunoglobulin. Nat Biotechnol 20:889–894 Kuvshinov V, Anissimov A, Yahya BM (2004) Barnase gene inserted in the intron of GUS – a model for controlling transgene flow in host plants. Plant Science 167(1):173–182 de Laat W, Grosveld F (2003) Spatial organization of gene expression: the active chromatin hub. Chromosome Res 11:447–59 Labosky PA, Barlow DP, Hogan BL (1994) Mouse embryonic germ (EG) cell lines: transmission through the germline and differences in the methylation imprint of insulin-like growth factor 2 receptor (Igf2r) gene compared with embryonic stem (ES) cell lines. Development 120:3197–3204 Lal P, Ramachandran VG, Goyal R, Sharma R (2007) Edible vaccines: current status and future. J Med Microbiol 25(2):93–102 Review Lavitrano M, Busnelli M, Cerrito MG, Giovannoni R, Manzini S, Vargiolu A (2006) Sperm-mediated gene transfer. Reprod Fertil Dev 18:19–23

2.6 References

69

Lee SJ, Park CI, Park MY, Jung HS, Ryu WS, Lim SM, Tan HK, Kwon TH, Yang MS, Kim DI (2007) Production and characterization of human CTLA4Ig expressed in transgenic rice cell suspension cultures. Protein Expr Purif 2:293–302 Lillico SG, Sherman A, McGrew MJ, Robertson CD, Smith J, Haslam C, Barnard P, Radcliffe PA, Mitrophanous KA, Elliot EA, Sang HM (2007) Oviduct-specific expression of two therapeutic proteins in transgenic hens. Proc Natl Acad Sci U S A 104:1771–1776 Love J, Gribbin C, Mather C, Sang H (1994) Transgenic birds by DNA microinjection. Biotechnology (N Y) 12:60–63 Ma JK, Drake PM, Christou P (2003) The production of recombinant pharmaceutical proteins in plants. Nat Rev Genet 4(10):794–805 Maga EA, Walker RL, Anderson GB, Murray JD (2006) Consumption of milk from transgenic goats expressing human lysozyme in the mammary gland results in the modulation of intestinal microflora. Transgenic Res 15:515–519 Manzini S, Vargiolu A, Stehle IM, Bacci ML, Cerrito MG, Giovannoni R, Zannoni A, Bianco MR, Forni M, Donini P, Papa M, Lipps HJ, Lavitrano M (2006) Genetically modified pigs produced with a nonviral episomal vector. Proc Natl Acad Sci U S A 103:17672–7677 Marshall B (2007) PMPs in clinical trials and advanced PMIPs - Research findings first presented at FinMed 2006, on 30th March 2006 - table last updated Oct.07, http://www.molecularfarming.com/PMPs-and-PMIPs.html (July, 2008) Masarik M, Kizek R, Kramer KJ, Bilova S, Brazdova M, Vacek J, Bailey M, Jelen F, Howard JA (2003) Application of avidin-biotin technology and adsorptive transfer stripping square-wave voltammetry for detection of DNA hybridization and avidin in transgenic avidin maize. Anal Chem 75:2663–2669 Mason HS, Lam DM, Arntzen CJ (1992) Expression of hepatitis B surface antigen in ransgenic plants. Proc Natl Acad Sci USA 89:1174–11749 Matsumoto S, Ikura K, Ueda M (1995) Plant Molecular Biology 27:1163–1172 McCreath KJ, Howcroft J, Campbell KHS, Colman A, Schnieke AE, Kind AJ (2000) Production of gene-targeted sheep by nuclear transfer from cultured somatic cells. Nature 405:1066–1069 Meissner A, Wernig M, Jaenisch R (2007) Direct reprogramming of genetically unmodified fibroblasts into pluripotent stem cells. Nat Biotechnol 25:1177–1181 Moreira PN, Pozueta J, Pérez-Crespo M, Valdivieso F, Gutiérrez-Adán A, Montoliu L (2007) Improving the generation of genomic-type transgenic mice by ICSI. Transgenic Res 16:163–168 Nayernia K, Nolte J, Michelmann HW, Lee JH, Rathsack K, Drusenheimer N, Dev A, Wulf G, Ehrmann IE, Elliott DJ, Okpanyi V, Zechner U, Haaf T, Meinhardt A, Engel W (2006a) In vitro-differentiated embryonic stem cells give rise to male gametes that can generate offspring mice. Dev Cell 11:125–132 Nayernia K, Lee JH, Drusenheimer N, Nolte J, Wulf G, Dressel R, Gromoll J, Engel W (2006b) Derivation of male germ cells from bone marrow stem cells. Lab Invest 86:654–663 Nykiforuk CL, Boothe JG, Murray EW, Keon RG, Goren HJ, Markley NA, Moloney MM (2006) Transgenic expression and recovery of biologically active recombinant human insulin from Arabidopsis thaliana seeds. Plant Biotechnology Journal 4(1):77–85 Oback B, Wells DN (2007) Donor cell differentiation, reprogramming, and cloning efficiency: Elusive or illusive correlation? Mol Reprod Dev 74:646–654 Okita K, Ichisaka T, Yamanaka S (2007) Generation of germline-competent induced pluripotent stem cells. Nature 448:313–317 Petitte JN, Liu G, Yang Z (2004) Avian pluripotent stem cells. Mech Dev 121:1159–1168 Pfeifer A (2004) Lentiviral transgenesis. Transgenic Res 13:513–522

70


Prelle K, Zink N, Wolf E (2002) Pluripotent stem cells-model of embryonic development, tool for gene targeting, and basis of cell therapy. Anat Histol Embryol 31:169–186 Pujol M, Gavilondo J, Ayala M, Rodríguez M, González EM, Pérez L (2007) Fighting cancer with plant-expressed pharmaceuticals. Trends in Biotechnology Vol 25, Issue 10:455–459 Rapp JC, Harvey AJ, Speksnijder GL, Hu W, Ivarie R (2003) Biologically active human interferon alpha-2b produced in the egg white of transgenic hens. Transgenic Res 12:569–575 Sanford JC, Klein TM, Wolf ED, Allen N (1987) Delivery of substances into cells and tissues using a particle bombardment process. Journal of Particulate Science and Technology 6:559–563 Sanford JC (1988) The Biolistic Process. Trends in Biotechnology 6:299–302 Sato M (2006) Direct gene delivery to urine testis as a possible means of transfection of mature sperm and epithelial cells lining epididymal ducts. Reprod Med & Biol 5:1–7 Sauter A (2005) Grüne Gentechnik – transgene Pflanzen der 2. und 3. Generation. Arbeitsbericht des Büros für Technikfolgen-Abschätzung beim Deutschen Bundestag, No. 104 Schlappi M, Hohn B (1992) Competence of Immature Maize Embryos for Agrobacterium-Mediated Gene Transfer .The Plant Cell, 4(1):7–16 Schnieke AE, Kind AJ, Ritchie WA, Mycock K, Scott AR, Ritchie M, Wilmut I, Colman A, Campbell KHS (1997) Human factor IX transgenic sheep produced by transfer of nuclei from transfected fetal fibroblasts. Science 278:2130–2133 Shadwick FS, Doran PM (2007) Infection, propagation, distribution and stability of plant virus in hairy root cultures. Journal of Biotechnology 131(3):318–329 Shadwick FS, Doran PM (2004) Foreign Protein expression using plant cell suspention and hairy root cultures. In: Fischer R, Schillberg S (eds) Molecular farming. Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Sharp JM, Doran PM (2001) Strategies for enhancing monoclonal antibody accumulation in plant cell and organ cultures. Biotechnol Prog 17(6):979–992 Stöger E, Vaquero C, Torres E, Sack M, Nicholson L, Drossard J, Williams S, Keen D, Perrin Y, Christou P, Fischer R (2000) Cereal crops as viable production and storage systems for pharmaceutical scFv antibodies. Plant Mol Biol 42(4):583–90 Swanson ME, Martin MJ, O’Donnell JK, Hoover K, Lago W, Huntress V, Parsons CT, Pinkert CA, Pilder S, Logan JS (1992) Production of functional human hemoglobin in transgenic swine. Biotechnology (N Y) 10:557–559 Takahashi K, Tanabe K, Ohnuki M, Narita M, Ichisaka T, Tomoda K, Yamanaka S (2007) Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell 131:861–72 Trombetta ES, Parodi AJ (2003) Quality control and protein folding in the secretory pathway. Annu Rev Cell Dev Biol 19:649–676 Twyman RM, Stoger E, Schillberg S, Christou P, Fischer R (2003) Molecular farming in plants: Host systems and expression technology. Trends Biotechnol 21 (12):570–578 Vajta G, Gjerris M (2006) Science and technology of farm animal cloning: state of the art. Anim Reprod Sci 92:211–30 Wakayama T, Perry AC, Zuccotti M, Johnson KR and Yanagimachi R (1998) Fullterm development of mice from enucleated oocytes injected with cumulus cell nuclei. Nature 394:369–374 Walsh G (2006) Biopharmaceutical benchmarks 2006. Nat Biotechnol 7:769–776 Walsh G (2003) Biopharmaceuticals: Biochemistry and Biotechnology. Wiley Wells DN (2005) Animal cloning: problems and prospects. Rev Sci Tech 24:251–264

2.6 References

71

Whitelaw CB, Radcliffe PA, Ritchie WA, Carlisle A, Ellard FM, Pena RN, Rowe J, Clark AJ, King TJ, Mitrophanous KA (2004) Efficient generation of transgenic pigs using equine infectious anaemia virus (EIAV) derived vector. FEBS Lett 571:233–236 Willadsen SM (1986) Nuclear transplantation in sheep embryos. Nature 320:63–65 Wilmut I, Schnieke A, McWhir J, Kind AJ, Campbell KHS (1997) Viable offspring derived from foetal and adult mammalian cells. Nature 385:810–813 Wobus AM, Boheler KR (2005) Embryonic stem cells: prospects for developmental biology and cell therapy. Physiol Rev 85:635–678 Wright G, Carver A, Cottom D, Reeves D, Scott A, Simons P, Wilmut I, Garner I and Colman A (1991) High level expression of active human alpha-1-antitrypsin in the milk of transgenic sheep. Biotechnology 9:830–834 Yu J, Vodyanik MA, Smuga-Otto K, Antosiewicz-Bourget J, Frane JL, Tian S, Nie J, Jonsdottir GA, Ruotti V, Stewart R, Slukvin II, Thomson JA (2007) Induced pluripotent stem cell lines derived from human somatic cells. Science 318:1917–1920 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, Winters-Digiacinto P, Wong S, Zhou W, Etches RJ (2005) Production of human monoclonal antibody in eggs of chimeric chickens. Nat Biotechnol 23:1159–1169

3 Risk assessment of plant pharming and animal pharming

In August 2006, the European Commission approved the first animal pharming product for human use, ATryn®, which is now in phase III clinical testing in USA. More animal pharming products are expected to be commercialized soon, and plant-made pharmaceuticals from transgenic plants are also at an advanced stage before commercialization. However, these new production platforms for pharmaceuticals may have consequences to the environment, including animals and humans. Production of pharmaceuticals will now take place “out of the laboratory” – in some cases it will be totally uncontained. Thus pharming animals and plants may expose the near environment to the active products, or the GM animals and plants may disperse and affect other ecosystems. The new production platforms challenge the present regulation which governs conventional pharmaceutical products, GM animals and plants, and animal welfare. Current legislation on these issues is complicated, in some cases overlapping, in other cases insufficient, and therefore new procedures and regulation are demanded. The intention of the following chapters is to present the unwanted environmental effects in relation to production of pharmaceuticals in GM animals and plants, and to shed some light on the current procedures used for estimation of effects. The benefits of pharming are included in chapters 1 and 2, risks relating to animal welfare are dealt with in chapter 4, and the legal aspects of this production are analysed in chapter 8.

3.1 Environmental risks and co-existence of plants genetically modified for production of pharmaceuticals Production of genetically modified pharming crops brings up several challenges for regulators, risk assessors and plant pharming companies. Most of the challenges arise from the cultivation of these pharming plants in the open field, where dispersal of pharming plants, their transgenes and nontarget effects to the environment are possible scenarios. Identification and evaluation of environmental hazards may be rather uncertain, as pharming plants often represent new combinations of trait, plant and environment – combinations not known from traditional plant breeding or previous trans-

74


genic plants. An uncontained field production is a risk, not only for the environment but also for plant pharming companies and the food and feed industry. Serious economic consequences might result if they are not able to prevent adventitious presence of pharming products in food and feed. Even in cases where there are no, or very minor, health or environmental risks associated with the production, the public is not likely to accept commingling, for social or ethical reasons (chapters 5 and 6).

3.1.1 Legal framework and basic principles of risk assessment of GM plants To avoid adverse effects to human health and environment, the release of a GM plant to the open environment is preceded by an environmental risk assessment. The EU Directive 2001/181 sets forth procedures for releasing GMOs into the environment and principles for the environmental risk assessment necessary in all cases. However, this regulation has been developed for food and feed and may not apply to the release issues raised by plants being used as production platforms for pharmaceuticals. Risk assessment of GM plants is based on the information provided by the GM plant producer. For food/feed, EFSA (European Food Safety Authority) has provided a guidance document2 that sets out the information and procedure. No such guidelines exist for GM pharming plants yet, but they are under development by EFSA. EFSA has recently drafted its opinion on the risk assessment of GM plants for non-food and non-feed purposes including GM plants producing pharmaceuticals3: EFSA considers that the general requirements found in their guidance documents on GMO for food and feed, will also apply to GM plants for non-food and nonfeed. However, as risks to humans and animals from accidental exposure might be a key point for some of the medicinal plants, the importance of exchange of information between EMEA (European Agency for the Evaluation of Medicinal Products (now European Medicines Agency) responsible for the clinical testing of medicinal products)) and EFSA is stressed. Box 3.1 summarizes the information that has to be provided for the risk assessment of GM food/feed plants. The EU Directive 2001/18 on the deliberate release GMOs sets forth two different authorization tracks. One track is for deliberate releases of GMPs for any purpose other than for placing on the market (Part B); these releases are limited both in time and area, and they are thoroughly monitored by the authorities. The other track (Part C) is for placing on the market of GMPs as or in products (includes cultivation, import, transport, processing, handling, storage). Part B procedures are national and authorization can be granted by the respective Member State; 1 2 3

Directive 2001/18/EC. EFSA 2004. www.efsa.europa.eu/EFSA/efsa_locale-1178620753812_1178716609288.htm (July 2008).

3.1 Environmental risks and co-existence of plants genetically modified Box 3.1:

75

Guidance document of the scientific panel on genetically modified organisms for the risk assessment of genetically modified plants and/or derived food and feed (EFSA 2004, revised 2006)

Information required in applications for GM plants and/or derived food and feed A. General information B. Information relating to the recipient or (where appropriate) parental plants C. Information relating to the genetic modification 1. Description of the methods used for the genetic modification 2. Nature and source of vector used 3. Source of donor DNA, size and intended function of each constituent fragment of the region intended for insertion D. Information relating to the GM plant 1. Description of the trait(s) and characteristics which have been introduced or modified 2. Information on the sequences actually inserted or deleted 3. Information on the expression of the insert 4. Information on how the GM plant differs from the recipient plant in: reproduction, dissemination, survivability 5. Genetic stability of the insert and phenotypic stability of the GM plant 6. Any change to the ability of the GM plant to transfer genetic material to other organisms 7. Information on any toxic, allergenic or other harmful effects on human or animal health arising from the GM food/feed 7.1 Comparative assessment 7.2 Production of material for comparative assessment 7.3 Selection of material and compounds for analysis 7.4 Agronomic traits 7.5 Product specification 7.6 Effect of processing 7.7 Anticipated intake/extent of use 7.8 Toxicology 7.9 Allergenicity 7.10 Nutritional assessment of GM food/feed 7.11 Post-market monitoring of GM food/feed 8. Mechanism of interaction between the GM plant and target organisms (if applicable) 9. Potential changes in the interactions of the GM plant with the biotic environment resulting from the genetic modification 9.1 Persistence and invasiveness 9.2 Selective advantage or disadvantage 9.3 Potential for gene transfer 9.4 Interactions between the GM plant and target organisms 9.5 Interactions of the GM plant with non-target organisms 9.6 Effects on human health 9.7 Effects on animal health 9.8 Effects on biogeochemical processes 9.9 Impacts of the specific cultivation, management and harvesting techniques 10. Potential interactions with the abiotic environment 11. Environmental Monitoring Plan 11.1 General 11.2 Interplay between environmental risk assessment and monitoring 11.3 Case-specific GM plant monitoring 11.4 General surveillance for unanticipated adverse effects 11.5 Reporting the results of monitoring

76


the plants or products must not be used for commercial purposes. Part C entails a centralized procedure involving the European Commission and all Member States in both risk assessment and decision making, and the authorizations grant commercialization in all EU countries. For part B and part C releases, the information requested for risk assessment of the GM plant and its interaction with the environment is the same. What differentiates part C releases is that information is also requested on labelling, use, handling, storage, traceability, monitoring and issues in relation to administration of the marketed product (see chapter 8 on legal issues). The introduction of a GM plant into the environment is a stepwise process. Initial releases to the environment are small field trials. The normal practice is that the scale of the releases increases gradually (step by step), if the risk evaluation of the preceding step indicates that the next step can be taken. Finally, the approved GM crop can be cultivated in the agro-ecosystem together with other Non-GM crops. This “mixed” cultivation is governed by co-existence legislation, which aims at ensuring the free choice of production system and thus free choice among products. This implies that Non-GM and GM production must be kept separate. One assumes that GM pharming plants are only released into the environment if they are considered safe; however, to manage potential risks most pharming plants will be released with a statutory management regime attached to them. In addition, rules of co-existence can help to mitigate risks of mixing. The assessment of risks to the environment and to health preceding the release of a GM crop into the environment should ensure that direct and indirect effects, as well as immediate and delayed effects, are assessed on a case by case basis, taking into account the nature of the GMO and the receiving environment (including other GMOs already released to the environment). In the case of a market approval of the GMO, labeling, monitoring and information to the public are also demanded. The legislation calls for assessment of the accumulated effects, meaning that all effects are evaluated, and the total effect is settled. Here too, the management strategies for coping with the possible effects should be taken into consideration before the total effect can be estimated. The environmental risk assessment consists of: 1. Hazard identification: what can go wrong (identify the events) and why 2. Probability analysis: how often do these events happen (events/time) 3. Consequence analysis: how much harm is caused by the event (consequences/event) 4. Risk calculation: Risk = probability x consequence 5. Scientific reasoning, uncertainty and significance analysis: how sure are we of the risk estimated and how important is this type of risk 6. Risk management: What can be done to reduce the risk?

3.1 Environmental risks and co-existence of plants genetically modified

77

In the case of plant pharming, the hazards are the unwanted effects on health and environment from the pharming plant itself, its exudations or decomposing parts. However, as they are new inventions – compared to conventionally bred crops and first generation GM crops (with herbicide tolerance and insect resistance) – many pharming plants and their transgene products will challenge our ability to identify the hazards. Probability analysis for pharming plants will mostly be an exposure analysis. Exposure routes of plant pharming products and dose-response from the exposure may also be difficult to evaluate and thus, render the probability analysis difficult. Probability and consequence are combined in the risk algorithm, risk = probability x consequence, and make it appear that risk assessment is exact and mathematical. However, GM plants are interacting with very complex ecosystems. Therefore, it is difficult to put figures on probability and consequence, and thus the risk is never expressed as a precise figure but in more general terms like “no perceived risk”, “low risk”, “medium risk”, “high risk” etc. Often the available scientific knowledge is limited, which may increase uncertainty. The uncertainty of risk assessment can also be influenced by the applicant, who produces and collects the information on which risk assessment is based. Even though the new directive (2001/18) sets forth the principle for the environmental risk assessment, various authorities responsible for the environmental risk assessment may identify or consider risks differently. Due to the uncertainty always inherent in the risk assessments, the assumptions behind the assessments should be clearly presented, and the risk assessment procedure as a whole should be transparent to laymen. Once the GMP has been approved for commercial release, its cultivation is regulated by legislation on co-existence. The purpose of the co-existence rules is to limit unintended mixture of GMP and Non-GMP. Co-existence of GMPs with conventional and organic crops requires care during production, and specific control measures which often go beyond good farming practice. The co-existence regulation is based on national legislation and non-mandatory EU recommendations. The EU has, however, common rules for labelling and traceability of food and feed with contents of GMOs (Regulations 1829/20034 and 1830/20035). The allowable adventitious level of GMOs in food and feed is presently 0.9 % for GMOs having passed all authorization stages, including a full risk assessment and final approval by the respective national and European Food Safety Authority (EFSA); the testing for GM contents in food and feed in European countries is done by spot testing. Defining how the thresholds should be determined is under discussion by the EU6. At the moment the threshold for GM pharm4 5 6

Regulation (EC) 1829/2003 on genetically modified food and feed. Regulation (EC) 1830/2003 concerning the traceability and labeling of genetically modified organisms. Joint Research Centre, Explanatory document on the use of “Percentage of GMDNA copy number”.

78


ing plants in food and feed is 0.0 %. The present legislation on GM plants was intended for use in food and feed, and it does not cover GM pharming plants in a satisfactory way (see chapter 8 on legal matters).

3.1.2 Risks of pharming plants When cultivating pharming plants in the open, two main processes may cause risks to environment and health: – unintended exposure, – transgene dispersal. These two processes are of course linked, as transgene dispersal may bring about unintended exposure at new localities. Magnitude of exposure and gene dispersal will depend on the size of the production area. Pharming plants are currently supposed to be cultivated only in medium to small scale fields7 (see also chapter 2), as the demand for most pharming products can be satisfied by highly productive plants grown in a small area. Therefore, environmental effects may also be restricted in space. However, within these areas the pharming plant cultivation could entail not only environmental effects but also changes to the agricultural structure, as most pharming crops will require intense agricultural management to enhance product quality and minimize effects to the environment. The farming practices will depend on crop type and the surrounding agricultural landscape. It should be kept in mind that even if pharming plants are only cultivated on limited areas, gene flow, plant invasions and elution of degraded plant parts may have long-range effects on other areas. 3.1.2.1 Risks of unintended exposure

The pharming plants are designed and optimized to produce products with biological effects on humans or animals. Therefore, unintended exposure to these pharming products might be of concern. For example, the erroneous intake or exposure to pharming plants or their pollen by humans, herbivores or pollinating insects could be a risk, due to toxicity or effects on physiology or behavioral patterns. Most of the plant species used for pharming are also used for human or domestic animal nutrition, and wildlife in the agro-ecosystem will use these crop plants as feed, for shelter or breeding. After intake, the pharming products could be passed on along the food chain. The effects to the environment and health could be caused by toxicity, hormonal or allergenic effects from the product, for example unintended exposure to plant-produced vaccines may lead to desensitization, and then a proper immune response might not develop when the patient is vaccinated8. Effects from exposure to the pharming plants in the field will, 7 8

Fischer et al. 2001; Sparrow et al. 2007; Spök 2007. Kirk et al. 2005.


79

of course, depend on bioactivity of the pharming product; in some cases bioactivity is first reached by processing after harvest. A detrimental effect to biodiversity is one risk scenario of pharming plants. Pharmaceutical products could decrease population size of affected plants and animals, for example production of human lactoferrin in rice could affect birds and rodents that consume rice seeds spilled at harvest. It has been shown that certain pathogenic microorganisms can use the lactoferrin as an iron source9, and therefore it is possible that animals that feed from the rice would have a higher frequency of severe infections. Conversely, pharmaceutical products, to use the same example with lactoferrin, could increase population sizes and thereby affect the composition of ecosystems, for example it was shown that chickens fed the transgenic lactoferrin rice had improved health and growth compared to controls10. Exudations from roots or decomposing pharming plants (for example from waste disposal) may affect flora, fauna, soil and water quality. Today’s use of medicines and cosmetics is already jeopardizing some ecosystems through sewage systems and water run-off11, and likewise leaching from pharming plants may pose problems. It has been shown that exudations of biopesticides from plant roots may have transient effects on protozoan soil Assessment of the toxicity Identification of the hazard

Risk assessment Assessment of the exposure

Characterize physical setting of exposure

Estimate absorption or intake of toxin

Identify potential exposure pathways

Estimate exposure concentrations Identify potential exposed populations

Figure 3.1: Steps in quantifying exposure and how exposure relates to the

overall risk assessment12

9 10 11 12

Weinberg 1999; Dhaenens et al. 1997; Vogel et al. 1997. Humphrey et al. 2002. Nash et al. 2004. Modified from Sutherland and Poppy 2005.

80


communities13. Many pharming products are proteins that may persist in the soil for a long time, for example Donegan et al.14 showed the persistence in the soil of a proteinase inhibitor from GM tobacco for 57 days. Figure 3.1 illustrates the steps involved in quantifying exposure. With regard to unintended exposure, a special group of pharming plants deserves mention: oral vaccines produced in fruits and vegetables and intended for direct intake. For example, producing vaccines in bananas is an attractive idea, especially for use in developing countries. The bananas can be consumed directly by both adults and children without any preprocessing. The drawback is that the identity among banana batches with vaccine could be poor and therefore the dose would be difficult to adjust. Therefore, the system has an inherent element of uncertainty of exposure. The exposure (dose) will be difficult to adjust correctly, and the risk of unintended exposure is also high, as the fruit containing the vaccine could easily be mistaken for unmodified fruit. Present experiences with effects from exposure. As yet, no experimental data have been reported on exposure effects from GM pharming plants, but for other categories of GM plants such investigations exist. Most of these studies are on non-target effects from Bt-plants (with toxin-producing transgenes from Bacillus thuringiensis). In a number of laboratory tests, effects of Bt-plants on different insect herbivores and their predators/parasitoids were analyzed. The results differed according to Bt-plant type and insect species studied15, and therefore no general conclusion can be reached on non-target effect of Bt-plants, but the effects have to be evaluated in every risk scenario. Examples of different pharming plants and their supposed environmental and health effects are presented in the appendix at the end of the book. 3.1.2.2 Transgene dispersal

Vertical gene flow and co-existence: Gene transfer from one plant to another through sexual crosses. The dispersal routes of GM plants are no different than those of other plants, and thus the routes will depend on the biology and life cycle of the species in question. The main routes of gene dispersal for a crop plant are shown in figure 3.2. For many types of pharming, dispersal of the inserted pharming genes is considered very problematic – gene flow and exposure to the pharmaceutical products are often closely linked – and therefore the choice of production species is of utmost importance. The inserted genes can be spread through pollen and seeds during: – Intraspecific hybridization: Hybridization through pollen transfer between the crop and plants of the same species in the same field, in neigh13 14 15

Griffiths et al. 2000. Donegan et al. 1997. For a review see Sutherland and Poppy 2005.

81


bouring fields or in feral populations (naturalized populations of crop plants). – Interspecific and intergeneric hybridization: Hybridization through pollen transfer between the crop and wild, weedy or cultivated relatives of related species. – Dispersal of seeds or other propagules through space and time: – to other locations for example through transport and handling; – to subsequent crops/plants in the same location due to spilling.

Farm-saved seed and seeds in feed and manure

Commercially available certified seed

Seeds with machinery Preparation sowing

Seeds in soil

Growing

Pollen from crop

Harvest

Pollen from weeds

Transport

Storage

Sale

Seeds from volunteers and weeds

Figure 3.2: Dispersal routes of a plant pharming crop at different stages in

crop production. Man-made dispersal routes are at the top (light gray); biological dispersal routes are below (dark grey)16.

Factors affecting the likelihood of gene dispersal. The magnitude (probability) of gene flow between plants is determined by a number of isolating mechanisms. When choosing a pharming production platform, these isolating factors should be evaluated17 so that dispersal is minimized or prevented. Table 3.1 lists some of the most important crop plants presently applied as pharming production platforms and reports on the potential for outcrossing in these species. Factors governing gene flow. The reproduction system of the plants. Generally, plants which are outcrossing, or have effective clonal reproduction (vegetative reproduction or agamospermy), will have higher gene dispersal. Perennial plants such as poplar (Populus sp.), fescue (Festuca sp.), rye (Lolium sp.) and meadow grasses (Poa sp.) often have these characteristics. The pollen vehicle – be it wind, insects or both – is also a determinant of 16 17

Modified from Tolstrup et al. 2003. Tolstrup et al. 2003; Jørgensen and Wilkinson 2005; Richards 2005.

Table 3.1:

Main European crops and their biological characteristics influencing gene flow to the environment: reproduction system, centre of diversity, dispersal agents and presence of wild relatives

Sunflower Seed (Helianthus annuus) Sugar beet (Beta vulgaris ssp. vulgaris) Potato (Solanum tuberosum) Alfalfa (Medicago sativa) Tobacco (Solanum nicotiana) Rice (Oryza sativa) Soy bean (Glycine max) Pea (Pisum sativum) Carrot (Daucus carota)

Centre of diversity Middle East

Means of dispersal Self-fertilisation, Pollen (wind dislow frequency of outcrossing persal) and seeds Self-fertilisation, Middle East Pollen (wind) and low frequency of outcrossing seeds Cross-fertilisation, Central Pollen (wind) and low frequency of selfing America seeds Self-fertilisation, some crossMultiple Pollen (insects fertilisation (10–30 % of seeds) origins? and wind) and seeds Cross-fertilisation USA Pollen (insects) and seeds Cultivated types are harvested be- Mediterranean, Pollen (wind and fore bolting. Seed production from Near East insects) and seeds crosses between male sterile and pollen producing lines Vegetative (tubers), crossLatin America Tubers (sensitive fertilization (in low frequency). to frost). Pollen (insects) Cross-fertilization Iran, Anatolia Seed, pollen (insects) Self-fertilization South America Seed, pollen (low levels of outcrossing) (insects) Self-fertilization Asia Seed, tillers (low levels of outcrossing) (pollen (wind)) Self-fertilization Asia Seed, (< 1 % outcrossing) (pollen (wind)) Self-fertilization Pakistan Seed (outcrossing very low) (pollen (insects)) Cross fertilization Afghanistan Pollen and seed

Wild cross-compatible relatives in Europe Triticum species (i.e. wild emmer T. turgidum), Aegilops species (i.e. A. cylindrica) H. vulgare ssp. spontaneum None. (Pollen dispersal between fields very likely) Brassica sp., Raphanus sp. (Sinapis arvensis, Hischfeldia incana). (Pollen dispersal between fields very likely) None. (Pollen dispersal between fields very likely) Wild and weedy beets (i.e. B. vulgaris ssp. maritima) Solanum nigrum and Solanum dulcamara: Generally no outcrossing M. falcata None Weedy red rice (O. sativa), Oryza rufipogon None None Wild carrot (same species as cultivated carrot)


Wheat (Triticum aestivum) Barley (Hordeum vulgare) Maize (Zea mays) Oilseed rape (Brassica napus)

Modes of reproduction

82

Crop


83

gene flow18. Gene dispersal generally occurs over long distance in species with wind dispersal; insects normally disperse pollen over shorter distances (though several km can be normal19), but are more efficient in their gene dispersal, as they specifically target flowers that are ready for pollination20. From a dispersal point of view, it would be wise to choose species for pharmaceutical production that do not reproduce before harvest time. That is, species where vegetative biomass constitutes the harvest and in this respect beet, tobacco and potato could be candidate species. Ramsey21 presents an overview of the many reproductive barriers that affect the extent of gene flow. When the production platform is the seed, the degree of overlap of distribution area and flowering period for pharming plant and potential recipient is critical. If no other crossing barriers are present, close physical contact should enhance gene dispersal. For pharming purposes, as well as some other GM traits, it would then make sense to choose crops that have no known hybridization partners in the cultivation area or choose varieties that have staggered flowering times, for example in an area with predominant cultivation of winter varieties of the crop to select a spring variety for pharmaceutical production. In relation to hybridization with wild relatives, pharming plants with cross-compatible wild relatives in the area may present a higher risk of outcrossing genes. Table 3.1 lists some potential crop recipients that may disperse genes, and their wild relatives with which they hybridize. The distance/isolation between donor and recipient is of importance for pollen dispersal. However, pollen dispersal by wind fits a leptokurtic curve and usually levels off a few meters from the pollen source22 . For co-existence, DEFRA in the UK has recently recommended an isolation distance of only 35 m for oilseed rape, an outcrossing crop with pollen dispersal by both insects and wind 23. This relatively short distance may reduce pollen flow substantially; however, in a landscape study on oilseed rape Rieger and colleagues found that the highest frequencies of pollination had taken place several km from the oilseed rape donor24. This shows that pollen dispersal can be difficult to predict especially when insects are involved, which is the case for oilseed rape being pollinated by both wind and insects. As some of the pollen from a wind-pollinated crop can be brought up into the atmosphere (to a height of approximately 2 km), some long distance pollination is always possible. Insects can also 18 19 20 21 22 23 24

Proctor et al. 1996; Ramsey 2005. Ramsay 2005. Tolstrup et al. 2003. Ramsey 2005. Ramsay 2005. DEFRA 2006. Rieger et al. 2002.

84


disperse the pollen widely in the landscape25, but most bee visits will be between neighbouring plants, limiting the long-range gene dispersal. Also affecting the degree of hybridization are environmental factors such as size and form of fields26 the surrounding landscape (for example hedges and roads) and the pollinators present. The least dispersal will be from small plots of the gene donor (the pharming plant) to large fields of the potential recipient; the large amounts of self-pollen from the recipient will lower the frequency of pollinations from plots outside the field. As an extra safeguard, the outermost border of the recipient could be discarded as alien outcrossing decreases towards the centre of the field 27. In relation to wild relatives and ferals, these rarely form large populations. Therefore, they may be exposed to large amounts of pollen from the donor, especially in wind-pollinated species. Preferably, a pharming species should be chosen that is not cultivated elsewhere in the region and has no relatives in that area. To reduce seed dispersal, crops that have a large seed spillage before or during harvest (for example species as oilseed rape, fescue and meadow grass) should be avoided. Likewise, species that have large seed dispersal with animals, wind and humans should be avoided. If spilled seeds are incorporated into the deeper layer of the soil seed bank after harvest, they may survive for long periods and germinate when tilling brings them up to the germination layer. After harvest it is, therefore, important to allow or initiate the germination of spilled seeds before preparing for the new crop. Generally, the agricultural practices, for example the crop rotation schemes, are very important for control of volunteer plants germinating from spilled seeds. Table 3.2 shows the seed survival of some important crops species. The cross compatibility between crop and recipient. As a rule, the closer the relationships between donor and recipient, the more cross compatible they are, and the more gene flow will occur. It follows that intraspecific gene flow (between plants of the same species) occurs more readily than interspecific or intergeneric gene flow (between different species of the same genus or between species of different genera). In the case of gene flow between different species, genetic barriers may reduce the production of offspring and the offspring often suffer from a reduced survival 28.

25 26 27 28

Hayter and Cresswell 2006. Damgaard and Kjellsson 2005. Tolstrup et al. 2003; Damgaard and Kjellsson 2005. See for example review by Arnold 1997.


85

Table 3.2: Seed survival in the soil of some important crops29. Survival periods

represent averages in undisturbed soil. Seed survival will be longer if the seeds are ploughed in deeply, and survival will be shorter under extensive management of the seed-containing soil layer.

Type of seed bank, survival interval in number of years

Plant species

Temporary survival, normally < 1 year

Oats (Avena sativa), wheat (Triticum aestivum), maize (Zea mays), rye (Secale cereale), onion (Allium cepa)

Short-term seed bank, 1–4 years

Barley (Hordeum vulgare), perennial rye grass (Lolium perenne)

Short- long-term seed bank, 1– >10 years

Italian rye grass (Lolium multiflorum), lucerne (Medicago sativa), parsnip (Pastinaca sativa), carrot (Daucus carota)

Long-term seed bank, 5– >20 years

Oilseed rape (Brassica napus), sugar beet, fodder beet (Beta vulgaris), hop medic (Medicago lupulina), red clover (Trifolium pratense), white clover (Trifolium repens), celeriac (Apium graveolens), potato (true seed) (Solanum tuberosum)

The fitness advantage provided by the transgene or crop genes linked to the transgene. If the pharming gene is advantageous, for example has a repellent or toxic effect to herbivores, there will be a selection for the pharming plant, which in turn might provide it with a better survival in time and space, and thereby increase the population size or perhaps make the transgenic plant more invasive to new habitats. Even though crop-wild hybridization may only occurs in low frequencies30, a moderately advantageous transgene would be expected to spread in the population and environment. After the escape of the transgene, the plant fitness of the transgenic hybrids is the best indicator of allelic spread31. Plant fitness can be evaluated from changes in survival and reproduction, and adaptation to biotic and abiotic stress32. In cases where the trangene is introgressed into a new genetic background, this may alter transgene expression33 and thus plant fitness. Present experiences with transgene dispersal. Many reports have been published on adventitious mixture of transgenes in the harvest of Non-GM crops34. Some of these incidents have received much public attention and 29 30 31 32 33 34

Modified from Tolstrup et al. 2003. Jørgensen and Wilkinson 2005; Ellstrand et al. 1999 and 2003. Snow et al. 1999; Hails and Morley 2005. White 2002. Ammitzbøll et al. 2005. Légère 2005; Beckie et al. 2003; Cerdeira and Duke 2006; Mellon and Rissler 2004.

86


created anxiety. One example is the co-mingling of transgenic seeds with conventional seeds during processing: StarLink was a variety of Bt corn patented by Aventis Crop Sciences. Sale of StarLink seed was allowed for feed, but as there was a possibility that a few persons might develop allergic reactions to the Bt protein, as it is less rapidly degraded than other Bt toxins, StarLink was only to be used for feed. StarLink corn was subsequently found in food destined for human consumption in many places all over the world. This led to an economic smack in the eye for Aventis and a PR disaster for the biotechnology industry as a whole. Another case of co-mingling occurred for the company ProdiGene. ProdiGene suffered substantial losses when GM seeds from maize volunteers with trypsin-producing transgenes were found in the subsequent soya harvest intended for human consumption. ProdiGene was forced to buy up the harvest of around 13,500 tonnes of soya beans, worth two million dollars, and destroy it, and also had to pay a fine. Lately, similar incidents of co-mingling have been reported from Europe. There are a number of examples of non-approved GM varieties found mixed with conventional seed: a herbicide-resistant GM maize line not yet approved for commercialization (LLRICE601, from Bayer) was found in long grain rice for human consumption, and the insect-protected maize Herculex RW (from Pioneer/Dow Agrosciences) was identified in maize imported to EU for feed. Herculex RW is approved in US, but not in EU. Ironically, the majority of the incidents of transgene contamination have been disclosed by NGOs. Most cases have been verified using authorized laboratories for GM testing. Genewatch UK and Greenpeace have created a register of cases of contamination35; the register also gives references to additional information on the cases. Adult, single and multi tolerant transgenic volunteers or ferals from herbicide-tolerant oilseed rape varieties have been reported from North America36 and from Japan37. In Japan these varieties were not cultivated, but the plants derived from seed spillage of imported seed. Gene flow via hybridization is also evident from the finding of a GM herbicide-tolerant hybrid between oilseed rape and the related species B. rapa in western Canada38. There are probably many more cases of spontaneous transgene flow, but as there is little or no monitoring of wild or feral populations, gene flow is rarely detected. In relation to the risks of gene dispersal, APHIS has stated that crops with multiple years of seed dormancy, which are bee-pollinated, and which are cross compatible with weedy or feral types that grow close to fields, are inappropriate for the open production of pharmaceuticals. In the US, where many releases of GM pharming plants have taken place, there is no obligatory monitoring of the wider environment, though the actual site 35 36 37 38

Greenpeace and Gene Watch UK. GM Contamination Register. Simard et al. 2002; Beckie et al. 2003; Yosimura et al. 2006a. Yosimura et al. 2006b. Yosimura et al. 2006a.


87

of the release must be monitored usually two years after cultivation. The specific conditions for individual pharming plant releases in the US can be found on the APHIS homepage39. Modelling and mitigating gene flow. As the adventitious presence of the transgene is dependent on so many biological, climatic and physical factors, the extent to which this adventitious presence will occur is discussed and modelled. To estimate the gene flow from pollen and seeds, landscape models such as GENESYS for oilseed rape40 and MADPOD for maize have been developed for key crops for example within the EU-project SIGMEA (the SIGMEA project is described in the appendix to this chapter). These very complicated models can estimate gene flow from the different admixture sources to a number of fields in a region, taking into consideration the biology and genotype of the crop, the agricultural practice and the regional topography. The modelling can help predict the major sources of adventitious presence of GM, thereby indicating which mitigating measures would be most efficient. However, it is important to remember that the models are simplifications of very complicated interactions in nature, and as such their output is a probabilistic statement. Commonly used management procedures (good farming practice) to minimize adventitious presence of GM plants in Non-GM crops: – establishing isolation distances between fields to minimize pollen flow between fields; – increasing length of cropping intervals to control volunteers that might contaminate the harvest through pollen and seeds; – changing crop types in the rotation to control volunteers, for example control of broadleaved GM species such as soybean and oilseed rape could be eased by cultivating cereals in the next cropping season; – direct control of volunteers and cross compatible weeds in the field to reduce contamination of the harvest from pollen and seeds of volunteers; – changing field size and form as generally large fields are less prone to cross pollination than small fields (with the same pollen source, quadratic fields are less exposed than rectangular fields with an identical area); – establishing buffer zones as catch crops of GM pollen; buffer zones can be sold as GM; – beehives in the field will (for a period depending on the crop) saturate the field with internal bees and limit the arrival of bees and pollination from outside; – testing of purity of certified seed for sowing or seed lots for use in food and feed; 39 40

APHIS Release permits. Colbach et al. 2001a and b.

88


– cleaning of sowing equipment, combines, transport vehicles and storage rooms; – special education of personnel responsible for GM plant growing. The above mentioned mitigation measures are interrelated; for example larger isolation distances will allow for smaller buffer zones, etc. It is difficult, however, to lay down general guidelines for the minimum mitigating measures that are necessary. That will depend on the product being produced, the crop type in question, the environmental setting and the legislation. This lack of generality is also apparent from the two examples on pharming crops presented in the Appendix to the book. Box 2 below gives an overview of different mitigating measures to prevent mixing of GM and Non-GM plant products for white/red biotechnology suggested by authorities and biotech experts in the US and Canada. These measures are largely the same as those suggested by the Union of Concerned ScienBox 3.2:

Measures to avoid co-mingling of GM plants for white and red biotechnology with Non-GM plants proposed by USDA, CFIA and the US Biotech Industry Platform41

– Distinct visual markers of the GM type. – Cultivation in remote areas will minimize cross pollination with other fields (zoning). – Time shift (compared with nearby food or feed crops) in planting will also provide temporal separation. Separation using varieties with differences in flowering or harvest times will reduce cross pollination and co-mingling of harvest products. – Extended isolation distances (e.g. 800–1600 m for normal pollinating maize), fallow zones, using other plants as pollen barriers, removing or covering of inflorescence. Biological confinement or indoor production could also be an option. – Fencing and other restrictions to entry. This will to some extent delimit feeding and dispersal by herbivores. – Dedicated equipment, machinery and processing facilities. – Preliminary on-farm processing to avoid dispersal of seed and propagules outside the farm. – Post-release monitoring. – Standard operating procedures for seeding, transplanting, side-maintenance, harvesting, seed cleaning; storage, drying and processing of biomass; disposal of biomass, for example autoclaving or incineration; handling and cleaning of machinery, equipment and containers; monitoring during growing seasons and post-harvest land use, dealing with non-compliance with terms and conditions for confinement. – Records and reporting of all activities dealing with the cultivation and transport of seeds and plant material, documentation and logs for seeds and biomass. – Training of staff and workers to handle adequately the plant material, both during growing, harvest, transport and possible processing. – Emergency response and/or contingency plans. – Strict control of compliance to measures imposed, either by regulators or by other independent institutions (third-party audits). – Test for GMOs in the raw agricultural commodity. 41

Modified from Spök 2007.


89

tists42 for pharming crops. In addition, the Union of Concerned Scientist suggested disallowing food and feed crops in order to obtain virtually zero contamination. Genetic containment of pharming plants. To a certain extent the dispersal of pharming genes from open fields to the environment can be controlled by “good farming practice”. Physical isolation of pharming fields from other fields can be brought about by distance and border or barrier crops. Besides the different management strategies (isolation distances or barriers between fields etc.) that may isolate the pharming crop from possible recipients in time and space, different genetic containment strategies can “keep the transgene in the pharming plant” by hindering dispersal via pollen or producing sterile plants through genetic changes in the genome of the pharming plant. Table 3.3 summarizes the state of the art of genetic containment. At present, there are three operational types of genetic containment that could limit or restrict gene flow, and more technologies are in the pipe line for example from the EU project TRANSCONTAINER (described in the appendix). The different types of genetic containment could be combined with physical isolation to obtain the right combination, so that containment fits with a particular transgene, recipient plant and environment. Presently the most common and applicable containment strategies are: – Engineering the transgene into the chloroplast genome (the plastid genome). These plants are named transplastomics. This will prevent dispersal with the pollen in the many plant species (the majority of flowering plants) that transmit their plastids exclusively through the seed. – Male sterility provided by nucleases with a pollen tapetum specific promoter. The promoter ensures that the nucleases are only expressed in the tapetum of the pollen sack. The tapetum is degraded by the nuclease activity, allowing no functional pollen to be produced. – Seed sterility. This technology, the GURT technology (Genetic Use Restriction Technology) also called “the terminator technology”, will cause second generation seeds to be sterile. There are two types of GURT technology based on different mechanisms to switch on and off the lethal gene that provides sterility. The technology was under development in the 1990s and is not yet commercially available, because some stakeholders expressed concerns that the terminator technology might prevent the smaller farmers using farm-saved seeds for their next crop. Monsanto, one of the world's biggest seed suppliers, has refrained from commercializing the technology. 42

Union of Concerned Scientists 2004.

90


Table 3.3: An overview of different containment strategies43 Technologies for biological transgene containment Technology

Advantages

Disadvantages

Status

Chloroplast engineering

Prevents gene flow through the pollen. High and stable expression and accumulation of the transgenic product.

Leakage rate very low, but little investigated

Still only available for a number of crops. Systems for control of expression needed (e.g. tissuespecific). Better expression in non-photosynthetic tissues

Male sterility (Ms)

Prevents outcrossing

Cytoplasmic-based systems can be leaky; nuclear systems leaky if Ms-genes are silenced

Cytoplasmic-based systems in most crop species. Nuclear based systems only in few crops

Seed sterility

Can control both outcrossing and volunteers

Leaky if sterility-constructs are silenced or recombined

Terminator technology not on the market due to public opposition

Cleistogamy

Pollination occurs before flowers open, theoretically preventing outcrossing

Probably leaky to some extent

Not ready for use though genes for cleistogamy have been identified

Apomixis

Seeds of vegetative origin. Controls outcrossing. Fixation of hybrid genotypes

Genes for apomixis not yet identified

Spontaneously found only in a limited range of species. Not demonstrated in transgenic crops

Incompatible genomes

Prevents recombination in hybrids formed with related species/types. Stable introgression in these species is prevented

Targeted integration on incompatible genomes or nonhomologous part of the genome is not yet straight forward

Will only be applicable for some cross combinations between crop and relative

Inducible promotors to ensure temporal and tissuespecific expression

Transgene activated only at the time or in the tissue when and where expression is needed

Inducibility may be unstable resulting in leakiness

Not yet demonstrated in transgenic crops

Transgene mitigation

Mitigation traits (linked to primary transgene) advantageous to crops but disadvantageous to weeds (e.g. dwarfism)

Does not address gene flow between fields, and weedy/ wild populations can be endangered

Demonstrated in oilseed rape and tobacco

43

Modified from Daniell 2002 and Murphy 2007.


91

For some crop-wild species complexes – where there is only partial homology between the genomes of the crop and the wild relative – it has been speculated that targeted insertion of the transgene in the genome parts only found in the crop, and not in the wild species, would limit the rate of transfer in hybridizing plants. This strategy could be applied to crops like oilseed rape44 or wheat45. However, it needs to be proved to what extent this kind of strategy will limit gene transfer. Genetic and physical containment and confinement of the pharming plants are possible. However, no kind of containment or confinement strategy is water-proof46. Gene stability and expression of containment genes can break down, and therefore within a certain pharming plant a number of different gene flow reducing strategies should work in concert. 3.1.2.3 Horizontal gene flow

Non-sexual transfer of genes between organisms by way of vectors able to insert DNA. Horizontal gene dispersal has not been studied as extensively as the vertical gene transfer, but this study area should perhaps be allocated resources in order to clarify if (and to what extent) horizontal gene flow takes place. If transfer of plant genes to microorganisms takes place, such horizontal gene transfer could raise concerns in relation to dispersal of pharming genes. Horizontal gene transfer has been shown to take place from plant to microorganism in the laboratory47, and there are also studies that suggest horizontal gene transfer of mitochondrial genes and mobile elements between higher plants mediated by microorganisms48. Although, cases of horizontal gene transfer occur, the frequency seems extremely low49. Horizontal gene transfers could be possible, but each of the many steps involved, from the release of intact DNA from a plant cell to integration into a prokaryotic genome (and perhaps even onwards to another plant cell), has such a low probability that a successful transfer event would seem to be extremely rare50. However, once a gene has been transferred from GM plant to bacteria, the gene is easily transferred to other bacteria. Even though horizontal gene transfer seems unlikely based on the data available today, a precautionary approach could be applied in the case of some pharming plants. In many cases, pharming plants will only be cultivated on limited areas, making special treatment of the field possible; for example tilling after harvest could involve procedures as steaming/heating for disinfecting recombinant microorganisms carrying the pharming gene. If 44 45 46 47 48 49 50

Mikkelsen et al. 1996; Metz et al. 1997; Tomiuk et al. 2000. Schoenenberger et al. 2005. Ellstrand 2003. Tepfer et al. 2003. E.g. Diao et al. 2006; Bergthorsson et al. 2003. Bergthorsson et al. 2003. de Vries and Wackernagel 2004.

92


the vector used for transgene transfer is integrated in the transformed plant, and if the vector shows homology to DNA of naturally-occurring microorganisms, the risk of horizontal gene transfer will increase due to the chance of recombination between homologous sequences.

3.1.3 The environmental risks – will pharming plants differ from the current GM plants? What are the risks to to the environment and to human health from pharming plants? Basically, the types of risks from pharming plants to the environment will be no different from those of first generation GMPs (that is GM herbicide- and insect-resistant crops) – for example biodiversity, human health and soil fertility can be affected in both cases. Unless totally contained, genes will flow regardless of whether the genes encode a pharmaceutical, a Bt toxin or a native trait. In all plant pharming cases the risk will vary, according to the host plant, the inserted genes and the environment where production will take place. One can imagine that the hazards from the production of poultry vaccine against coccidiosis in oilseed rape and the production of insulin in safflower51 will have different effects on the environment. Generally, the environmental effects from pharming plants will be more difficult to picture than risks from first generation GM crops that have traits similar to crops bred by traditional methods. Also, the climatic changes that are predicted for the future may make the risk of GM plants more unpredictable. Environmental effects of GM plants are normally only tested in a few environments, but if the environment changes the effects may change too. The pharming crops will probably be bred to have an optimized yield, and they may be derived of natural (endogenous) toxic or undesirable metabolites that might jeopardize the quality of the pharming product. Human and environmental exposure risks could therefore be increased compared to more traditional GM crops. Especially when the transgene integration takes place in the plastids, production can be substantial52. For example, a peptide (2L21), which confers protection to dogs against canine parvovirus (CPV), was expressed in tobacco chloroplasts as a translational fusion with the cholera toxin B subunit (CTB)53. In mature tobacco plants a maximum of 7.49 mg/g fresh weight of CTB-2L21 protein was produced. This is equivalent to 3 % of the total soluble protein. This would constitute a 700-fold increase in transgene products compared to first generation GM crops54. With high concentrations of pharming products, which have never been produced in the agro-ecosystem, the likelihood of unintended effects might be higher. Such unintended non-target effects are already the most uncer51 52 53 54

Fox 2006. Daniell et al. 2005. Molina et al. 2004. Spök 2007.

3.2 Environmental risks of animal pharming

93

tain parameter of first generation GM crops, and the new genes and corresponding gene products of pharming plants will increase the likelihood of hazardous effects to the environment. Uncertainty can also increase because it is quite likely that the pharming plants will house other gene modifications stacked onto the pharming genes. Genes providing disease and pest resistance could be inserted in order to increase the quality of the pharming crop. Also, some pharming plants will be engineered with different types of traits that confer genetic containment, in order to reduce the dispersal of inserted genes to the environment.

3.1.4 Concluding remarks When discussing potential risks of pharming plants, it is important to remember that not all pharming plants may present risks to the environments – that will depend on the product and the environmental exposure. Therefore, it seems logical to adopt a case by case and step by step approach, which is already one of the main principles for the statutory risk assessment55. Many aspects of GM plant pharming will be controversial to the public; public perception is often linked to the usefulness of the product (see chapter 5). Risks to the environment could be extensive, and the risk assessment could be quite uncertain. This uncertainty calls for a broad regulation of GM pharming plants; ethical and social aspects of the production should accompany the traditional risk assessment based on natural science.

3.2 Environmental risks of animal pharming If pharming animals were to escape into the environment, further consequences depend on a number of factors. The first factor to consider is whether they would pose a risk of immediate harm. This would be the case if, for example, they had an infectious disease. In addition, should they be eaten by humans or wildlife, their ingestion could be detrimental. This is unlikely, and toxicity would depend on the bioactivity of the protein that they express, as well as on expression sites and levels56. Escapes of large mammals can probably largely be prevented by using relatively simple physical containment (for example double fencing). Escapes could be almost completely ruled out in housing with barrier facilities, although there will always be some risk of escape due to, for example, criminal acts. Small mammals, insects and fish may escape more easily and be irretrievable. If pharming animals should escape, an important question is the degree to which they can survive long-term and reproduce in a natural environment without human protection. For example, with domestic sheep and 55 56

Directive 2001/18/EC. Bruggemann 1993.

94


cattle it would be very unlikely that they disappeared, survived and reproduced. The same is true of domestic chickens and laboratory rabbits, who may be prone to disease and predation, and whose phenotype is probably not competitive with conspecific wild types and other competitors. However, the answer to this question may to some extent depend on the environment into which the animals escape57. The FDA recommends that in order to lessen the chance of inadvertent breeding into a nontransgenic population, transgenic animals should be neutered58. If escaped animals do survive, they may become pests or otherwise disturb the environment, possibly mixing with wild conspecifics genetically. As with short-term escapes, their potential for harm would also depend on the bioactivity of the protein that they express. However, their transgene expression (and that of their offspring) may be unstable. Such instability may, for example, be caused by different genetic backgrounds in the offspring or in response to the altered environmental stimuli, and it may result in unexpected phenotypes. Problems could also be caused if the escaped animals have further transgenic properties. For example, if they are disease resistant, they may become reservoirs of disease. Little is known about whether horizontal gene transfer takes place from animals to other organisms, which might be another source of spread of the transgene. Therefore, pharming animals kept for production should not be released into the environment, and the accidental entry of other animals into their facilities should be prevented.

57 58

Buehr and Hjorth 1994. FDA 1995.

3.3 References

95

3.3 References Ammnitzbøll HA, Mikkelsen T, Jørgensen RB (2005) Environmental effects of transgene expression on hybrid fitness – a case study on oilseed rape. Environmental Biosafety Research 4:3–12 APHIS Release permits http://www.aphis.usda.gov/brs/ph_permits.html (July 2008) Arnold M (1997) Natural Hybridization and Evolution. Oxford University Press Beckie HJ, Warwick SI, Nair H,, Séguin-Swartz G (2003) Gene flow in commercial fields of herbicide-resistant canola (Brassica napus). Ecological Applications 13:1276–1294 Bergthorsson U, Adams KL, Thomason B, Palmer JD (2003) Widespread horizontal transfer of mitochondrial genes in flowering plants. Nature 424:197–201 Boothe JG, Saponja JA, Parmenter DL (1997) Molecular farming in plants: Oilseeds as vehicles for the production of pharmaceutical proteins. Drug Development Research 42:172–181 Bruggemann EP (1993) Environmental safety issues for genetically modified animals. J Anim Sci 71:47–50 Buehr M, Hjorth JP (1994) Genetically modified animals. Perspectives in development and use. Miljoeprojekt nr. 277, Ministry of the Environment and Energy, Danish Environmental Protection Agency, Denmark Cerdeira AL, Duke SO (2006) The current status and environmental impacts of glyphosate-resistant crops: A review. Journal of Environmental Quality 35:1633–1658 Colbach N, Clermont-Dauphin C, Meynard JM (2001a) GENESYS: a model of the influence of cropping system on gene escape from herbicide tolerant rapeseed crops to rape volunteers – I. Temporal evolution of a population of rapeseed volunteers in a field. Agriculture Ecosystems and Environment 83:235–253 Colbach N, Clermont-Dauphin C, Meynard JM (2001b) GENESYS: a model of the influence of cropping system on gene escape from herbicide tolerant rapeseed crops to rape volunteers – II. Genetic exchanges among volunteer and cropped populations in a small region. Agriculture Ecosystems and Environment 83:255–270 Commandeur U, Twyman RM, Fischer R (2003) The biosafety of molecular farming in plants. Agbiotechnet 5:1–9 Council Directive 90/220/EEC. 12 March 2001 Damgaard C, Kjellsson G (2005) Gene flow of oilseed rape (Brassica napus) according to isolation distance and buffer zone. Agriculture Ecosystems and Environment 108:291–301 Daniell H (2002) Molecular strategies for gene containment in transgenic crops. Nature Biotechnology 20:581–586 Daniell H, Kumar S, Dufourmantel N (2005) Breakthrough in chloroplast genetic engineering of agronomically important crops. Trends in Biotechnology 23:238–245 de Vries J, Wackernagel W (2004) Microbial horizontal gene transfer and the DNA release from transgenic crop plants. Plant and Soil 266:91–104 DEFRA (2006) Consultation on proposals for managing the coexistence of GM, conventional and organic crops http://www.defra.gov.uk/environment/gm/crops/pdf/gmcoexist-condoc.pdf (July 2008) Dhaenens L, Szczebara F, Husson MO (1997) Identification, characterization, and immunogenicity of the lactoferrin-binding protein from Helicobacter pylori. Infection and Immunity 65:514–518

96


Diao XM, Freeling M, Lisch D (2006) Horizontal transfer of a plant transposon. PLoS Biology 4:119–128 Directive 2001/18/EC of the European Parliament and the Council on the deliberate release into the environment of genetically modified organisms and repealing http://eur-lex.europa.eu/LexUriServ/site/en/oj/2001/l_106/ l_10620010417en00010038.pdf (July 2008) Donegan KK, Seidler RJ, Fieland VJ, Schaller DL, Palm CJ, Ganio LM, Cardwell DM, Steinberger Y (1997) Decomposing of genetically engineered tobacco under field conditions: persistence of the protein inhibitor I product and effects on soil microbial respiration and protozoa, nematode and microarthropod populations. Journal of Applied Ecology 34:767–777 EFSA (2004) Guidance document of the GMO Panel for the risk assessment of genetically modified plants and derived food and feed. 28 April 2004, revised in 2006 http://www.efsa.europa.eu/en/science/gmo/gmo_guidance/660.html (July 2008) Ellstrand N (2003) Dangerous Liaisons? When Cultivated Plants Mate with Their Wild Relatives. The Johns Hopkins University Press, Baltimore Ellstrand NC, Prentice HC, Hancock JF (1999) Gene flow and introgression from domesticated plants into their wild relatives. Annual Review of Ecology and Systematics 30:539–563 Ellstrand NC (2003) Going to “Great Lengths” to prevent the escape of genes that produce specialty chemicals. Plant Physiology 132:1770–1774 FDA (1995) Points to Consider in the Manufacture and Testing of Therapeutic Products for Human Use Derived from Transgenic Animals. Food and Drug Administration, Center for Biologics Evaluation and Research, August 1995 http://www.fda.gov/CBER/gdlns/ptc_tga.txt (June 2008) Fischer R, Schillberg S, Emans N (2001) Molecular farming of medicines: a field of growing promise. Outlook on Agriculture 30:31–36 Fox JL (2006) Turning plants into protein factories. Nature Biotechnology 24:1191–1193 Greenpeace and Gene Watch UK. GM Contamination Register http://www.gmcontaminationregister.org/ (July 2008) Griffiths BS, Geoghegan IE, Robertson WM (2000) Testing genetically engineered potato, producing the lectins GNA and Con A, on non-target soil organisms and processes. Journal of Applied Ecology 37:159–170 Hails RS, Morley K (2005) Genes invading new populations: a risk assessment perspective Trends in Ecology and Evolution 20:24–252 Hayter KE, Cresswell JE (2006) The influence of pollinator abundance on the dynamics and efficiency of pollination in agricultural Brassica napus: implications for landscape-scale gene dispersal. Journal of Applied Ecology 43:1196–1202 Humphrey BD, Huang N, Klasing KC (2002) Rice expressing lactoferrin and lysozyme has antibiotic-like properties when fed to chicks. Journal of Nutrition 132:1214–1218 Joint Research Centre, Ispra. Explanatory document on the use of “Percentage of GM-DNA copy number in relation to target taxon specific DNA copy numbers calculated in terms of haploid genomes” as a general unit to express the percentage of genes http://engl.jrc.it/docs/HGE %20release %20version %201.pdf (July 2008) Jørgensen RB, Wilkinson MJ (2005) Rare hybrids and methods for their detection. In: Guy M. Poppy GM and Wilkinson JM (eds) Gene Flow from GM Plants. Blackwell Publishing, pp 113–142 Kirk DD, McIntosh K, Walmsley AM, Peterson RKD (2005) Risk analysis for plantmade vaccines. Transgenic Research 14:449–462

3.3 References

97

Légère A (2005) Risks and consequences of gene flow from herbicide-resistant crops: canola (Brassica napus L) as a case study. Pest Management Science 61:292–300 Mellon M, Rissler J (2004) Gone to seed http://www.ucsusa.org/assets/documents/food_and_environment/seedreport_ fullreport.pdf (July 2008) Metz PLJ, Jacobsen E, Nap JP, Pereira A, Stiekema WJ (1997) The impact on biosafety of the phosphinothricin-tolerance transgene in inter-specific B-rapa x B-napus hybrids and their successive backcrosses. Theoretical and Applied Genetics 95:442–450 Mikkelsen TR, Jensen J, Jørgensen RB (1996) Inheritance of oilseed rape (Brassica napus) RAPD markers in a backcross progeny with Brassica campestris. Theoretical and Applied Genetics 92:492–497 Molina A, Hervas-Stubbs S, Daniell H, Mingo-Castel AM, Veramendi J (2004) High-yield expression of a viral peptide animal vaccine in transgenic tobacco chloroplasts. Plant Biotechnology Journal 2:141–153 Murphy DJ (2007) Improving containment strategies in biopharming. Plant Biotechnology Journal 5:555–569 Nash JP, Kime DE, Van der Ven LTM, Wester PW, Brion F, Maack G, StahlschmidtAllner P, Tyler CR (2004) Long-term exposure to environmental concentrations of the pharmaceutical ethynylestradiol causes reproductive failure in fish. Environmental Health Perspectives 112:1725–1733 Proctor M, Yeo P, Lark A (1996) The natural history of pollination. HarperCollins, London Ramsey G (2005) Pollen dispersal vectored by wind or insects. In: Poppy GM, Wilkinson JM (eds) Gene Flow from GM Plants. Blackwell Publishing, pp 43–77 Regulation 1829/2003 of the European Parliament and of the Counsil, on genetically modified food and feed. 22 September 2003 http://eur-lex.europa.eu/LexUriServ/site/en/oj/2003/l_268/ l_26820031018en00010023.pdf (July 2008) Regulation 1830/2003 of the European Parliament and of the Counsil, concerning the traceability and labelling of genetically modified organisms and the traceability of food and feed products produced from genetically modified organisms and amending Directive 2001/18/EC. 22 September 2003 http://eur-lex.europa.eu/LexUriServ/site/en/oj/2003/l_268/ l_26820031018en00240028.pdf (July 2008) Richards AJ (2005) Hybridization – reproductive barriers to gene flow. In: Guy M, Poppy GM, Wilkinson JM (eds) Gene Flow from GM Plants. Blackwell Publishing, pp 78–112 Rieger MA, Lamond M, Preston C, Powles SB, Roush RT (2002) Pollen-mediated movement of herbicide resistance between commercial canola fields. Science 296:2386–2388 Simard MJ, Légère A, Pageau D, Lajeunesse J, Warwick S (2002) The frequency and persistence of volunteer canola (Brassica napus) in Quebec cropping systems. Weed Technology 16:433–439 Schoenenberger N, Felber F, Savova-Bianchi D, Guadagnuolo R (2005) Introgression of wheat DNA markers from A, B and D genomes in early generation progeny of Aegilops cylindrica Host x Triticum aestivum L. hybrids. Theoretical and Applied Genetics 111:1338–1346 Snow AA, Andersen B, Jorgensen RB (1999) Costs of transgenic herbicide resistance introgressed from Brassica napus into weedy B-rapa. Molecular Ecology 8:605–615

98


Sparrow PAC, Irwin JA, Dale PJ, Twyman RM, Ma JKC (2007) Pharma-Planta: Road testing the developing regulatory guidelines for plant-made pharmaceuticals. Transgenic Research 16:147–161 Spök A (2006) From Farming to Pharming. Risks and policy challenges of third generation GM crops. Institute of Techolog Assessment http://epub.oeaw.ac.at/ita/ita-manuscript/ita_06_06.pdf (July 2008) Spök A (2007) Molecular farming on the rise – GMO regulators still walking a tightrope. Trends in Biotechnology 25:75–82 Sutherland JP, Poppy GM (2005) Quantifying exposure. In: Poppy GM, Wilkinson JM (eds) Gene Flow from GM Plants. Blackwell Publishing, pp 186–212 Tepfer D, Garcia-Gonzales R, Mansouri H, Seruga M, Message B, Leach F, Perica MC (2003) Homology-dependent DNA transfer from plants to a soil bacterium under laboratory conditions: implications in evolution and horizontal gene transfer. Transgenic Research 12:425–437 Tolstrup K, Andersen SB, Boelt B, Buss M, Gylling M, Holm PB, Kjellsson G, Pedersen S, Østergård H, Mikkelsen SA (2003) Report from the Danish working group on the co-existence of genetically modified crops with conventional and organic crops. DIAS Report no. 94, Frederiksberg Bogtryk http://pure.agrsci.dk:8080/fbspretrieve/504564/DIAS_report__Plant_ Production_94 (July 2008) Tomiuk J, Hauser TP, Bagger Jorgensen R (2000) A- or C-chromosomes, does it matter for the transfer of transgenes from Brassica napus. Theoretical and Applied Genetics 100:750–754 Union of Concerned Scientists (2004) A growing concern http://www.ucsusa.org/food_and_environment/genetic_engineering/ pharmaceutical-and-industrial-crops-a-growing-concern.html (July 2008) Vogel L, Geluk F, Jansen H, Dankert J, van Alphen L (1997) Human lactoferrin receptor activity in non-encapsulated Haemophilus influenzae. Fems Microbiology Letters 156:165–170 Weinberg ED (1999) The role of iron in protozoan and fungal infectious diseases. Journal of Eukaryotic Microbiology 46:231–238 Weintraub JA, Hilton JF, White JM, Hoover CI, Wycoff KL, Yu L, Larrich JW, Featherstone JDB (2005) Clinical Trial of a Plant-Derived Antibody on Recolonization of Mutans Streptococci. Caries Research 39:241–250 White JL (2002) U.S. Regulatory Oversight for the Safe Development and Commercialization of Plant Biotechnology. In: Ecological and Agronomic Consequences of Gene Flow from Transgenic Crops to Wild Relatives http://www.biosci.ohio-state.edu/~asnowlab/Proceedings.pdf (July 2008) Yoshimura Y, Beckie HJ, Yasuda K (2006a) Transgenic oilseed rape along transportation routes and port of Vancouver in western Canada. Environmental Biosafety Research 5:67–75 Yoshimura Y, Matsuo K, Yasuda K (2006b) Gene flow from GM glyphosate-tolerant to conventional soybeans under field conditions in Japan. Environmental Biosafety Research 5:169–173

3.3 References

99

Appendix: Status of the risk assessment and co-existence research funded by

the EU

Presently, there is almost no EU-funded research on risk assessment, but there are projects on co-existence and on genetic confinement methods. SIGMEA (www.sigmea.go.dyndns.org/ (March 2008)) The overall objective of SIGMEA was to set up a science-based framework, strategies, methods and tools for assessing the ecological and economical impacts of GM crops and for an effective management of their development within European cropping systems, i.e. to create a practical toolbox. To achieve this overall objective, SIGMEA studied gene flow and related subjects in a wide number of countries across Europe. Crops under study were oilseed rape, beet and maize. SIGMEA started on 3 May 2004 and ran for three and a half years (ended November 2007). CO-EXTRA (www.coextra.org/ (March 2008)) studies and validates biological containment methods and models supply chain organisations and provides practical tools and methods for implementing co-existence. In parallel, CO-EXTRA designs and integrates GMO detection tools, develops sampling plans, and elaborates new techniques to meet the challenges raised by increased demands for cost-effective multiplex methods to detect as yet unapproved or unexamined GMOs (e.g., with stacked genes). All of the methods and tools that are studied and developed are assessed not only from the technical point of view, but also with regard to economic and legal aspects. CO-EXTRA started in 2005 and will run for four years. TRANSCONTAINER (http://www.transcontainer.wur.nl/uk (March 2008)) The project started in May 2006 and will run for three years; aims at developing efficient and stable biological containment systems for genetically modified plants. More specifically the objectives are to promote co-existence of GM and non-GM (including organic) agriculture in Europe by using stable, environmentally safe and commercially viable biological containment strategies in crops economically relevant for Europe, and improve and simplify rules for co-existence. Model species of the project are oilseed rape, grasses, sugar beet, birch and poplar. There are work packages dealing with chloroplast transformation, controllable flowering, controllable fertility and technology impact.

4 The welfare of pharming animals

4.1 Introduction This chapter addresses the welfare of animals used for pharming. Public concern is well documented with regard to animals and biotechnology generally, and in the context of this study also with regard to animals genetically engineered to produce pharmaceutical proteins (see chapter 5). Also, potential animal suffering is one of the major ethical concerns in animal pharming (see chapter 6). Many of the animal welfare concerns that arise in pharming are similar to those in conventional animal husbandry for production or experiments, for example: are the animals housed, fed, bred and handled in ways that suit their species-specific requirements, are they protected from disease and injury, are they subjected to painful or frightening procedures? Further concerns are typical for transgenesis: the generation of a transgenic animal involves reproductive and gene technological procedures that can disturb its physiology, especially its development. In pharming specifically, the expression of a medicinal protein may interfere with the animal’s physiology. Long-term consequences of the genetic intervention are also possible. In animal pharming, efforts are made to rule out problematic effects of transgenesis in the experimental and development phases, i.e. in the course of establishing a production line of animals expressing the desired medicinal protein (see section 2.3). However, negative effects on animal welfare may remain undetected, or they may be tolerated in trade-offs with the desired production. Therefore, although it is not a goal of animal pharming to generate animals with compromised health (as would be the case, for example, in transgenic animal disease models), it is clear that animal pharming is a type of animal utilisation in which health problems and harm to the animals may occur. The animals involved in pharming can be classified as follows: – In the production phase, there are transgenic animals for breeding and for production, and some that cannot be used for either (for example males if the recombinant protein is expressed in milk, or individuals that poorly express the recombinant protein). The latter will in most cases be killed.

102


– In the experimental phase, classes of animals used or generated may include egg cell donors, foster mothers, transgenic founder animals, further generations of transgenic animals for evaluation, and animals that cannot be used (for example those that are ill or poorly expressing the transgene), see section 2.31. In European animal protection legislation, generally speaking, animals may be harmed (only) if this serves a sufficient good (see chapter 8). Underlying these laws and guidelines is an ethical stance that in principle prohibits harming animals, but that allows a certain (also harmful) instrumentalisation of animals, provided it benefits humans (see chapter 6). The evaluation of what is acceptable thus depends on the analysis of benefits and costs to the industry or the public, and of good and adverse effects on the animals (benefits and costs to the animals). This kind of cost-benefit analysis is also part of research proposals that involve utilisation of animals. Many regulatory institutions at EU and national levels have scientific animal health and welfare advisory committees. In order to facilitate costbenefit analyses and moral and legal evaluations in a variety of contexts of animal utilisation, the scientific advisory committees provide answers to empirical questions on the basis of the current state of knowledge. In this chapter empirical animal welfare questions of pharming will be addressed to inform our evaluation of animal welfare concerns in pharming (see chapter 6): – What is the potential harm to the different classes of animals that are utilised in the pharming experimental and production phases? – What are the sources and probabilities of harm? – What measures can be taken to avoid harm?

4.2 Animal welfare risks In the animal pharming experimental phase, as part of their research licence applications (see chapter 8), scientist have to make predictions about the expected impact of their research on the health and welfare of animals used or generated in the research, and to argue what benefits are expected from it. Typically, local ethical committees or authorities evaluate the quality of the research proposal and the animal welfare risk assessment and conduct a cost-benefit analysis, and as a consequence grant permission or not. The information available on potential animal harm should be included in decisions on methodology. If there is a choice between different methods, the argument that one method is likely to inflict less harm on the animals than 1

There will also be use of laboratory animals in the pre-clinical stages of testing the pharmaceutical derived from the desired protein. This aspect is not included here.

4.2 Animal welfare risks

103

the other can be a reason for refusal of authorisation of the more harmful option. The predictions about the expected impact of the research on animal health and welfare can be made on the basis of prior investigations and experience with similar procedures. How much information actually is available influences the precision of these predictions – in the case of routine procedures, these may be based on firm knowledge, whereas with novel procedures, the predictions may amount to educated guesses. Genetic engineering in the past has sometimes had unforeseen negative consequences for animal welfare. The most well-known example is maybe the so-called “Beltsville pigs” which expressed additional growth hormones and had serious health problems2. Being a leading-edge technology, genetic engineering involves a high degree of novelty and, in many cases, very little is known about the longterm impact of the suggested procedures. Novelty and variability lead to higher degrees of uncertainty and can result in unforeseen effects. This is why the Canadian Council on Animal Care applies a “high animal welfare risk” mark to gene technological procedures, initially classifying all experiments involving the creation of novel transgenics as the second most severe on a five-level scale, comparable to major surgery. Further, if approval is merited, it should be provisional, limited to a 12-month period, and subject to the requirement that the investigator reports back to the ACC [Animal Care Committee] as soon as feasible on the animals’ phenotype, noting particularly any evidence of pain or distress.3

While the novelty predisposes to unforeseen effects, risk assessment may nevertheless help to avoid many problems. As Sandøe and colleagues4 have pointed out, some of the health problems experienced by the Beltsville pigs could probably have been foreseen in a careful risk assessment prior to the genetic engineering. Genetic engineering involves not only a high degree of novelty, but also a high degree of variability, for example due to random gene insertion, where each transgenic founder animal is different. Random transgene insertion can alter the expression level or even disrupt an endogenous gene. Therefore, elaborate testing of the animals’ geno- and phenotypes is part of the research routine (see section 2.3). To some extent such testing also involves animal welfare parameters (i.e. at present, mainly health parameters). Extrapolation from such (and more general) knowledge to future studies amounts to assessment of animal health and welfare risks. There are at present no standardised procedures for animal welfare risk assessment, but the Animal Health and Animal Welfare panel of the European Food Safety 2 3 4

Pursel et al. 1989. CCAC 1997 section b.iv. Sandøe et al. 1997.

104


Authority is in the process of developing guidelines to this effect5. The uses of animals in pharming include well-known risk factors (for example management and housing of laboratory or production animals) and others that involve some uncertainty (for example in vitro reproduction techniques) or large degrees of uncertainty (for example transgene expression). Many animal procedures in pharming are not pharming-specific – they may be part of conventional food animal husbandry or of laboratory animal science. However, neither the animal welfare risk assessment required for permission to conduct research with animals, nor ethical evaluation generally, is restricted to potential hazards that are specific to the novel procedures involved. Rather, they address all potential harm to animals in the proposed research. This means that if welfare is compromised by aspects of housing, handling or reproduction procedures that are not specific to pharming, risk assessments and ethical evaluations still need to include these aspects. In this chapter, we try to briefly address several relevant aspects, but focus on those that are most specific to pharming.

4.3 The concept and assessment of animal welfare As a topic for empirical investigation animal welfare is within the scope of veterinary science, animal husbandry, animal behaviour science, and related disciplines. The assessment of animal welfare has in the past decades become a scientific field of its own (as reflected in an accepted theoretical framework, common methodological approaches, data, institutionalisation, and an improvement of the match between theories and empirical data). However, many theoretical and methodological issues are still controversial. Definitions of animal welfare and ways in which animal welfare assessment is approached depend on assumptions regarding what constitutes good and bad lives for animals6. Here, we focus on two aspects: – aspects of their lives that probably matter to sentient animals (and we assume that the animals involved in pharming are sentient), for example freedom from pain and fear; – aspects of the animals' ability to cope with their lives. Although death is typically not included in scientific definitions of welfare or health (a dead individual cannot be ill or suffer, death of foetuses and killing or euthanising of “surplus” or “excess” animals are included in this overview. Various aspects of animal welfare are intricately linked together. For example, a pig exhibiting a behavioural anomaly such as bar-biting (stereotypically mouthing and biting the pen’s bar) has a psychological disorder 5 6

EFSA 2006; EFSA 2007; Müller-Graf et al. 2007. Dawkins 1980; Broom 1991; Duncan 1993.

4.4 Animal welfare considerations in the animal pharming production phase

105

which has most likely been caused by negative feelings: boredom and frustration with a barren environment. Negative feelings can lead not only to psychological disorders, but also to physical health and production problems such as poor immune system functioning or decreased fertility. The causal relationship in the other direction is obvious: disease often leads to pain and suffering (although it does not necessarily). It is not within the scope of this text to provide an overview of indicators of welfare; a large body of literature is available on this topic7. However, some examples are given below. The welfare of an animal can be examined by systematically looking for symptoms of disease, injuries, stress and aversive states, and by conducting diagnostic tests. Overt signs such as morbidity, death, malformations, lack of reproductive ability, feeding behaviour/ stomach fill, body weight patterns, coughs and altered mucus membranes, or lameness, may be easily spotted. Important welfare indicators may also be based on subtly altered behaviour or physiology. For example, scratching of the body can be a sign of itching and parasitic skin lesions or scrapie in sheep. Species-specific adaptive responses need to be taken into account: for example, prey species such as sheep can be very quiet when experiencing acute pain, whereas a pig would squeal. Stress, which is often reflected in abnormal behaviour (for example spontaneous defecation, shortened lying and feeding intervals, stereotypies, aggression, self-mutilation, apathy) is sometimes confirmed with physiological measures. Physiological indicators of short-term stress in vertebrates typically aim at quantifying the activity of the sympathetic nervous system, which is active initially in a stress response through the release of adrenaline, heart rate increases, and associated physiological changes. Medium- and long-term stress is investigated in terms of activity and functionality of the hypothalamic-pituitaryadrenal axis (“stress axis”), i.e. levels of hormones, notably glucocorticoids. Increasingly, the tonus of the parasympathetic system (as reflected in the variability of blood pressure and heart rate) is also being used as a noninvasive welfare indicator. Besides examination of the physiological stress system, it is also common to carry out behaviour tests (for example to identify fearfulness, or to quantify sensorimotor functioning) and preference/ aversion tests, where the animals’ own choices shed light on their needs.

4.4 Animal welfare considerations in the animal pharming production phase In the production phase, only the welfare considerations arising directly from the altered genotype and the expression and harvesting of the recombinant protein can be said to be specific to animal pharming. Ideally, poten7

E.g. Broom and Johnson 1993; Moberg and Mench 2000; Duncan 2005.

106


tial problems related to altered protein expression are detected in the research and development phase and lines with known problems thus need not enter the production phase. In this section we briefly sketch out some general considerations, before addressing the aspects specific to animal pharming.

4.4.1 Housing and management As detailed in section 2.3, the species most relevant to animal pharming are cattle, goats, sheep, pigs, rabbits and chicken. Each of these species is highly domesticated, and there are breeds available that have been bred for production traits (for example milk yield, protein content in milk, reproduction rate, docility) for a very long time. Each of these species has also been subject to close scrutiny with regard to their species-specific behaviour and needs in production environments. The housing and management requirements of livestock have been described in detail elsewhere8, and are to some extent required in current legislation (see chapter 8). Only a few general points will be made here to sum up the situation for readers not familiar with the topic. Domestic animals are genotypically and phenotypically different from their wild progenitors in many ways. Nevertheless, as was first shown by Stolba and Wood-Gush for pigs9, our domestic species will – if given the opportunity – carry out many of the behaviours that occur in their wild relatives. For example, in pigs this includes building nests for their offspring and rooting for food, and in chickens it includes dust-bathing and elevated perching. Preference tests have been used to show that being able to carry out some behaviours is highly important to the animals; they have behavioural needs10. From a welfare point of view it is important to take species-specific (and gender- or age-specific) behavioural needs into account11. It is also crucial to pay attention to specific needs that may arise in some breeds or transgenic lines. For example, cattle of some breeds can be milked easily, while for others it is highly stressful and requires the presence of calves. Some breeds of sheep will suffer more than others when kept indoors, and some breeds of goats are more aggressive than others and therefore need more space and hiding possibilities when group-housed. Each of the species used in pharming is social, and individuals suffer when kept in social isolation, which often results in detrimental effects on the immune and reproductive systems. Species and groups (breeds, age 8 9 10 11

E.g. Fraser and Broom 1997; Kaliste 2004; Perry 2004. Stolba and Wood-Gush 1984, 1989. Duncan 2005. Behavioural needs are therefore also mentioned in animal protection legislation, see chapter 8. In reality, the behavioural needs of production animals in conventional animal husbandry are typically met only to a very limited extent.

4.4 Animal welfare considerations in the animal pharming production phase

107

groups, and sexes) differ in their group housing requirements with regard to group sizes and composition. Frequent re-grouping is stressful to the animals because they recognise each other (or at least their hierarchical orders or relative social standings, depending on the species) and form stable social structures within groups. The animals naturally search for food and have a need to carry out their species-specific food-searching and food-manipulating behaviours. At a minimum, for cattle, goats and sheep, this involves walking and grazing, for pigs walking and rooting, for chickens pecking and using their legs and wings, and so forth. When deprived of such opportunities, the animals can develop behavioural abnormalities such as stereotypies (for example barbiting, feather-pecking). Other natural behaviours that the animals need to carry out include, for example, nest building and maintenance behaviours. In the absence of the “natural” possibilities to perform such behaviours, provision of artificial opportunities to perform “similar-to-natural” food manipulation or species-typical behaviours such as playing, roaming, dustbathing, hiding, solving problems etc., for example by providing the animals with substrates for manipulation or with hiding opportunities, can improve welfare and health, including reproductive, cognitive and emotional functioning. Such provisions are known as “behavioural enrichment” and “environmental enrichment”12 of animal housing. Careful handling and appropriate veterinary care are also important aspects of animal welfare: with careful handling (also based on taming), procedures involved in daily care, veterinary care, research protocols and production become less stressful and less dangerous for the animals and staff alike. Use of anaesthesia and analgesics are a mandatory part of invasive procedures and should be controlled by veterinarians. Further requirements for good health and welfare include adequate nutrition and physical parameters such as space allowance, appropriate flooring, airing, temperature, etc. While good production often indicates good health and nutrition, overproduction has become a problem in many livestock breeds. When the animal is producing at too high a level, this becomes a welfare problem in itself, straining the animal’s physiology and leading to so-called production diseases13 (for example overly heavy muscles, overly large udders, metabolic imbalances, stomach problems caused by too nutritious diets). The housing and management of transgenic animals in a commercial biopharmaceutics production process may be subject to specific requirements, for example to avoid infectious pathogens in the final product14. Although such requirements regarding the production environment may 12 13 14

Smith and Taylor 1996. Mills et al. 1997. Costa 1997; Schmitt 2004.

108


help to assure a comprehensive health monitoring system, they may also result in particularly barren environments or movement restrictions that can infringe on animal welfare, as described above. To some extent, these can be counteracted with environmental enrichment, but there are limits: for example, a tethered or crated animal will always be too restricted in its movement. The restrictions that the requirements may impose should be considered at the outset of an animal pharming project, particularly with regard to the choice of species, breed and protein expression sites.

4.4.2 Protein collection and excess offspring The extent to which the collection of the fluids or tissues containing the transgenic protein is detrimental to the animals depends on where the protein is expressed, on the species, and on the collection procedures. For milking, standard husbandry recommendations apply in cattle, sheep and goats. Care should be taken to choose breeds adapted to being milked by humans; animals of some breeds can in some cases only be milked after hormonal treatment to provoke milk let-down. The same applies to non-dairy species: milking of pigs and rabbits requires careful development of methodology and monitoring of stressfulness, as comparatively little is known about their milking physiology. Another aspect with regard to expression of recombinant protein in milk is that it is not desirable to aim at very high expression levels that substantially alter the milk’s composition, as this could lead to discomfort during milking (cheesy milk). Semen can be collected relatively non-invasively provided there are good animal husbandry routines. Collection of urine would probably involve metabolic crates and therefore be highly restrictive on the animals’ ability to move. Collection of blood at the production level would be highly invasive or involve slaughter. In the case of painless slaughter, as with excess animals, this would strictly speaking not be an animal welfare problem. However, it could pose problems of acceptance and acceptability, as discussed in chapter 6. Excess offspring could be a considerable problem in the production phase, for instance if only one of the genders can be used for collection (i.e. product is in milk). Similar problems exist in food production, where male chicks of laying hens, for example, are also excess and normally destroyed as soon as their gender is known. Excess pharming animals are not allowed to enter the food production chain, so they have to be destroyed. One way around this would be reproductive methodology that can select the gender of the offspring.

4.4.3 Reproduction Transgenic animals can in principle be bred with conventional animal husbandry procedures. In some species (for example cattle), this already involves routine artificial insemination, whereas others (for example sheep,

4.5 Animal welfare considerations in the development phase

109

rabbits) are typically bred by mating. Increasingly, reproductive technologies like embryo transfer are used. Stress can occur in these contexts, for example if restraint, hormonal treatments and invasive procedures are necessary – this is a situation similar to that on farms producing conventional products. If the animals are cloned in order to ensure genetic identity (reproductive cloning), a number of problems may occur; again, this would be comparable with problems that can occur when traditional farm, sport or companion animals are cloned. Typical problems occurring with cloning are described in the context of the research and development phase, because as detailed in chapter 2, cloning can be used not only to generate genetically identical offspring, but also to generate transgenic animals, and it is in that context that it is most likely to be used for pharming.

4.4.4 Effects of genotype In principle, because pharming animals are utilised for production (and not as disease models), the aim is to generate healthy animals. Therefore, not only from an animal welfare point of view, but also from a production point of view, it is a goal to rule out adverse effects of the altered genotype in the research and development phase, rather than waiting for them to become a problem in a line that has entered production. Only such adverse effects of transgenesis that have been accepted as a side effect, or that have remained undetected in the experimental phase, are therefore relevant for the production phase. Effects of transgenesis may remain undetected in the experimental phase for various reasons – for example, they may only arise at a certain age or production stage, or they may only be visible in production rather than laboratory conditions. Another source of health and welfare problems in the production herd may be long-term effects due to manipulations carried out at the embryo stage in order to create the transgenic founder. To date, there are few studies monitoring the long-term effects of transgenesis and in vitro reproductive techniques. With regard to cell nuclear transfer, the few long-term studies indicate that cloned animals can have normal zootechnical characteristics15 and produce healthy offspring16. However, further studies are required.

4.5 Animal welfare considerations in the development phase The general considerations with regard to farm animal housing and management that were made in the previous section in principle apply in the laboratory, too. The laboratory may pose challenges such as individual housing, barren environment and invasive handling. Abundant literature is available advising on how to avoid such conditions and keep stress arising 15 16

Enright et al. 2002; Govoni et al. 2002; Pace et al. 2002. Heyman et al. 2004.

110


from such conditions to a minimum17, and self control of the laboratories is coordinated by for example the American Association for the Accreditation of Laboratory Animal Care (AAALAC). Animal protection legislation specific to animals used in scientific procedures is analysed in chapter 8. In the experimental phase (as opposed to the production phase), the generation of transgenic animals is a trial and error process, during which animals with adverse genotypes will be created. Also, the generation of transgenic animals involves in vitro reproductive technologies that may disrupt development. At present, the methods for producing transgenic mammals work best in mice, and are still associated with high losses or low efficiency in the species most relevant to pharming (see section 2.3).

4.5.1 Transgenesis, expression of medicinal protein, and transgene evaluation The animal welfare consequences of transgene expression depend on the nature of transgenesis, i.e. the bioactivity, tissue-specificity, route of secretion, temporal expression pattern and concentration of the protein to be produced. Harm to the animals can be caused by the bioactivity of the foreign protein: either at the intended expression site, because it enters the body’s circulation, or because of its expression at unintended sites. Preliminary studies with mice can highlight problems with bioactivity. While this means conducting additional animal experiments, it may nevertheless be a sensible approach as the transgenic technology is well-established and more efficient in mice than in large animals. For animal pharming, transgene expression is at present mainly considered in milk, urine, seminal fluid, blood and, in chickens, eggs. As mentioned in section 2.3, blood is generally problematic as an expression tissue because it means that a bioactive foreign protein is directly available to the animal’s entire physiology. Milk can also be problematic because proteins can leak from the mammary gland into the animal’s blood18. Proteins secreted exclusively in the milk thus may not stay there, and although their presence may not be adverse in the mammary gland, they may have adverse effects on the animal’s physiology once they have entered its circulation. Proteins secreted into milk may also have adverse effects in the mammary gland itself19. Expression sites that produce protein strictly for secretion out of the body (semen, urine) do not pose this risk. An example of insufficient tissue-specificity of transgene expression was reported by Massoud20 and colleagues, who found that rabbits engineered to express human erythropoietin (EPO) in their mammary glands also expressed EPO at low levels in other organs, resulting in elevated num17 18 19 20

E.g. Poole 1999; Kaliste 2004; Institute of Laboratory Animal Research 1996. E.g. Lubon 1998, Devinoy et al. 1995. Shamay et al. 1992. Massoud et al. 1996.


111

bers of red blood cells, infertility and premature death. Ectopic expression of mouse whey acidic protein (which is thought to be less detrimental than EPO) in transgenic sheep may have caused serious health problems in another study21. Insufficient tissue-specificity can be caused by lack of regulatory elements22, or it can be due to position effects. Position effects can occur with all techniques for generating transgenic animals, except with gene targeting (see section 2.3). This increases the possibility of aberrant expression, as regulatory parts of host genes near the transgene can influence its expression with regard to concentrations, temporal patterns and sites. Attempts are being made to refine the technology to better “insulate” the transgenes against such effects. In addition, the random placement of the transgene in the genome involves the possibility of insertional mutations which disrupt endogenous gene expression. Depending on which host gene has been disturbed, there can be harmful consequences for the animal. At present, typical transgene analysis therefore involves analysis of transgene mRNA expression and veterinary examination of the transgenic animals to identify obvious health problems. Dominant insertional mutations have been reported, for example in transgenic mice that developed nephrotic syndrome23 and craniofacial abnormalities24, but typically, insertional mutations are recessive. Therefore, transgene insertion is primarily a problem if the animals are bred to homozygocity. However, there may be hidden subtle dysfunctions of recessive changes in hemizygotes. Therefore, ectopic expression is checked in all organs, but subtle defects and their effects on health and welfare may remain unnoticed. From an animal welfare point of view it is important to quickly detect and treat or eliminate pain or distress caused by inappropriate transgene expression, and to avoid breeding a line that turns out to have welfare problems. This can be achieved by including welfare assessment protocols in the routine evaluation of transgenic animals. While admittedly a measure fraught with practical difficulties, such inclusion has been tried with mice and is feasible25. Analysis of such welfare protocols could, in addition, lead to new knowledge that could be used to make more informed statements about the welfare of pharming animals, and more precise predictions in further research and production projects26. While essential and required by the regulating agencies, transgene evaluation may also in itself compromise welfare. The aversiveness of the laboratory routines partly depends on the handling, as restraint and related fear 21 22 23 24 25 26

Wall et al. 1996. Wells and Wall 1999. Weiher et al. 1990. Ting et al. 1994. Costa 1997; Mertens and Rülicke 1999; van der Meer et al. 2001. Van Reenen et al. 2001; Olsson and Sandøe 2004.

112


can cause stress reactions in the animals. In addition to standard laboratory procedures, such as blood sampling and tissue biopsies, animal pharming also sometimes involves more special procedures in transgenic animal analysis. The early induction of lactation (in sexually immature females and males) is an invasive hormonal treatment that is likely to be uncomfortable to the animals and may also be questionable on ethical grounds for reasons related to animal dignity (see chapter 6). However, if the recombinant protein is expressed in the mammary gland and the transgenic founder animal to be evaluated is male, the alternative is to breed the animal and analyse its female offspring. This of course is a time-consuming procedure that involves the use of additional animals.

4.5.2 Reproductive technologies The reproductive technologies used to generate transgenic pharming animals are by no means exclusive to pharming or to the generation of transgenic animals: companion and farm animals have been cloned to produce genetically identical animals, and techniques like in vitro fertilisation and embryo transfer are increasingly common in breeding programmes. However, because the reproductive technologies are a major risk factor in the creation of transgenic animals, they will be dealt with here. While an animal’s development can be disturbed by the genetic effects of transgenesis (i.e. the nature or placement of the transgene, see above), it is possibly more frequently disrupted by epigenetic changes caused by the in vitro reproductive technologies used in the generation of transgenesis (see section 2.3). It is not yet known in detail how the various reproductive and gene technological procedures contribute to abnormal development27, which can cause suffering to the foster mothers, to the foetuses if they are sentient, to the new born animals, and possibly also longer-term. Problems with foetal development and around birth appear to be particularly prevalent with in vitro manipulations. For example, calves from in vitro embryo production showed an increased incidence of high birth weights, malformations and perinatal mortality28. In recent years, refinement of the in vitro techniques (for example the choice of medium) has led to higher success rates and healthy offspring. 4.5.2.1 Developmental problems in somatic cell nuclear transfer (cloning)

Somatic cell nuclear transfer (i.e. cell cloning) is sometimes the technology of choice to generate transgenic founder animals (see section 2.3). It involves an in vitro phase, reprogramming of the somatic cell nucleus, and the reconstructed oocyte is exposed to facilitating and activating stimuli. 27 28

Young and Fairburn 2000. Van Reenen and Blokhuis 1997.


113

At the present state of the technology, the net effect is disturbance in the regulation of gene expression in early embryogenesis and inefficiency in producing live offspring. Death of cloned embryos and foetuses is most prevalent in early pregnancy but can occur throughout pregnancy, and a high proportion of those that survive to term die soon after birth29. Placental dysfunction and abortions compromise foster mother welfare. Foetuses that die in the 3rd trimester have symptoms such as amniotic squames and meconium in the lungs that would be indicative of suffering, provided they have the ability to suffer. It is not known at what point sentience arises in the development of an individual. According to a report by the European Food Safety Authority’s Animal Health and Animal Welfare Panel30, current knowledge indicates that mammals are probably not conscious until they breathe air, although it is accepted that foetuses are responsive and have the possibility of associative learning. It is therefore safe to say that very little is known about prenatal ability to experience pain and distress. If foetuses develop to term, there is often impaired hormonal signalling in preparation for birth, and offspring are often unusually large. This can lead to birth complications and often makes caesarean delivery necessary. Different approaches are taken to manage the newborn animals to decrease the risks of suffering and death. These include keeping the young with their foster mothers to avoid artificial feeding and handling of the newborns31 or very intensive veterinary observation and care32; strategies that can be combined to a limited extent only. In cloned foetuses or neonatal offspring, a typical cluster of symptoms, including abnormally large birth weight, is called “Large Offspring Syndrome”. Coculture or serum in the culture medium is thought to be the reason for LOS33 and it is hoped that it will be overcome in the future. Postnatal complications reported in cloned cattle include lung dysmaturity, pulmonary hypertension, respiratory distress/failure, decreased oxygen supply in body tissues, decreased body temperature, hypoglycemia, metabolic acidosis, enlarged umbilical vessels, development of sepsis in umbilical structures or lungs, increased birth weight, asynchronously large organs, musculoskeletal abnormalities (especially contracted flexor tendons), abnormal immune function, anaemia, brain lesions, depression and prolonged recumbency34. 29 30 31 32 33 34

Edwards et al. 2003. EFSA 2005. Panarace et al. 2006. Fecteau et al. 2005. Young et al. 1998. Brem and Kuhholzer 2002; Chavatte-Palmer et al. 2002; Chavatte-Palmer et al. 2004; Fecteau et al. 2005; Heyman et al. 2004; Li et al. 2005; Panarace et al. 2006; Renard et al. 1999; Tsunoda and Kato 2002; Wells et al. 2004.

114


These types of complications have been found not only with adult cell nuclear transfer, but also with foetal35 and embryo36 cell nuclear transfer. In some cases, the severity of such complications has not been apparent directly after birth37. On the other hand, not all somatic cell nuclear transfer derived offspring are ill38. It also appears that physiological differences decrease with increased age39 and need not be found in animals that survive to adulthood: there have been few long-term studies so far, but cloned cattle can have growth rates and reproductive characteristics that do not differ from non-cloned40 cattle. 4.5.2.2 Donor animals and foster mothers

Reproductive technologies involve not only the offspring that is being created, but also foster mothers and, in some cases, egg cell donor animals. As described in section 2.3, egg cells can either be collected from donor animals, or they can be obtained from ovaries that are by-products in slaughterhouses. For live donor animals, the adverse effects of either the explantation of ovaries or the collection of in vivo fertilised eggs depend on method and species. Superovulation, which is used to increase the number of eggs, is known to lead to some physical discomfort and abdominal pain41. Ultrasound-guided transvaginal oocyte recovery is a mildly aversive invasive procedure. From a donor welfare point of view, the alternative, obtaining ovaries from slaughterhouses, is preferable. However, the in vitro reproduction techniques are not well established for all species in question. Also, in vitro reproduction results in less viable embryos, increasing the number of recipients needed, and potential problems with foetal and perinatal health. For the recipient animals (foster mothers), hormonal priming or induction of pseudopregnancy are probably mildly aversive procedures. Transfer of embryos into the oviducts is more or less invasive depending on the species; non-surgical transfer is possible in large animals and rabbits, i.e. all typical pharming mammals. The number of recipients needed depends on the viability of the implanted embryos and the efficiency of transgenesis. The efficiency of transgenesis is much higher with nuclear transfer than with microinjection, because nuclear transfer allows for in vitro analysis of the transgenic embryos (see section 2.3). Recipients are negatively affected in the case of abnormal foetal development and the necessity of caesarean sections for delivery, which are particularly common with somatic cell nuclear transfer (see above). 35 36 37 38 39 40 41

Hill et al. 1999. Garry et al. 1996. Gibbons et al. 2002. Lanza et al. 2001; Piedrahita et al. 2002. Chavatte-Palmer et al. 2004. Enright et al. 2002; Govoni et al. 2002; Pace et al. 2002. Boivin and Takefman 1996.

4.6 Conclusions

115

4.5.3 Excess offspring With pronuclear DNA microinjection, because analysis of the transgene’s expression must be carried out in the founder animals and cannot, as with cell nuclear transfer, be carried out at the cell stage, large numbers of animals who do not carry the gene or do not express it are produced: the percentage of offspring carrying the transgene is 1–5 %. If the remainder are killed painlessly this is not strictly speaking an animal welfare problem (although it is still an ethical problem). Excess animals produced in the experimental phase of animal pharming can sometimes be used in another experiment. The number of non-expressing (or poorly expressing) animals born can be reduced with various methods, but there is often a trade-off: methods that produce more transgenic embryos sometimes reduce their viability. As described above, however, the use of cell nuclear transfer, allowing a considerable reduction of non-transgenic numbers, may lead to high incidence of foetal and perinatal diseases, abnormalities, and death.

4.6 Conclusions The choice of methodologies that takes place at the early planning stage of an animal pharming project provides opportunities to avoid animal welfare problems. An integrated risk assessment, covering not only the experimental phase but also the potential animal welfare hazards in the production phase, is therefore desirable. Such risk assessment would benefit from more knowledge about the animal welfare effects of different types of transgenesis and of the procedures involved in creating the transgenic animals. Profound animal welfare concerns arise in the experimental phase, where 1) some laboratory procedures are invasive, 2) developmental problems may arise, and 3) there is a degree of trial and error with regard to the effects of transgenesis. With regard to the first point, non-invasive options should be chosen where possible. Such choice of non-invasive options may, however, in some cases be ambivalent. For example, it is less invasive to obtain oocytes from slaughterhouse material rather than from live oocyte donors, but oocyte viability is less. The hormonal induction of lactation in males and pre-pubertal females should be avoided from an animal welfare point of view, but this may lead to increased numbers of animal to be bred. With regard to the second point: while the amount of developmental problems is substantially lower with pronuclear DNA microinjection than with cell nuclear transfer, the inefficiency of producing transgenic founders with microinjection means that greater numbers of foster mothers will be needed and more excess animals will be produced with this approach. With regard to the third point: the monitoring of the effects of transgenesis (phenotyping) needs to include animal welfare parameters both on the level

116


of the individual animal (to allow early euthanasia) and on the level of the transgenic line that will later enter production. Before a transgenic pharming line enters production, it should thus have been carefully monitored with regard to negative effects of transgenesis. However, health and welfare monitoring programmes should be continued in the production phase to ensure detection of subtle, previously undetected genetic or epigenetic changes. This would also help to reveal potential recombination or copy-loss, and long-term effects of the intended recombinant protein expression. Unintended adverse effects of recombinant protein expression in the production phase are prevented to some extent through choices that are made in the experimental phase about the expression site for the recombinant protein. Urine, eggs and semen are desirable expression sites: they pose the least risk of the transgene entering circulation. However, it is also important that the expression site allows for non-invasive protein collection. It would therefore be advisable to develop methods of urine collection that do not involve restrictions to animal movement. The mammary glands can be good expression sites in dairy animals, but in species like pigs and rabbits milking may be stressful. Also, side-effects of transgene expression cannot be excluded as easily in milk as in urine, eggs or semen. Another aspect of the animal pharming production phase, that ought to be taken into account early on during a risk assessment in an animal pharming project, is that special requirements with regard to husbandry and management may apply. For example the animals may not be allowed to be outside. Choice of species, or of breeds within species, may be important determinants of the magnitude of problems that may arise from such requirements. If compromises are made in this respect, appropriate environmental enrichment has to be ensured. Environmental enrichment is often advisable even where no special hygienic housing requirements apply. Conventional husbandry conditions may, in many cases, cause the largest amount of welfare problems in an animal pharming project. Such welfare problems that are not pharming-specific, but typically occur in other types of animal utilisation too, are not trivial. Efforts are being made in many countries and internationally (for example European Food Safety Authority Animal Health and Animal Welfare Panel, World Organisation for Animal Health) to improve this. Animal pharming could be at the forefront of animal welfare provisions rather than remaining comparable to current levels of welfare. Some of the suggestions made here to promote the welfare of animals used for pharming may involve costs to the experimenters, producers, industry or beneficiaries of the envisaged pharmaceutical products – or indeed to other animals involved in the process. Costs and benefits will thus, in many cases, have to be weighed against each other, or other judgements may overrule such an analysis, as discussed in chapters 6 and 8.

4.7 References

117

4.7 References Boivin J, Takefman JE (1996) Impact of in-vitro fertilization process on emotional, physical and relational variables. Human Reprod 11:903–907 Brem G, Kuhholzer B (2002) The recent history of somatic cloning in mammals. Cloning Stem Cells 1:57–63 Broom DM (1991) Animal welfare concepts and measurement. J Anim Sci 69:4167–4175 Broom DM, Johnson KG (1993) Stress and Animal Welfare. Chapman and Hall, London Canadian Council on Animal Care (1997) Canadian Council on Animal Care (CCAC) Guidelines on Transgenic Animals www.ccac.ca/en/CCAC_Programs/Guidelines_Policies/GDLINES/ TRANSGEN/TRANSGE1.HTM (April 2008) Chavatte-Palmer P, Heyman Y, Richard C, Monget P, LeBourhis D, Kann G, Chilliard Y, Vignon X, Renard JP (2002) Clinical, hormonal, and hematologic characteristics of bovine calves derived from nuclei from somatic cells. Biol Reprod 66:1596–1603 Chavatte-Palmer P, Remy D, Cordonnier N, Richard C, Issenmann H, Laigre P, Heyman Y, Mialot JP (2004) Health status of cattle at different ages. Cloning Stem Cells 6:94–100 Costa P (1997) Production of transgenic animals: practical problems and welfare aspects. In: van Zutphen LFM, van der Meer M (eds) Welfare aspects of transgenic animals. Springer, Berlin, pp 68–77 Dawkins MS (1980) Animal suffering: the science of animal welfare. Chapman and Hall, London Devinoy E, Stinnakre MG, Lavialle F, Thépot D, Ollivier-Bousquet M (1995) Intracellular routing and release of caseins and growth hormone produced into milk from transgenic mice. Experimental Cell Research 221:272–280 Duncan I (1993) Welfare is to do with what animals feel. J Ag & Appl Ethics 6:8–14 Duncan I (2005) Science-based assessment of animal welfare: farm animals. Rev sci tech Off int Epiz 24:483–492 Eaton M (2004) Ethics and the Business of Bioscience. Stanford University Press, Stanford Edwards JL, Schrick FN, McCracken MD, van Amstel SR, Hopkins FM, Welborn MG, Davies CJ (2003) Cloning adult farm animals: a review of the possibilities and problems associated with somatic cell nuclear transfer. Am J Reprod Immun 50:113–123 EFSA (2005) Aspects of the biology and welfare of animals used for experimental and other scientific purposes. EFSA-Q-2004-105. Annex to the European Food Safety Authority Journal 292:1–136 EFSA (2006) Basic information for the development of the animal welfare risk assessment guidelines www.efsa.europa.eu/EFSA/DocumentSet/AHAW_report_basicinfo_AWRA_ en,0.pdf (June 2008) EFSA (2007) Opinion of the Scientific Panel on Animal Health and Welfare on a selfmandate on the Framework for EFSA AHAW Risk Assessments, The EFSA Journal 550:1–46 http://www.efsa.europa.eu/EFSA/Scientific_Opinion/ahaw_op_ej550_ framework_en,1.pdf (June 2008) Enright BP, Taneja M, Schreiber D, Riesen J, Tian XC, Fortune JE, Yang X (2002) Reproductive characteristics of cloned heifers derived from adult somatic cells. Biol Reprod 66:291–296

118


Fecteau ME, Palmer JE, Wilkins PA (2005) Neonatal care of high-risk cloned and transgenic calves. Vet Clin Food Anim 21:637–653 Fraser AF, Broom DM (1997) Farm animal behaviour and welfare. CAB International Garry FB, Adams R, McCann JP, Odde KG (1996) Postnatal characteristics of calves produced by nuclear transfer cloning. Theriogenology 45:141–152 Gibbons J, Arat S, Rzucidlo J, Miyoshi K, Waltenburg R, Respess D, Venable A, Stice S (2002) Enhanced survivability of cloned calves derived from roscovitinetreated adult somatic cells. Biol Reprod 66:199–203 Govoni KE, Tian XC, Kazmer GW, Taneja M, Enright BP, Rivard AL, Yang X, Zinn SA (2002) Age-related changes of the somatotropic axis in cloned Holstein calves. Biol Reprod 66:291–296 Heyman Y, Richard C, Rodriguez-Martinez H, Lazzari G, Chavatte-Palmer P, Vignon X, Galli C (2004) Zootechnical performance of cloned cattle and offspring: preliminary results. Cloning Stem Cells 6:111–120 Hill JR, Roussel AJ, Cibelli JB, Edwards JF, Hooper NL, Miller MW, Thompson JA, Looney CR, Westhusin ME, Robl JM, Stice SL (1999) Clinical and pathological features of cloned transgenic calves and fetuses (13 case studies). Theriogenology 51:1451–1465 Institute of Laboratory Animal Research, Commission on Life Sciences, National Research Council (1996) Guide for the care and use of laboratory animals. National Acadamy Press Kaliste E (ed) (2004) The welfare of laboratory animals. Springer, Berlin Lanza RP, Cibelli JB, Faber D, Sweeney RW, Henderson B, Nevala W, West MD, Wettstein PJ (2001) Cloned cattle can be healthy and normal. Science 294:1893–1894 Li S, Li Y, Du W, Zhang L, Yu S, Dai Y, Zhao C, Li N (2005) Aberrant gene expression in organs of bovine clones that die within two days after birth. Biol Reprod 72:258–265 Lubon H (1998) Transgenic animal bioreactors in biotechnology and production of blood proteins. Biotechnol Annu Rev 4:1–54 MacKenzie AA (2005) Applications of genetic engineering for livestock and biotechnology products. World Organisation for Animal Health (OIE): 73 SG/10 www.oie.int/downld/SG/2005/A_73 %20SG_10.pdf (January 2007) Massoud M, Attal J, Thépot D, Pointu H, Stinnakre MG, Théron MC, Lopez C, Houdebine LM (1996) The deleterious effects of human erythropoietin gene driven by the rabbit whey acidic protein gene promoter in transgenic rabbits. Reprod Nutr Dev 36:555–563 Mertens C, Rülicke T (1999) Score sheets for the monitoring of transgenic mice. Animal Welfare 8:433–438 Mertens C, Rülicke T (2000) Phenotype characterisation and welfare assessment of transgenic rodents (mice). Journal of Applied Animal Welfare Science 3:127–139 Mills AD, Beilharz RG, Hocking PM (1997) Genetic selection. In: Appleby MC, Hughes, BO (eds) Animal Welfare. CABI, pp 219–231 Moberg GP, Mench JA (eds) (2000) The biology of animal stress: basic principles and implications for animal welfare. CABI, New York Müller-Graf C, Candiani D, Barbieri S, Ribó O, Afonso A, Aiassa E, Have P, Correia S, De Massis F, Grudnik T, Serratosa J (2007) Risk assessment in animal welfare – EFSA approach. Alternatives to Animal Testing and Experimentation 14, Special Issue:189–794 http://altweb.jhsph.edu/wc6/paper789.pdf (June 2008) Olsson IAS, Sandøe P (2004) Ethical decisions concerning animal biotechnology: what is the role of animal welfare science? Animal Welfare 13:139–144

4.7 References

119

Pace MM, Augenstein ML, Betthauser JM, Childs LA, Eilertsen KJ, Enos JM, Forsberg EJ, Golueke PJ, Graber DF, Kemper JC, Koppang RW, Lange G, Lesmeister TL, Mallon KS, Mell GD, Misica PM, Pfister-Genskow M, Strelchenko NS, Voelker GR Watt SR, Bishop MD (2002) Ontongeny of cloned cattle to lactation. Biol Reprod 67:334–339 Panarace M, Agüero JI, Garrote M, Jauregui G, Segovia A, Cané L, Gutiérrez J, Marfil M, Rigali F, Pugliese M, Young S, Lagioia J, Garnil C, Forte Pontes JE, Ereno Junio JC, Mower S, Medina M (2006) How healthy are clones and their progeny: 5 years of field experience. Theriogenology 67:142–151 Perry GC (ed) (2004) Welfare of the Laying Hen (Poultry Science Symposium, No. 27). CABI Piedrahita JA, Wells DN, Miller AL, Oliver JE, Berg MC, Peterson AJ, Tervit HR (2002) Effects of follicular size of cytoplast donor on the efficiency of cloning in cattle. Mol Reprod Dev 61:317–326 Poole TB (ed) (1999) The UFAW Handbook on the Care and Management of Laboratory Animals. Blackwell Pursel VG, Pinkert CA, Miller KF, Bolt DJ, Campbell RG, Palmiter RD, Brinster RL, Hammer RE (1989) Genetic engineering of livestock. Science 244:1281–1288 Renard JP, Chastant S, Chesne P, Richard C, Marchal J, Cordonnier N, Chavatte P, Vignon X (1999) Lymphoid hypoplasia and somatic cloning. Lancet 353:1489–1491 Russell WMS, Burch RL (1959) The Principles of Humane Experimental Technique. Methuen & Co., London Sandøe P, Forsman B, Hansen AK (1997) Transgenic animals: the need for ethical dialogue. In: van Zutphen LFM, van der Meer M (eds) Welfare aspects of transgenic animals. Springer, Berlin, pp 90–101 Schmitt EH (2004) Regulatory background in the development of medicinal products for human use produced by transgenic animals – current situation and perspective in the EU and USA. Wissenschaftliche Prüfungsarbeit zur Erlangung des Titels “Master of Drug Regulatory Affairs“ der Rheinischen Friedrich-Wilhelms-Universität Bonn Shamay A, Pursel VG, Wilkinson F, Wall RJ, Henninghausen L (1992) Expression of whey acidic protein in transgenic pigs impairs mammary development. Transgenic Research 1:124–132 Smith CP, Taylor V (eds) (1996) Environmental enrichment information resources for laboratory animals: 1965–1995. Birds, cats, dogs, farm animals, ferrets, rabbits, and rodents. Diane Publishing Company, Darby Stolba A, Wood-Gush DGM (1984) The identification of behavioural key features and their incorporation into a housing design for pigs. Annales de Recherches Véterinaires 15:287–298 Stolba A, Wood-Gush DGM (1989) The behaviour of pigs in a seminatural environment. Animal Production 48:419–425 Ting CN, Kohrmann D, Burgess DL, Boyle A, Altschuler A, Gholizadeh G, Samuelson LC, Jang W, Meisler MH (1994) Insertional mutations of mouse chromosome 18 with vestibular and cranofacial abnormalities. Genetics 136:247–254 Tsunoda Y, Kato Y (2002) Recent progress and problems in animal cloning. Differentiation 69:158–161 Van der Meer M, Rolls A, Baumans V, Olivier B, van Zutphen LFM (2001) Use of score sheets for welfare assessment of transgenic mice. Laboratory Animals 35:379–389 Van Reenen CG, Blokhuis HJ (1997) Evaluation of welfare of transgenic animals; lessons from a case study in cattle. In: Nilsson A (ed) Proc. transgenic animals and food production workshop, Stockholm, Sweden. J Royal Swedish Academy of Agriculture and Forestry 136:99–105

120


Van Reenen CG, Meuwissen THE, Hopster H, Oldenbrock K, Kruip TAM, Blokhuis HJ (2001) Transgenesis may affect animal welfare: a case for systematic risk assessment. J Anim Sci 79:1763–1779 Wall RJ, Rexroad CE, Powell A, Shamay A, McKnight R, Henninghausen L (1996) Synthesis and secretion of the mouse whey acidic protein in transgenic sheep. Transgenic Research 5:67–72 Weiher H, Nod T, Gray DA, Sharpe AH, Jaenisch R (1990) Transgenic mouse model of kidney disease: insertional inactivation of ubiquitously expressed gene leads to nephrotic syndrome. Cell 62:425–434 Wells DN, Wall RJ (1999) One gene is not enough. In: Murray JD, Anderson GB, Oberauer AM, McGloughlin MM (eds) Transgenic animals in agriculture. CABI, Oxon pp 37–56 Wells DN, Forsyth JT, McMillan V, Oback B (2004) The health of somatic cell cloned cattle and their offspring. Cloning Stem Cells 2:101–110 Young LE, Sinclair KD, Wilmut I (1998) Large offspring syndrome in cattle and sheep. Rev Reprod 3:155–163 Young LE, Fairburn HR (2000) Improving the safety of embryo technologies: possible role of genomic imprinting. Theriogenology 53:627–648

5 Public views and attitudes to pharming

5.1 Introduction A singular phenomenon arose in the last third of the 20th century and remains with us today. Its origins lie in the confluence of two contrasting vectors. Firstly, an intense dependence on science and technology as the powerhouse of economic growth, and a precondition for the satisfaction of a broad range of needs and expectations from economic prosperity to leisure by way of healthcare and responsible management of the natural environment. Secondly, pockets of cultural and social uneasiness about the implications and effects of scientific advances. An indirect indicator of the first vector – the weight of science and technology – is the many terms coined, with varying degrees of accuracy, to denote the structure and dynamics of contemporary society bearing the stamp of high-profile scientific-technological developments. A few names should suffice to illustrate the importance accorded to the technology base, especially the information technologies frequently enthroned as the defining force of the planet’s most developed societies: The Computerized Society (Martin and Norman 1970), Postindustrial Society (Touraine 1971, Bell 1973), Telematic Society (Nora and Minc 1978, Martin 1981), The Information Society (Martin and Butler 1981), The Information Era (Dizard 1982), The Control Revolution (Beniger 1986), High-Tech Society (Forester 1987), Network Society (Castells 1996). The most all-embracing terms, “knowledge economy” and “knowledge society”, have won themselves a privileged place in the vocabulary of social analysts, the mass media and policy-makers alike. Not only the way we refer to certain 20th-century scientific and technological advances but also a large body of empirical evidence attests to the importance and degree of (inter)dependence between economic growth and scientific advance. We might reasonably expect, then, that the values, attitudes and worldviews characteristic of the population, that is, the high culture “intellectual appropriation of science”1 alongside the culture of everyday life, would be thoroughly imbued with science and technology and enthusiasm about the constant rolling back of the frontiers of knowledge. But the picture is a lot more complex. Culturally, we observe attitudes of 1

Hard and Jamison 1998.

122


ambivalence, at times even rejection, towards specific techno-scientific developments and, if some essayists are right, a more general disenchantment with science and technology as a whole, on the cognitive plane (as a way of knowing) and, more so, with regard to the consequences of the advancement of science-based knowledge and its practical manifestation through technology. What a little over a decade ago was referred to as “science wars” was an expression of this cultural unease about science, typical of “postmodernity”, which went so far as to question the core epistemological principles that underpin essential modes of scientifically understanding the natural and social world, and triggered the response of a part of the scientific community2. The roots of this cultural phenomenon lie in the past, in particular the Neo-Romantic critique of science arising in the 1970s3. While acknowledging a cultural change in the way science was viewed in the last stretch of the 20th century, the postmodernist critique has proceeded, with a very limited empirical base, to generalize what occurs with specific subsets of science to science as a whole, and even the culture of our time. However, if we wish to characterize the current standing of science in the culture of advanced societies by reference to public perceptions of science and technology, the dominant profile emerges from the following points.4 Most areas of science and its application to satisfying social needs are strictly non problematic for the majority of people, and a good number of them are seen as clearly beneficial. The standard case is still that technological and scientific developments take their place silently in the background of the complex mode of the collective satisfaction of needs and, more weakly, in the cognitive schema of individuals, helping them interpret the world and organize the realm of everyday experience. In general, the attention paid to these advances beyond the scientific community is modest and short-lived. To put it another way, nowadays scientific themes have to vie for the interest of a public faced with a choice of information channels and subject areas way behind the scope of their interest, cognitive ability and the time at their disposal. The segment known as the “attentive public” (meeting the twin conditions of being “interested in” and “informed about” science) stands at around 10 %–15 %t of the adult population in advanced societies.5 Resistance or rejection phenomena are currently focused on one of the most dynamic scientific areas – that of biotechnology, while information 2 3 4 5

Gross and Levitt 1994; Gross, Levitt and Lewis 1996; Ross 1996; Sokal and Bricmont 1998; Levitt 1999; Weinberg 2001; Haack 2003. Holton 1992; Marx 1988. A fuller description can be found in Pardo 2003, chapter 4 of Solter et al. 2003:159–164. The “attentive public” concept was introduced by political scientist Gabriel Almond in 1950, referring to the public’s interest and involvement in foreign policy matters, then taken up by Jon D. Miller et al. 1980, 1983a in connection with scientific and technological issues.

5.1 Introduction

123

technologies are viewed in a clearly positive light and eagerly embraced in a growing number of areas.6 Public policies, which play a vital role in the development of biotechnology, exhibit traces of these critical or ambivalent views, placing strict external constraints on research as such and the market transfer of research outcomes. Researchers, entrepreneurs and, at times, the potential beneficiaries of biotechnological advances (particularly patients suffering certain diseases) can only look on in frustration and, at times, bafflement at the attitudes of the public in advanced societies and the excessive constraints imposed by regulators. A frequent diagnosis of this situation, viewed as anomalous by researchers in the biotechnology field compared to the treatment given to other areas of scientific knowledge, is the very low level of scientific literacy in general and, in particular, the public’s lack of familiarity with basic genetic concepts and principles. There is ample evidence for this deficit of public knowledge (in Europe, for instance, the mistaken belief and/or ignorance, repeatedly documented by the European Union’s “Eurobarometer” survey, that “genetically modified tomatoes contain genes while ordinary tomatoes do not”). But, as we will later discuss, this cognitive deficit finding is not enough to explain the complexity of public perceptions regarding biotechnology. The institutional and cultural framework for scientific research in biotechnology is very different from the one surrounding other knowledge areas which emerged earlier in the twentieth century (including nuclear power in its initial years and almost up to the mid 1960s). Specifically, the principle of “self regulation” by the scientific community has given way to stringent external regulation, a continuous outpouring of recommendations and guidelines from bioethical committees and, more recently, a range of participation mechanisms to give “voice” to the public and, in so doing, to avoid the alternative course of action, i.e., the public’s “exit” or alienation from science.7 Another widespread phenomenon of the closing years of the last century has brought added pressure to bear on science areas like biotechnology; namely, the extension of the democratic principle to what were once the exclusive preserves of expert opinion. In effect, since the start of modernity certain areas have been reserved for those groups who, through a lengthy and rigorous process of knowledge acquisition and parallel mechanisms of formal accreditation of the competencies attained, could legitimately exhibit their credentials in a clearly demarcated field of knowledge.8 The opening to the public of the science domain marks a far-reaching institutional change that is neatly summed up in the dictum of French physicist 6 7

8

Nelkin 1995. On the pair of concepts “exit” and “voice”, see the seminal contribution by Albert O. Hirschman 1970. On the favoured route for public participation in science policies, see Joss and Durant 1995; Frewer 1999; Einsiedel, Jelsøe and Breck 2001; Dietrich and Schibeci 2003. On the professionalisation of science, see Morrell 1990 and Ben-David 1985.

124


and essayist Jean Marc Lévy-Leblond that “conscience should take precedence over competence”.9 Until recently, however, the view of this author was very much a minority one, with no real practical repercussions. The response among science and communication analysts, policy-makers and, naturally, the scientific community was mainly to lament the failure of transmitting scientific knowledge effectively to society, and to lay almost exclusive blame for opposition to certain scientific-technological developments at the door of the low scientific literacy documented in numerous surveys. In the last three decades, a number of science policy agencies have been promoting regular “tests” of the public’s level of scientific knowledge and attitudes to science; a labour complemented by that of academic analysts and scientific societies (the National Science Board in the USA, the European Commission’s Eurobarometer). The field known as PUoS (Public Understanding of Science) came into being in the early 1990s. The then prevalent concern about the public’s low familiarity with the sciences (what came to be labelled the “deficit model”) was patent in the name of its first academic journal, “Public Understanding of Science”, which appeared in 1992 (though, since the outset, the issues it addresses have been broader in scope than the mere understanding of science by non experts). The first generation of PUoS research endorsed a program for the promotion of “public scientific literacy”, whose success, it was presumed, would bring multiple individual and collective benefits ranging from the empowerment of citizens in their private lives in economically advanced societies to others of an aesthetic nature, by way of the improvement of competitiveness and sustained growth.10 The literature was also confident that the public’s grasp of scientific knowledge would improve democratic government in the societies of the latter half of the twentieth century by encouraging public involvement in complex decision-making processes and simultaneously improving the effectiveness of these processes, producing if not an automatic consensus then at least more informed decisions than those that would emanate from a public with no understanding of science. Some authors, interested in the science-society relationship from the standpoint of decision making in a democratic framework, saw the formidable challenge posed for the conceptual foundations of democracy by the coexistence of, on the one hand, a society extremely dependent on science and technology and a policy-making process increasingly reliant on specialist science-based knowledge, and, on the other, the pretensions to “give voice” to a public alien to science.11 This is a challenge that would mean reformulating the theory of democracy and would “involve the public education 9 10 11

Lévy-Leblond 1992. See Thomas and Durant 1987; Pardo and Calvo 2002. Miller 1983b; Prewitt 1983.

5.1 Introduction

125

system in a task not unlike that of an earlier century, when a mass population was introduced to the norms and rules of democratic politics”.12 At the end of the 1990s this approach began to be disputed from a number of angles. Some of these were strictly analytical, like the difficulties of achieving statistically significant effects of medium-high magnitude in predicting attitudes to science on the basis of the level of knowledge variable. Other authors criticised the “public understanding of science” movement for its allegedly patronizing view of the public; at odds with the principles of democracy. And champions emerged for the values of “local knowledge” and lay knowledge, supposedly better adapted to the understanding of concrete phenomena than detached and abstract scientific knowledge, harking back to the Romantic and Neo-Romantic critique of science13. The result was the erosion and finally overthrow of the “deficit model”, and its replacement by the approach known as the “3Ds” – dialogue, discussion, and debate between the scientific community and the public.14 The goal of familiarizing the public with science did not disappear, but was joined by a new dimension that has been gaining in importance since the start of the present century. This shift in emphasis is clearly perceptible in the “Science and Society” report of the UK’s House of Lords15. The idea is to create a new climate for the science-society relationship based on dialogue, blurring the classic divide between public and scientific community as regards the former’s contribution to formulating and implementing extremely complex public policies. The thesis that there is no significant link between knowledge and attitudes has found its way into reports like that of the House of Lords. The buzzwords now are participation, dialogue, consensus and engagement. The search for institutional mechanisms aimed at “engaging the public” (to borrow a chapter heading from the House of Lords report) absorbs a large part of the analytical efforts and action programs underway in European countries with the support of the European Commission. Thus, the longstanding Scandinavian tradition of public participation in technology assessment and policies was embraced by other advanced societies in the closing years of the past century. Parallel with this new direction in science-public relations, some authors decided to take a second look at the empirical evidence on the link between knowledge and attitudes. In contrast to the previous finding that no link existed, this recent literature has been able to show, using more robust conceptual schemas and statistical analyses, that the public’s scientific knowledge goes some way to explaining their attitudes and that, on the whole, favourable predispositions to science are associated with familiarity with the scientific method and certain key scientific concepts and 12 13 14 15

Prewitt 1983:64. Wynne 1996; Marx 1988. Miller 2001. House of Lords 2000.

126


findings16. Furthermore, these and other recent analyses regarding the public’s perceptions of biotechnology offer a more complex explanatory model for attitudes to science, which reintroduces the knowledge variable along with a broader repertoire of constructs; notably values and ethical criteria, risk perceptions, worldviews (among them, views of nature), trust in the scientific community and regulators, and a series of contextual variables like the level of salience and content of media coverage of scientific advances and issues.17 In the specific case of biotechnology, room is also found for the variable of general attitudes towards science and technology or, as the literature has it, the role of general attitudes to the object x (in this case science) in explaining attitudes and behaviour towards a subset of the same.18 Focusing on the general area of biotechnology, which contains the emerging subset of public views on pharming, we can define its basic profile as follows. The public’s views and attitudes to biotechnology in advanced societies are characterized by virtually no familiarity with genetics, a rejection of, or alternatively, a failure to perceive the benefits to be derived from the ends of such research and a distrust of the means employed, i.e., the genetic engineering of the blueprint of plant and animal life. But this picture ignores the variability of views on different areas. In general, the socalled “red” biotechnologies, of a biomedical nature, are favourably perceived or, at least, do not meet with significant reservations, whereas “green” biotechnologies, focusing on the genetic modification of plants for agriculture and the production of foods (not for pharmaceuticals or cleaning up environmental pollution), are critically perceived19. In addition, the more or less active resistance of the first half of the 1990s has given way at the turn of the century to a moderate opposition or even a positive evaluation of some applications and, at any rate, to a more flexible perspective that discriminates according to the goal of the research and the specifics of the means that are utilized. This more differentiated evaluation will be of particular interest in the case of pharming.

5.2 Methodological considerations Before going on to examine attitudes to pharming, it is worth making a few methodological observations about the nature of the “attitudes” construct, and how to interpret the connections between series of attitudes of varying degrees of abstraction converging on an object or issue. 16 17 18 19

Pardo and Calvo 2006a; Bauer et al. 2007; Allum et al. 2008. Bauer and Gaskell 2002; Gaskell and Bauer 2006. Ajzen 2005. Gaskell 1997; Priest 2001; Pardo and Calvo 2002; Bauer and Gaskell 2002; Sturgis et al. 2004; Sturgis et al. 2005; Gaskell and Bauer 2006; Pardo and Calvo 2006b; Pardo and Calvo 2008.

5.2 Methodological considerations

127

Attitudes are predispositions toward more or less abstract objects (people, institutions, situations, symbols, ideas), and have three main components: cognitive (information and knowledge about the properties of the object), evaluative (affect and feelings of approval or disapproval toward the object) and conative (disposition to behave in a particular way towards an object). In Thurstone’s terms, the concept of attitude denotes “the sum total of a man’s inclinations and feelings, prejudice or bias, preconceived notions, ideas, fears, threats, and convictions about any specific topic”20. Opinions, in his seminal contribution, were the linguistic or symbolic expression of an attitude. Later research tried to ascertain the degree of strength and consistency of attitudes held by individuals and thereby gave a different meaning to the concept of “opinion”. Attitudes could be represented as embedded in a tree in which all parts are interconnected. The tree’s most basic component – the “root” – would correspond to an individual’s socio-psychological traits (from personality characteristics to social status), the stem would be analogous to core values (basic, general attitudes or worldviews), the branches to specific attitudes toward a vast array of objects and situations, and finally the leaves to opinions. The stability and depth of these different elements would be greater at the stem and the roots, while opinions would be subject to a high rate of change. Even a single piece of new information or a change in the framework in which information or the object of interest is presented can have a major impact on the opinions expressed by individuals. Since attitudes have generally been inferred not from the observation of actual behaviour but from individuals’ responses to survey questionnaires, the question of the validity and strength of attitudes toward a particular object or domain immediately arises. It is a well-known fact of research on survey technique that many individuals are willing to respond to an interviewer’s questions as a matter of courtesy, even regarding issues in which the individual has little interest, or that he/she may not have heard about. This kind of conduct generates noise in the matrix of opinion and attitude data of a population sample . Even when individuals have some information and evaluative orientation about a specific object that is of low salience to them, chances are that we will be measuring no more than weak “opinions”, or attitudes that are not rooted in their enduring interests and values, and which are therefore extremely unstable and of virtually no value for predicting behaviour.21 20 21

Thurstone and Chave 1929:7. Bishop 2005 offers a useful review of the myriad problems encountered in capturing an elusive object like “public opinion”. To overcome the most typical problems encountered in survey research, a large number of rules and principles, based on experiments and cognitive science, are applied in the design of questionnaires developed in the context of social science studies. Central pieces in the literature dealing with survey measurement issues are Turner and Martin 1984; Schuman and Presser 1996; Lyberg et al. 1997; Tourangeau et al. 2000.

128


In one of the most influential pieces of research on political attitudes, Philip Converse found that the American electorate fell short of the claims made by canonical theories about the sophistication and role of the public in democratic systems.22 The actual positions held by many individuals on logically connected issues (as assessed by the researcher) were found to be unrelated for many individuals: the intercorrelations between positions on a family of issues were extremely weak, which pointed to virtual independence among the value judgments (lack of attitudinal structure), and their volatility was high (lack of attitudinal stability over time). After Converse and others, the search for structure became one of the central tasks for researchers dealing with public opinion topics: “we speak of an ‘attitude structure’ when two or more beliefs or opinions held by an individual are in some way or another functionally related” […] and “attitude structures are often thought of as hierarchies in which more specific attitudes interact with attitudes toward the more general class of objects in which the specific object is seen to belong”23. The structuring of attitudes could range from a low level of constraint between positions on logically or semantically connected issues (i.e., a modest correlation between the issues) to a highly developed type of structure, that of belief systems or ideologies: “an ‘ideology’ may be seen as a particularly elaborate, close-woven, and far-ranging structure of attitudes. [...] An ideology [...] shares some of the characteristics of a taxonomic system”.24 It is clear that at least since the turn of the twenty-first century, not many individuals embrace an ideological system or Weltanschauung. The general rule is precisely the reverse: most people hold only fragmentary or “patchy” arrays of disparate attitudes, some of which are logically inconsistent, particularly when they deal with extremely complex objects that have simultaneous linkages with many logical classes or sets. “Any complex object may be located in a variety of general classes at the same time, and the values engaged may be in conflict. In the practical situation, evaluation may be strongly affected by extraneous concerns”.25 Biotechnology applications represent one of these cases. From the perspective of the scientific community, attitudes toward biotechnology would be expected to follow from the more general class of attitudes to which they logically pertain: predispositions toward science and technology in general. But they may also be related to attitudes towards progress, towards religious and moral beliefs, the natural environment, animals, economic competitiveness, health, and several other sets. It is plausible to expect that most of the citizens in the small segment of the informed or attentive public hold defined attitudes toward biotechnology, whereas it is unlikely that the least-informed stratum of citi22 23 24 25

Converse 1964, 1970. Campbell et al. 1960:190. Campbell et al. 1960:192–93. Campbell et al. 1960:191.

5.2 Methodological considerations

129

zens will have structured and stable attitudes in the sense defined by Converse. The largest group of all, i.e., people with a moderate level of interest and an even lower level of information will also have a medium to low level of structure in their evaluative views of an object. At any rate, the analyst must be aware of the many difficulties in the process of isolating attitudes toward an emerging and complex object of public opinion – as biotechnology pharming applications are – from a larger domain of values and attitudes. It is important to bear in mind that the explanatory schemas may suffer from the classical problem of misspecification (i.e., non inclusion of some relevant variables in the models). In this chapter, we will explore attitudes toward pharming while attempting to overcome some of the difficulties that arise in dealing with an object that is still fuzzy for many individuals. Apart from self-explanatory contingency tables, we will make use of summated scales (which try to capture attitudes using aggregated responses to a number of different evaluations of the object of interest, cancelling out, in part, the noise and measurement error associated with single variable measurements) and schemas or frames such as reservations and expectations about science at large, views of nature and animals (schemas which don’t impose strong logical dependencies between issues, but are open to many kinds of linkages, from associations based on previous experience to the “glue” provided by emotions, and social influences ranging from the immediate network of social relationships all the way to the culture of a particular society at a specific point in time). For the great majority of the population, moral issues and dilemmas neither present themselves as abstract or isolated matters nor are evaluated by reference to a single, uniform criterion or principle (although this can be the case for some minorities in relation to high salience issues involving their core values, which is particularly true for individuals affiliated to “single issue” organizations).26 Rather, they arise in specific contexts composed or integrated by several overlapping domains, in which multi26

The philosopher Leszek Kolakowski (1967) sees a degree of “inconsistency” as strictly non problematic and indeed argues in its praise: “[...] the race of inconsistent people continues to be one of the greatest sources of hope that possibly the human species will somehow manage to survive. [...] Total consistency is tantamount in practice to fanaticism; while inconsistency is the source of tolerance”. Kolakowski defines the consistent individual as “he who is in possession of a series of non-contradictory universal principles and strives to abide by them strictly in all he does and in all his opinions about what is right”. The stance maintained in this chapter is not evaluative (preference for a particular outcome or its opposite) but merely descriptive, on the grounds that, except for the minorities typified by certain “single issue” groups, an individual will in normal circumstances apply several principles and viewpoints when dealing with complex issues or objects, possibly giving rise to logically inconsistent views (particularly when such views touch on series of objects with a strong formal relationship, like views on specific objects falling within a broader class of views of which it would logically form a subset).

130


ple and diverse values or ethical principles may apply27. Typically there is a complex mix of convergent, mutually reinforcing values and beliefs, and also of contrasting or competing ones, leading to different possible decisions or courses of action. Any explanatory model aiming to give account of 30–50 % of attitude variance would need to call on a range of very different variables. This suggests that relations or linkages between such diverse influences cannot conform to a strictly logical pattern, as if we were talking about a formal system or scientific theory. It is possible to predict logical relations between some components of the structure underpinning attitudes (like the consistency principle referred to in chapter 6 of this monograph). But usually the linkages between structural components will be of a more relaxed or loosely connected nature, among other reasons because the object would be typically perceived with fuzzy boundaries (i.e., extending over several domains). In a highly idealized scenario, individuals would somehow have to “retrieve” from memory and list all the relevant beliefs and values connected to the issue at stake and “weigh up” each of them against the others, in order to make up their minds or adopt an evaluative position. In a more realistic setting, let’s say a “bounded rational” context in which the “actor” must face a number of constraints, chief among them a limited interest and cognitive capacity (“computational” resources), other factors besides the algorithmic calculation of the relative importance of each element in an array of relevant variables come into play.28 The result is a reduced or simplified “search space”, formed by a subset of highly salient variables, which the individual then uses to arrive at an evaluative position.29 Typically an individual in normal situations will deploy the minimum necessary cognitive resources to reach a “satisficing” or good enough decision, not an optimal one, as described by Simon and, after him, many other cognitive scientists. This will be especially the case when his or her interest or understanding of the question at stake is in the low to medium range. It is precisely this simplified frame that the pharming questionnaire has sought to simulate or approximate, and that is addressed by the proposed explanatory model. Pharming is clearly a new and highly complex issue that is hard to grasp for the majority of individuals lacking scientific training. Yet at the same time, it touches on core aspects of the “moral and ideational landscape” of society. Given the opposing influences of these two characteristics (com27

28 29

Campbell et al. 1960; Pardo and Calvo 2002. The analysis seeks to identify the varying evaluative angles that the typical individual will employ to reach a decision, for or against, regarding a given object. The Nobel laureate Herbert Simon is the main author of the “bounded rational” decision-making notion (Simon 1982). We borrow the notion of “search space” from the area of problem solving, part of the field of Artificial Intelligence and Cognitive Science. See the chapter “Exploring Alternatives” in Winston 1984.

5.3 Attitudes to pharming in advanced societies

131

plexity, discouraging the formation of value judgments; core nature of the values and cultural representations involved, encouraging an evaluative position), it is fair to suppose that most individuals – particularly those with a poor understanding of the issue’s scientific dimension – will fall back on general schemas or worldviews in forming their judgment about the acceptability of pharming research, as well as on the weight they accord to the goals of the research and, particularly, their own views of nature, plants and animals.

5.3 Attitudes to pharming in advanced societies: awareness and evaluative perspectives In this section we present a characterization of attitudes to plant and animal pharming in 15 advanced societies, twelve of them European plus Israel, Japan and the United States, spanning a wide spectrum of values, cultural traditions and social conditions.30 Commonalities and differences between countries in attitudes toward this novel area of science and technology will be explored in the rest of the chapter. Among the core issues studied for the first time, to the best of our knowledge, we can single out level of public awareness on pharming, views on animals and plants and their genetic modification for the production of medical drugs and of specific scenarios of pharming (both in terms of the means used, i.e., plants and different types of animals, and the goals or intended uses of the drugs to be obtained through genetic modification). A number of relevant worldviews (expectations and reservations about science, views of nature, images and beliefs about animals), trust in the regulators and canonical socio-demographic variables are also part of the survey, and will be used to explore the sources of the differences in attitudes to pharming. 30

The survey was based on a representative sample of the population aged 18 and over in fifteen countries: Austria, Czech Republic, Germany, Denmark, Spain, France, Ireland, Italy, Netherlands, Poland, United Kingdom, Sweden, United States, Japan and Israel. Information was gathered through 1,500 face-to-face interviews in each country using a multistage sample distribution stratified by region (NUTS or common classification of territorial units for statistics in the European Union that divides up the economic territory of the Member States or equivalent)/size of habitat, with primary units selected at random and individual respondents by the last birthday rule. The sampling error estimated is ±2.6 %, for a confidence level of 95.5 %. The survey was coordinated by TNS opinion, with fieldwork conducted between April and June 2007 and January and February 2008. The pharming module forms part of a broader study on Attitudes to Biotechnology prepared and funded by the BBVA Foundation in Spain. The conceptualisation of the questions on pharming in this study owed much to the Europäische Akademie project group authoring the present publication. Mariana Szmulewicz (BBVA Foundation Social Studies Department) assisted with the data analysis for this chapter and I wish to take this opportunity to express my thanks to her, to the Europäische Akademie and to the BBVA Foundation.

132


5.3.1 Awareness about pharming What occurs in the laboratory in a relatively early phase of a scientific development tends to go unnoticed by public opinion, except by minorities that vary in size from one country to the next. It is therefore interesting to have a simple measure of how much a given issue or scientific area catches the attention of society at a given time. This is what is known as “level of awareness” with regard to object x. Differences between countries and their fluctuations over time will suggest the level of salience the media and other social agents accord to a particular topic. At this early stage, the degree of awareness about the production of pharmaceutical drugs to treat human and animal diseases through the genetic modification of plants and animals, as was to be expected, tends to be low in all societies. The greatest awareness is found in Sweden (42 %), and it varies between 20 % and 35 % of the population (aged 18 and over) in the remaining survey countries (see table 5.1).

5.3.2 Evaluative perspectives That awareness is currently in the medium-low range does not mean the population cannot evaluate the implications and acceptability of pharming from different angles, though it does indicate that the evaluative target will be fuzzily perceived and that views of the same, especially for the majority subset of the non-aware, will rest more on general schema than on fine-grained judgments. This indeed is a common feature even with issues attracting wide media attention, including those of a political nature.31 In any case, we should not underestimate the power of general frames and schema in shaping attitudes towards specific objects (in our case, plant and animal pharming) on which the public is little informed. In its study of public perceptions of biotechnology, the European survey known as the Eurobarometer applies four evaluative angles to specific biotechnological applications; namely, “usefulness”, “morality” and “risk”, and the degree of support (“encouragement”) each should be given. Implicit in these four evaluative perspectives is the idea that acceptance or support for a given application will be a function of the first three criteria. The “Eurobarometer 58.0” (2002), one of the most widely analysed in the literature, employed a technique know as the “split ballot”32 to capture the public’s views on the following applications: in Split A, genetic testing, xenotransplantation and production of foods, and in Split B, crops, enzymes and clon31 32

Delli Carpini and Keeter 1996. The split ballot technique is the aplication to each half of a sample of one or more questions using different wording referring to a supposedly equivalent object, either to gauge the impact on responses of the way a question is worded or else for practical reasons of questionnaire cost or duration, allowing more items to be covered. In the case that concerns us each half of the sample was asked about different applications, summing six in all (three per split).


133

ing human cells.33 However as opposed to the model of three different evaluative angles, exerting clearly differentiated influences on the acceptance of each application, factor analysis reveals the existence of an identical structure: four factors in each split, three corresponding to the type of application (each grouping the properties “usefulness”, “morally acceptable” and “encouraged”) and a fourth factor, risk, constructed the opposite way, i.e., the common component or latent dimension is the evaluative criterion (risk), grouping all the applications. These results show the high specificity of judgments on each application, which rest on an evaluative framework formed out of the interrelation of the first three criteria. The other salient aspect – the independence of risk perception – exhibits a surprising profile. In effect, the perceived risk of biotechnology applications (which, according to the Eurobarometer, would move in the medium-high interval, with means of 2.6 or more points in a range from 1 to 4) was shown not to have any significant influence on the other three criteria and, particularly, on the belief that a given application should or should not be encouraged. Statistically it is a separate factor, its negative correlations with the other three criteria are very low and, in a multiple regression model to predict level of support, the “beta” coefficients of “risk” are very small, meaning “risk” has virtually no role in predicting the “encouragement” score of each application. Applications may be useful, morally acceptable and encouraged to diverse extents, but these differences of degree have very little to do with the scale of risk perceived.34 In the research on which the results in this chapter are based, the list of evaluative criteria was enlarged while acknowledging that several of them may form part of a more general evaluative structure. Another premise is that the judgments of a large part of the population are likely to rely on the “cues” about the general and/or the specific practical benefits given in the questions asked (“to produce pharmaceutical drugs for treating human diseases” or for treating specific diseases) as well as drawing on general beliefs 33

34

The specific items of the “Eurobarometer 58.0”, 2002 read as follows (the first three corresponding to Split A and the rest to Split B): Using genetic testing to detect diseases we might have inherited from our parents such as cystic fibrosis, mucoviscidosis, thalassemia. Introducing human genes into animals to produce organs for human transplants, such as into pigs for human heart transplants. Using modern biotechnology in the production of foods, for example, to make them higher in protein, keep longer or change the taste. Taking genes from plant species and transferring them into crop plants, to make them more resistant to insect pests. Using genetically modified organisms to produce enzymes as additives to soaps and detergents that are less damaging to the environment. Cloning human cells or tissues to replace a patient’s diseased cells that are not functioning properly, for example, in Parkinson’s disease or forms of diabetes or heart disease. Pardo and Calvo 2006b.

134


and worldviews. With these methodological cautions in mind, we can go on to offer an exploratory profile of evaluations and attitudes concerning plant and animal pharming. There is a clear consensus that the genetic modification of plants to obtain pharmaceutical drugs is a useful technique that should be supported and that this application is neither immoral nor reckless, even though it is perceived as going against nature. Perceptions of risk are more divided. The majority do not believe it is a product of the arrogance of scientists, but they do believe it is a product of the interests of a small number of pharmaceutical multinationals (see table 5.2). Finally, the majority trust the laws and controls established by each country’s government to regulate the growing of genetically modified plants in order to obtain pharmaceutical drugs, and are prepared to take drugs produced through this procedure (see table 5.3). In general, the perception that plant pharming should be supported correlates with the various facets or evaluative angles (table 5.4). The perception of usefulness is the dimension that evokes the highest level of correlation in all countries. In principle, this suggests that in the trade-off between means (genetic modification of plants) and ends (biomedical goals), the perceived utility will play a particularly significant role in inducing a compromise about the tools to be used, but all the other factors (such as moral considerations, worldviews and risk perceptions) will exert an independent influence, either reinforcing or diminishing the level of support. The interest of pharmaceutical multinationals in this technique, however, produces a lower level of correlation with its acceptance and, in contrast with the areas of genetic modification of crops and food, does not currently seem to be a salient component of pharming’s perception by the public. Although each of the countries studied behaves differently according to the dimension evaluated, there is a split between the opinions more favourable to plant pharming observed in the United States (mean value of disposition to support it, 6.4), Israel (6.4), the Czech Republic (6.3), Denmark (6.2), and the Netherlands (5.9), and the more guarded opinions found in Austria (4.2), Germany (4.6), and Japan (4.8). The perception that plant pharming is a technique involving risks is clearer in three countries where it achieves a relative majority: Austria, Japan and Germany, and in eight countries, the group that does not perceive important risks is larger than the group that does perceive them: this is the case in the Netherlands, Israel, United States, Denmark and, to a lesser extent, the Czech Republic, Poland, France and Spain. In the other countries, opinions are distributed more homogeneously. The following results are noteworthy: the high percentage of Austrians who declare they totally disagree with the idea that this technique does not involve important risks; the high percentage of Americans and Israelis who express total agreement; and finally, the significant percentage of intermediate or neutral scores among the Japanese (probably a cultural


135

style of responding, characteristic of that society and observed in many different surveys and contexts).35 The exception to the predominant medium-to-high level of trust in the laws and controls on plant pharming established by their national governments is found in Japan and Poland, whereas opinions are more divided in Germany and Austria. Danish, Spaniards, Dutch, Swedes and Israelis stand out for their declared trust in official controls and regulations. Finally, the majority or relative majority in 11 of the 15 countries included in the study would be prepared to take drugs produced from genetically modified plants. This willingness is greater (scores of 8–10) in Denmark and, some way behind in the Netherlands, Sweden, Great Britain, Czech Republic, Israel and the United States. In Austria and Japan, on the other hand, the majority would not be prepared to take a drug produced this way, whereas in Germany and France opinions are more divided. In contrast, the genetic modification of animals provokes a completely different response: a very broad consensus exists that it is a risky, immoral, reckless technique that goes against nature, will transmit diseases from animals to human beings and should not be supported. The majority consider it also to be incompatible with animal rights and believe that it will produce great suffering for animals. Opinions are divided, however, on the perception of the usefulness of the technique. Specifically, they are less homogeneous regarding scientists’ motives in developing the technique, and there is a clear consensus that it is a product of the interests of a small number of pharmaceutical multinationals (table 5.5). Views are also divided regarding the trust that citizens express in the laws and controls established by the governments of their countries to regulate this technique. Finally, the majority or the relative majority in most of the countries included in the study declare that they would not be prepared to take pharmaceutical drugs produced from genetically modified animals (table 5.6). As has been observed in regard to the genetic modification of plants to obtain pharmaceutical drugs, the perception of usefulness is, also in the case of animal pharming, the dimension with the greatest level of correlation in all countries, usually around 0.7. But, in this case, we find another perception of a similar magnitude in the opposite direction: the level of correlation with the view that it “is an unacceptable exploitation of animals” is also high in several countries, with values between 0.6 and 0.7 in 9 of the 15 countries studied. This suggests that, in this case, the usefulness of the goals being pursued weighs practically the same as concern about the instrumentalization of animals. In principle, and unless the specific nature of the medical goal ranks really high in perceived usefulness (such as treatments for life-threatening diseases), this dual perception could generate ambivalent attitudes. The other perceptions or evaluations of animal 35

Ladd and Bowman 1996.

136


pharming could either reinforce opposition or support. The correlation values of support for this type of pharming with the perception of immorality, risk, recklessness and with the ideas that it goes against nature, is incompatible with the inherent dignity of animals and with the view that it will cause them a great deal of suffering tend to be equal to or higher than 0.5 in all countries. As in the case of plant pharming, the dimension showing the lowest correlation with support for animal pharming is the view that this area “is a product of the interests of a small number of pharmaceutical multinationals” (between 0.2 and 0.4). Also, the views that it is “playing God”, that it “will transmit diseases from animals to humans” and that it is “a product of the arrogance of scientists” display relatively low levels of correlation with animal pharming support (see table 5.7). Factor analysis (principal component analysis), applied to the just-mentioned evaluative criteria in the consolidated sample of European countries, shows that the same structure is present in the cases of plant and animal pharming, with a single exception. In both cases this structure rests on two components. In the case of animal pharming, the first component is made up of nine reservation items (negative valence), with a variance explained of 55.80 %, while the second contains three promise items (with a positive valence, i.e., the items “should be supported”, it is “useful”, poses “no risks”) representing a modest 8.24 % of explained variance (total or cumulative explained variance with two components, 64.04 %). The composition of the evaluative structure of plant pharming is identical to that of animal pharming (two components with a 65.15 % total variance explained, breaking down 53.33 % and 11.82 % for the first and second component respectively). The only difference is that in this case the item on morality doesn’t neatly load on just one component, but has crossloadings of virtually the same magnitude on the two factors, suggesting that the notion of morality applied to the genetic modification of plants is not as clear as in the animal pharming domain.

5.4 A differentiated landscape of perceptions of pharming As we have seen, a majority of the population are prepared to accept a tradeoff between means (genetic modification of plants and animals) and goals. And the specific values assigned to one and the other (means and goals) help tilt the balance of attitudes towards approval or, alternatively, rejection. In effect, the fact of genetic modification to produce pharmaceuticals occurring in plants or in animals produces a first level of response differentiation. Then the type of plant or, especially, animal causes a further modulation, as we will later discuss. Views of pharming are also significantly influenced by the nature of the biomedical and social goals. The modulations that specific means and goals impose on attitudes to pharming exhibit

5.4 A differentiated landscape of perceptions of pharming

137

a similar structure in the case of plants and animals, but differ in intensity. In the case of plant pharming, approval predominates for most of the goals pursued and the type of plant used is not a strong enough motive to alter the valence of responses (generally positive). The space where plants are grown does, however, have some qualifying force. In the case of animal pharming, conversely, both the type of animal and the nature of the goals help determine whether attitudes fall on one or other side of the acceptance-rejection divide. The following sections present the public’s evaluations of plant and animal pharming, in sequence, as a function of the specificities of goals and means.

5.4.1 Ranking of biomedical and socio-economic goals and acceptance of plant pharming The biomedical purpose of the drugs is clearly a discriminatory factor in the evaluation of plant pharming (see table 5.8). Certain purposes noticeably increase acceptance, particularly finding treatments for life-threatening diseases, which is favourably evaluated in all countries. There is also a broad consensus regarding plant pharming to treat childhood diseases. Mean acceptance is over the midpoint in almost all countries (although with values somewhat below those found for the above purposes) in the case of obtaining antidotes or medicines to counter the effects of biological weapons and in that of obtaining vaccines for adults before they travel to areas where there is a risk of contracting certain diseases. On the other hand, opinions are more varied regarding purposes such as minor ailments, and the use of plant pharming to produce drugs for purposes such as lengthening years of life, delaying the effects of ageing and, even more so, for obtaining cosmetic products, is rejected by the majority of the population. This pattern varies markedly from one country to another. The citizens of Spain, Czech Republic, Poland, Israel and United States accept plant pharming in all cases with the sole exception of producing cosmetics. The highest acceptance scores are usually found in Spain, the United States and Israel. In contrast, the citizens of Germany and Austria show the most reservations regarding all scenarios: the Germans only express clear agreement with plant pharming in three scenarios (lifethreatening diseases, antidotes to biological weapons, diseases in children) whereas the Austrians disapprove across the board. In addition to medical goals, purposes of a specific socio-economic nature also play a role in the evaluation of plant pharming. In most countries, there is a tendency to accept plant pharming in order to obtain cheaper pharmaceutical drugs for the population of less developed countries and to eliminate the problems of shortages of certain kinds of drugs, whereas there is a pattern to reject its use for obtaining cheaper drugs for the population of advanced countries. Citizens of Spain, Denmark, Netherlands, Czech Republic, Poland, Israel and United States accept plant pharming in

138


all suppositions, whereas Italy and, more markedly, the citizens of Austria and Germany are the most critical, and disapprove of pharming in all the scenarios suggested (table 5.9).

5.4.2 The specifics of the means in the acceptance of plant pharming In addition to the specific purposes or goals, two particular aspects of the means – the kind of place where genetically modified plants to obtain pharmaceutical drugs are grown and the type of crop – further differentiate citizens’ views. Growing them in open fields is considered unacceptable by the majority in all countries. Extremely negative opinions stand out in Austria, France and Denmark. Growing them in open fields at a distance of several kilometres from other areas of crops or plants that have not been genetically modified is also considered unacceptable by the citizens of the majority of the countries, but with less intensity than when the distance from other kinds of crops is not specified. The group that considers this scenario unacceptable is larger than the group that considers it acceptable in 10 of the 15 countries contemplated. In the rest of the countries, opinions are distributed somewhat more homogeneously among the various levels of approval and rejection (this is the case in Spain, Italy, the Czech Republic, Israel and United States). Once again, the strongest rejection is recorded in France and Austria. Finally, growing genetically modified plants in completely enclosed precincts (such as greenhouses) gives a map that is completely different from that found in the previous suppositions. The segment that considers this scenario acceptable in all cases is larger than that holding the opposite opinion. The only exception is observed in Austria, a country where opinions are more divided. The group that expresses a wider and firmer acceptance of growing this kind of plant in closed precincts is noteworthy in the Czech Republic, Denmark, Israel and United States (table 5.10). While the main barriers to acceptance of animal pharming have to do with the ethical dimension of instrumentalizing and exploiting animals, in the case of plants they are more associated to fears about an eventual contamination of the environment and the food chain. This, in turn, explains the just-mentioned differences in attitudes to growing crops in one or other space. Regarding the possibility that genetic modification of plants may affect the environment or food safety, the majority express some degree of concern (table 5.11). Views on the environmental risk, in other words the possibility that growing genetically modified plants to obtain drugs could contaminate the environment, tend to be divided both across and within countries. In half of the countries included in the study, the segment that perceives risks (giving scores from 6 to 10) is larger than that which does not perceive them as clearly. Two countries, the Netherlands and the Czech Republic, exhibit the reverse trend. Perceptions are distributed more regu-


139

larly in the rest. The perception of risk is more intense (a higher percentage of scores from 8 to 10) in France, Germany and Austria. The percentage of intermediate responses in Japan and Spain is very significant. Finally, the importance of non-responses is noteworthy in many countries, with percentages of over 10 % in 7 of the 15 countries included in the analysis, and is specifically 20 % in Spain and Ireland. In the second supposition, the possibility that growing genetically modified plants in order to obtain pharmaceutical drugs could contaminate plants or seeds that are currently consumed as food, the perception of risk is seen to intensify. In 11 of the 15 countries included in the study, the group that perceives risks is clearly larger than the one that does not, whereas distribution in the four remaining countries is more equal. In this case, the perception of risk is, once again, particularly strong in France followed in second place by Germany and Austria. The percentage that cannot give an opinion on this subject is also significant, exceeding 10 % in 8 of the 15 countries and reaching almost 20 % in 3 of them (table 5.11). The kind of plant also influences acceptance of this technique, but with less intensity than the purpose or the place where it is grown (table 5.12). The use of tobacco leaves for the genetic modification of plants in order to obtain pharmaceutical drugs is accepted by the absolute or relative majority in 11 countries and rejected in 3 (Austria, France and Japan). Evaluations are distributed more homogeneously in Italy. The highest levels or thresholds of acceptance are observed among the citizens of the Czech Republic and the Netherlands. In the case of both genetically modified potatoes and corn, acceptance is higher than rejection in 9 of the 15 countries, although this trend is reversed in 5 (Germany, France, Italy, Austria and Japan). In the remaining country studied (Ireland), opinions are evenly distributed. Those who express most agreement with the use of genetically modified potatoes or corn to obtain pharmaceutical drugs are the citizens of the Netherlands, Czech Republic, Denmark and, some way behind, the United States and Israel. In contrast, the citizens that reject the growing of this kind of plant most strongly are the Austrians and Japanese. The highest level of discrimination in acceptance between the tobacco leaf scenario and that of the other two plants (potatoes and corn) is observed in Germany – where rejection increases by almost 15 points (reversing the tendency to approve) – and in Austria and Japan – where rejection increases 7 or more points from one scenario to another. In the remaining countries, the differences in terms of acceptance or rejection among the three scenarios tend to stand below 4 percentage points. A significant percentage of the population ventures no opinion regarding each of the plants surveyed: over 10 % in almost all countries and even above 15 % in several cases. The scenario that obtains the highest rate of non-responses in the one referring to genetically modified tobacco leaves.

140


5.4.3 Ranking of biomedical and social goals and acceptance of animal pharming The purpose of the pharmaceutical drugs is also a discriminatory factor in acceptance levels for the genetic modification of animals, although here the variations arise within a context of widespread rejection. The only purpose that activates a moderate level of acceptance in almost all countries is the treatment of very serious diseases and, in some countries, to treat childhood diseases. Rejection is the majority reaction to the rest of the suppositions (table 5.13). Furthermore, in almost all societies, the majority disapprove of genetic modification of animals to obtain cheaper pharmaceutical drugs for the population of less developed countries, to eliminate the problems of a shortage of some types of drugs and, even more so, to obtain cheaper drugs for the population of advanced countries (table 5.14). This is an indirect indication that when the means to be used are the genetic modification of animals, there is strong rejection in virtually all biomedical and social scenarios. The general map of trends shows that the citizens of Spain, Czech Republic, Poland, Israel and Japan express a mean acceptance score above the midpoint in a larger number of scenarios (between 4 or 5 of the 10 scenarios presented). On the contrary, the citizens of Austria and France tend not to accept this technique under any circumstances, and in Germany only life-threatening diseases reaches a minimal level of acceptance.

5.4.4 The specifics of the means in the acceptance of animal pharming The animal to be used and the medium in which the drug is produced also modulate the evaluation of genetic modification of animals (see table 5.15)36. In a context of overall rejection of this technique in almost all the societies evaluated, the use of animals like fruit flies and, in second place, mice generates moderate acceptance in various countries (the former animal in 9 of the 15 countries and the latter in 7). Nevertheless the mean acceptance scores are moderate in most cases. They tend not to exceed 6 points and to be located close to the midpoint (5) of the scale. The use of fish inspires widespread rejection in the majority of the countries and mean acceptance scores only reach the midpoint of the scale in Denmark, Spain, Czech Republic and Israel. Mean acceptance of the use of hens and rabbits approaches the midpoint only in Spain. In the remaining countries, acceptance of the use of these three types of animals tends to be under 4 points. Rejection is accentuated even more when we enquire about the genetic mod36

The literature on the use of animals in scientific research has found that the species used is among the main predictors of acceptance or rejection, with higher acceptance in the case of small rodents (considered pests) than in that of larger animals (some of them viewed as pets and other as endangered species close to humans) (see Hagelin et al. 2003; Crettaz von Roten 2008).


141

ification of sheep, cows, pigs and chimpanzees. In several countries (Germany, France, Austria, Ireland, Italy and Sweden), the mean score for these scenarios is less than 3 points. Relatively higher mean acceptance scores for all the scenarios presented can be observed among the citizens of Spain and the Czech Republic. In contrast, the most critical opinions tend to be concentrated among the citizens of Austria, France and Germany. A more precise picture of the evaluation of animal pharming can be obtained from the overall distribution on the agreement scale divided into five segments, which confirms as well as clarifies the trends observed based on mean values (table 5.16). In the case of fruit flies, in 8 countries the segment that accepts their use is larger than that rejecting it. The reverse is true in 4 countries, while in the remaining 3 opinions are more widely distributed. Acceptability scores also vary regarding the genetic modification of mice, which is accepted in 6 countries and rejected in 8, with opinions more evenly distributed in the one country remaining. The genetic modification of fish is only accepted by the relative majority in three countries (Denmark, Spain and the Czech Republic), with the majority in 9 countries against and replies more evenly distributed in the remaining 3. Genetic modification of the other animals considered in the survey (hens, rabbits, sheep, cows, pigs and chimpanzees) causes an even clearer and stronger rejection in almost all societies. Only in Spain is the group that considers some cases acceptable larger than the group that considers them unacceptable. On analysing the intensity of support or rejection, we find that the extremes of acceptance (scores from 8 to 10) tend to be observed among the citizens of Denmark, the Czech Republic and Israel. In contrast, the strongest rejection (scores from 0 to 2) tends to characterize citizens of Austria, Germany, France and Sweden. The Japanese and Spaniards, finally, are distinguished by their greater predisposition to give intermediate scores (5). In general, this is not a question where people have difficulties giving an opinion and non-response figures tend to be below 10 %, although they stand out slightly in Ireland. Within this context of general disfavour regarding the genetic modification of animals to obtain pharmaceutical drugs, rejection also varies in intensity according to the medium from which the protein is taken to produce the drugs (table 5.17). Rejection decreases slightly when the protein is obtained from the milk or eggs of the genetically modified animals compared to when it is obtained from their urine or blood. In any case, mean acceptance of drugs obtained from milk or eggs produced by genetically modified animals only exceeds the midpoint on a scale from 0 to 10 in Denmark, Spain, Czech Republic and Israel. In the Netherlands, Great Britain, Poland, United States and Japan, it varies from 4 to 5 points, whereas in remaining countries it stands below 4 points. The mean

142


acceptance of pharmaceutical drugs obtained from the urine or blood of genetically modified animals is between 4 and 5 points in six countries and lower in the other nine. The lowest mean acceptance scores are observed in Austria followed by France, Germany and Ireland. An examination of the distribution of the replies in five segments, according to their intensity of support or rejection, confirms that the segment that accepts the obtaining of pharmaceutical drugs from the eggs and milk of genetically modified animals is only larger than the segment that considers it unacceptable in Spain, the Czech Republic and Israel (table 5.18). In turn, the segment that considers obtaining drugs from the blood or the urine of these animals unacceptable increases in comparison with the two previous scenarios. With the exception of the Netherlands, Denmark and the Czech Republic, where the groups declaring themselves for and against have a similar significance, an absolute or relative majority in remaining countries is against both suppositions. In terms of intensity of rejection, Austrian citizens again stand out with a stronger rejection score (0 to 2). At the opposite extreme, the citizens of Israel, Denmark and Czech Republic tend to express higher levels of acceptance (scores from 8 to 10).

5.5 Preferences for methods of production of pharmaceuticals After examining the acceptance of pharming according to various suppositions, respondents were asked which methods of production they would prefer in order to obtain a certain medicine with an identical composition regardless of the procedure used. The production methods suggested as alternatives included the isolation of the drug from species of wild plants, through chemical synthesis, through the genetic modification of animal cells in culture, and through the genetic modification of plants or animals. The hierarchy of preference for methods of producing a medicine with identical characteristics is consistent with the views on pharming previously analysed. Responses gave the following ranking. First, in all countries, the majority opt in first place for the isolation of the drug from species of wild plants. This preference is particularly marked in France, Denmark and Japan, where more than 70 % of the population is in favour of this method. The second method selected by citizens in all countries is that of chemical synthesis. The percentage that mentions it in first place is near 20 %, with the exception of France and Japan where it is significantly lower. Conversely, in Italy and Sweden, though a relative majority also prefer isolation from wild plant species, the percentage favouring production of the medicine through chemical synthesis is very significant. Third, it is revealing that the genetic modification of animal cells in culture is mentioned by

5.6 Awareness and acceptance of plant and animal pharming

143

a very low percentage and the percentage supporting genetic modification of animals is much smaller, almost non-existent (table 5.19). The analysis of the method respondents would prefer in second and third place gives the following hierarchy: after isolation of the medicine from wild plant species comes production through chemical synthesis and, in third place, through the genetic modification of plants. The difficulty in selecting three methods of preference is significant and many respondents cannot select more than one or two. Analysis of total mentions confirms this hierarchy, and shows that those making most mention of genetic modification of plants somewhere in their order of preference are the Dutch and Danish and, in second place, the Czechs, Swedish, British and Americans. And those most frequently including a preference for the modification of animal cells in culture are the Israelis, the Dutch and the Italians. Finally, although the percentage of preference is low, Israelis stand out in selecting the genetic modification of animals in some place in their order.

5.6 Awareness and acceptance of plant and animal pharming One point widely debated in the literature, as stated in the opening section of this chapter, is whether the population subset that is more informed about science is also more favourably disposed towards it. In recent years, interest has focused on the relations between the “aware” public for a given development and attitudes to the same. Generally speaking, our data show that those who have previously heard or read information about the genetic modification of plants in order to obtain pharmaceutical drugs tend to highlight the positive facets of this technique to a greater extent and its negative aspects to a lesser extent. Of all the indicators of attitudes to pharming, perhaps the one capturing the most general or holistic perception is “declared level of support”, and the pattern of responses for this variable will also be presented in the following sections for most other evaluations of pharming (usefulness, morality, etc.). In 11 of the 15 countries included in the study, the mean agreement score on support for plant pharming among those having previously heard or read information was between half a point and one point above the mean score of the segment having no contact with such information prior to the survey (see table 5.20). In the case of animal pharming, having previously heard or read information about the genetic modification of plants and animals in order to obtain pharmaceutical drugs reduces rejection of this application to some extent. The mean score for agreement that it is an application which should be supported tends to increase by approximately 1 point among those having previous information, although it only falls above the midpoint on the

144


scale in four countries: Denmark, Spain, the Czech Republic and Israel (table 5.21). An additional test of this association between “awareness of pharming” and “positive views on pharming” was obtained through a more specific question about the creation of genetically modified hens whose eggs contain medical proteins. Generally speaking, the data show that the low percentage that has previously heard or read information about it express a greater acceptance of this application than those who have not heard or read anything about it. In most countries, the mean acceptance of this application among those who have previously heard or read some information is more than a half point, and in some countries over 1 point, above the mean observed in the segment lacking any prior information. Therefore, among those who have previously heard or read information, mean acceptance on the scale from 0 to 10 reaches or exceeds (though in some cases very slightly) the midpoint in all countries except Germany, France, Austria and Italy (table 5.22). Awareness, it seems, is related to more sympathetic views even of the type of pharming most opposed by the public, i.e., animal pharming (although with the qualification of producing drugs for serious diseases). It is an open question at this stage if providing more information to both the aware and the non-aware public would have the same or different effects (in this last case, it would be suggestive of the early aware public having a different profile, more in favour of these applications).

5.7 Elements of an explanatory model Although at this early stage we cannot expect to find specific, consolidated attitudes on a novel object like pharming, what we do have is an ample structure inside which perceptions or views are formed about this scientific-technological area. As with any complex object, the influences and pointers are multiple. In the stylized model discussed earlier, a tradeoff is apparent between the means used and the nature of the goals to be achieved. In general, the means, i.e., biotechnology, and much to the surprise of the researchers in the area, are (for the most part) negatively perceived, while the ends (biomedical applications) receive different levels of support depending on their specific nature. Regarding the means, in a context of modest appreciation of biotechnology, the modification of the genetic blueprint of plants is accepted to a certain extent, provided the goals are seen as worth pursuing, whereas in the case of animals the dominant pattern is one of resistance and opposition, only turning neutral or positive when the goal is of critical importance. Plants and animals are perceived in most cultures as occupying different levels in the ranking of life forms, with animals, in principle, especially pro-

5.7 Elements of an explanatory model

145

tected as being closer to humans (on the perception of animals, see below). A hierarchy also emerges with regard to biomedical goals, with minor ailments and life enhancing treatments situated at the bottom, and the treatment of life-threatening or childhood diseases occupying the top places. For the majority of the population, it is this balancing of specific means and specific goals which shapes either a positive or a negative response to pharming. Now there are some other factors at play in shaping attitudes to this scientific development.37 One of them, as we have just seen, is level of awareness, with the aware segment more in favour of both types of pharming. Immediately the question arises if more knowledgeable people, all else being equal, also take a more positive view of pharming. A test of elementary biological knowledge comprising 13 items (range of correct responses 0–13), with scores assigned to one of 3 groups (low knowledge: scores 0–4, medium knowledge: 5–9, high knowledge: 10–13), finds significant differences, in the expected direction, between knowledge of and support for pharming, of a magnitude in the medium-low range.38 In the specific case of plant pharming, a one-way ANOVA gives an F(2, 17497) value=77.10, p

Pharming: Promises and risks of Biopharmaceuticals derived from genetically modified plants and animals (Ethics of Science and Technology Assessment)

Genetically Modified Plants

Nanotechnology: Assessment and Perspectives (Ethics of Science and Technology Assessment)

Business Ethics of Innovation (Ethics of Science and Technology Assessment)

Genetically Modified Pest-Protected Plants: Science and Regulation

Environmental Risk Assessment of Genetically Modified Organisms

Genetically Modified Planet: Environmental Impacts of Genetically Engineered Plants

Science and Technology Ethics

Genetically Modified Crops: Their Development, Uses and Risks

Information und Menschenbild (Ethics of Science and Technology Assessment, 37)

Genetically Modified Athletes: Biomedical Ethics, Gene Doping and Sport

Introgression from genetically modified plants into wild relatives

Encyclopedia of Science Technology and Ethics

Encyclopedia of Science Technology and Ethics

Encyclopedia of Science Technology and Ethics

Encyclopedia of Science Technology and Ethics

The Encyclopedia of Science, Technology, and Ethics

Encyclopedia of Science Technology and Ethics

Encyclopedia of Science Technology and Ethics

Encyclopedia of Science Technology and Ethics

Encyclopedia of Science Technology and Ethics

Genetically Modified Food and Global Welfare

The GMO Handbook: Genetically Modified Animals, Microbes, and Plants in Biotechnology

Genetically Modified Crops

Innate Immunity of Plants, Animals and Humans

Plant Biotechnology: Current and Future Applications of Genetically Modified Crops

Environmental Risk Assessment of Genetically Modified Organisms: Vol. 4: Challenges and Opportunities with Bt cotton in Vietnam (Environmental Risk Assessment of Genetically Modified Organisms)

Environmental Impact of Genetically Modified Crops

Testing of Genetically Modified Organisms in Foods

Genetic Engineering in Livestock: New Applications and Interdisciplinary Perspectives (Ethics of Science and Technology Assessment)

Pharming: Promises and risks of Biopharmaceuticals derived from genetically modified plants and animals (Ethics of Science and Technology Assessment)

Genetically Modified Plants

Nanotechnology: Assessment and Perspectives (Ethics of Science and Technology Assessment)

Business Ethics of Innovation (Ethics of Science and Technology Assessment)

Genetically Modified Pest-Protected Plants: Science and Regulation

Environmental Risk Assessment of Genetically Modified Organisms

Genetically Modified Planet: Environmental Impacts of Genetically Engineered Plants

Science and Technology Ethics

Genetically Modified Crops: Their Development, Uses and Risks

Information und Menschenbild (Ethics of Science and Technology Assessment, 37)

Genetically Modified Athletes: Biomedical Ethics, Gene Doping and Sport

Introgression from genetically modified plants into wild relatives

Encyclopedia of Science Technology and Ethics

Encyclopedia of Science Technology and Ethics

Encyclopedia of Science Technology and Ethics

Encyclopedia of Science Technology and Ethics

The Encyclopedia of Science, Technology, and Ethics

Encyclopedia of Science Technology and Ethics

Encyclopedia of Science Technology and Ethics

Encyclopedia of Science Technology and Ethics

Encyclopedia of Science Technology and Ethics

Genetically Modified Food and Global Welfare

The GMO Handbook: Genetically Modified Animals, Microbes, and Plants in Biotechnology

Genetically Modified Crops

Innate Immunity of Plants, Animals and Humans

Plant Biotechnology: Current and Future Applications of Genetically Modified Crops

Environmental Risk Assessment of Genetically Modified Organisms: Vol. 4: Challenges and Opportunities with Bt cotton in Vietnam (Environmental Risk Assessment of Genetically Modified Organisms)

Environmental Impact of Genetically Modified Crops

Testing of Genetically Modified Organisms in Foods

Genetic Engineering in Livestock: New Applications and Interdisciplinary Perspectives (Ethics of Science and Technology Assessment)

Recommend Documents