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...

108 downloads 734 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 Europ¨aische 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 Europ¨aische 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 Europ¨aische Akademie. The Series The series Ethics of Science and Technology Assessment (Wissenschaftsethik und Technikfolgenbeurteilung) serves to publish the results of the work of the Europ¨aische 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

Comprehensive table of contents

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

Comprehensive table of contents

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

Comprehensive table of contents

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

Comprehensive table of contents

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

Comprehensive table of contents

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

2 The technology of pharming

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

2 The technology of pharming

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

2.2 Plants as a production platform for recombinant biopharmaceuticals

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

2 The technology of pharming

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.

2.2 Plants as a production platform for recombinant biopharmaceuticals

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

2 The technology of pharming

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.

2.2 Plants as a production platform for recombinant biopharmaceuticals

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 (<1 µm), for example gold particles, which are accelerated towards the target plant tissues. The tissue could for example be a plant callus or leave discs. A plant cell callus consists of somatic undifferentiated cells from an adult plant that has the ability to differentiate. The bombardment devices use a sudden release of compressed helium gas to accelerate the DNA-coated particles. After the particles pass through the plant cell wall, they enter the cytoplasm and preferably the nucleus, where the DNA comes off the particles and integrates into the genome. Depending on the host species, several physical and biological parameters have to be adapted to achieve transformation. Physical parameters include the nature, chemical, and physical properties of the particles; the nature, preparation,

18

2 The technology of pharming

and binding of DNA onto the particles; and the characteristics of the target tissue. Biological factors include choice and nature of explants, pre-and post-bombardment culture conditions, and interactions between the introduced DNA and cytoplasmic or nuclear components. After the bombardment transformed cells have to be screened. Based on the number of bombarded explants, the overall transformation frequency can be for example as high as 15 percent with germline transformation frequencies approximating 0.25 percent23. The screening and testing for successful integration and expression of new DNA is performed by molecular techniques, such as a using tandem selectable gene markers or molecular analysis methods such as northern blots. Selected single cells from the callus or leave disc can be treated with a series of plant hormones, such as auxins and gibberellins, which are capable to induce the re-differentiation into entire plants. This capability of total re-generation is called totipotency. The new plant that originated from a successfully shot cell may have new genetic (heritable) traits. Their ability to segregate in a Mendelian fashion in the next generations is tested in the next step. Then the families that express the transgene in the desired manner are screened. The choice of the transformation method depends largely on the plant species that is intended to be transformed. The soil pathogen Agrobacterium tumefaciens provides a simple method of transforming most dicotyledon24 plant species and is commonly used for pharming in tobacco, alfalfa, pea tomato and potato25. Some selected monocotyledons (i.e. grasses) can also be transformed by Agrobacterium, but in most monocotyledons particle bombardment is the preferred method. This is the case for cereals, such as rice, wheat, and maize, but also for soybean and other legumes26. Particle bombardment is also necessary, if the target tissues are plastids, as the Agrobacterium T-DNA complex is targeted to the nucleus and therefore unsuitable for gene transfer to chloroplasts. These transformation methods generally lead to the introduction of surplus DNA sequences into the genome of the host. In the case of Agrobacterium-mediated transformation, this is because inefficient processing of the T-DNA border sequences often results in cotransfer of flanking sequences27. In the case of particle bombardment, superfluous DNA transfer occurs because whole plasmids are generally used to coat microprojectiles28. Superfluous DNA transfer is a safety as well as a regulatory issue, when the result23 24

25 26 27 28

Sanford et al. 1987; Sanford 1988. Dicotyledons, or “dicots” are a flowering plants whose seed contains two embryonic leaves or cotyledons. Monocotyledons – in contrast – typically are having one embryonic leaf. Giddings et al. 2000. Ma et al. 2003. Gelvin 2003. Klein et al. 1992.

2.2 Plants as a production platform for recombinant biopharmaceuticals

19

Table 2.3: Particle bombardment mediated transformation of plants



Gene Gun: When the fire switch on the outside of the chamber, helium is released at high pressure. The blast ruptures a first disk and a shock wave is induced. This wave hits another disk, which is free to move. Attached to the front of that disk are particles 1 micron in diameter coated with DNA molecules. This disk travels another centimeter at the speed of a gun projectile, roughly 400 meter per second, and hits a screen which detains the disk but liberates the DNA coated particles toward the target cells.

Transformed cells are selected with selectable marker or by screening for the transgene with molecular methods. Selected single cells form a plant cell callus.

 Transgenic plant regeneration from plant cell callus and analysis of the transgenic plant lines.

ing genetically modified organisms (GMO) are to be released into the environment. Recently, several strategies to avoid the cotransfer of flanking sequences have been developed. One example is the incorporation of the barnase gene in the construct of Agrobacterium mediated transformation29. This gene ensures that all plant cells that contain sequences linked to the T-DNA are killed, as barnase gene expression is lethal. The barnase gene is derived from the soil bacterium Bacillus amyloliquefaciens and codes for a ribonuclease. In cases where flanking regions have undeliberatly incorporated into the genome of the plant, the barnase ribonuclease is synthesized. After activation the ribonuclease destroys the vital RNA molecules of the plant cell and is therefore lethal. By the addition of the barnase gene 29

Beals and Goldberg 1997; Kuvshinov et al. 2004.

20

2 The technology of pharming

in the flanking regions only plant cells survive that have not incorporated these additional pieces of DNA. One side effect of these new methods is that the transgenic loci are considerably simpler than those of whole-plasmid transformants, and the plants show a notable reduction in the frequency of transgene silencing.

2.2.2 Transient expression using viral vectors In addition to the incorporation of a transgene into the plants genome, a foreign gene can also be expressed transiently. That means that the transgene has not been physically incorporated into the genome but is carried as an episome that can be lost. In pharming genetically modified plant viruses have been applied to whole-plant systems for the production of transient biopharmaceutical expression30. In these cases the plant itself is not transformed with the gene for the biopharmaceutical, but a plant virus that uses the plant to propagate. This means that expression levels will not be constant over time, and will eventually fall away. It also means that the plant does not inherit the ability to produce the biopharmaceutical but the plant virus. Foreign proteins produced using viral vectors can be in the form of free cytosolic proteins or fusions to viral proteins. Viral expression systems exploit the ability of viruses to propagate rapidly and achieve high concentrations in plant tissues. For example, tobacco mosaic virus (TMV) can accumulate in infected tobacco leaves to levels greater than 60 mg/g dry weight and produce amounts of TMV coat protein accounting for 10–40 % of the total protein content of the leaves31. Provided the movement proteins on recombinant viruses remain functional, viral vectors are able to spread throughout the entire plant from a single infection point via the plasmodesmata between individual cells and the vascular system. Therefore, in principle, when foreign protein is co-expressed with the plant virus, large amounts of product can be found. However, currently the application of transgenic viruses in whole plants has not resulted in the production of foreign proteins to the same high levels as the viral proteins from non transgenic virus infections. This is probably because the genetic construct carried by the virus interferes to some extent with the normal folding, packaging, transmission and replication process. Nevertheless, foreign protein yields achieved using viral vectors can be substantial. For example, transgenic viruses with coat protein fusions have been reported to accumulate to levels of 1–3 mg/g plant tissue32.

30 31 32

Grill et al. 2005. Shadwick and Doran 2007. Shadwick and Doran 2004.

2.2 Plants as a production platform for recombinant biopharmaceuticals

21

2.2.3 Choice of species and site of production The choice of the expression platforms in plant pharming depends on a combination of factors, including environmental conditions of the intended growing area, local expertise and intellectual property issues. In addition factors that must be considered when choosing a production crop include biomass yield per hectare, yield of the recombinant protein per unit biomass, ease of transformation and scalability. Also it needs to be taken into account whether the biopharmaceutical is intended to be consumed together with the plant (for example edible vaccines) or whether it should be stored before isolation (for example in the grain) whether the biopharmaceutical is intended to be isolated immediately after harvest (for example from leafs). The following expression organs can be differentiated: leafs, seeds and fruits or vegetables. In addition to the utilization of whole plants, fermenter grown plant cell suspensions or root cultures are applied in plant pharming. 2.2.3.1 Leaves

Tobacco (Nicotiana tabacum) has an established history as a model system for pharming and is the most widely used species for the production of recombinant pharmaceutical proteins at the research laboratory level33. It is therefore one of the strongest candidates for the commercial production of recombinant proteins. The major advantage of tobacco includes the well-established technology for gene transfer and expression, high biomass yield, prolific seed production and the existence of a large-scale processing infrastructure. Because tobacco is neither a food nor a feed crop, there is little risk that tobacco material will contaminate either the food or the feed chain. Although many tobacco cultivars produce high levels of toxic alkaloids, low-alkaloid varieties are available that can be used for the production of pharmaceutical proteins34. Alternative leafy crops that are being investigated for pharming include alfalfa, soybean and lettuce. Alfalfa and soybean have the major advantage of using atmospheric nitrogen through symbiotic nitrogen fixation, which therefore reduces the need for chemical fertilizer. Alfalfa is in particular useful because it has a large dry biomass yield per hectare and can be harvested up to nine times a year. Both of these legumes have been used to produce recombinant antibodies. Lettuce is also being investigated as a production host for edible recombinant vaccines and has been used in one series of clinical trials for a vaccine against hepatitis B virus35. One great disadvantage of leafy crops is that recombinant proteins are synthesized in an aqueous environment and are often unstable, resulting in low yields. The leaves must be frozen or dried for transport, or processed soon after harvest to extract useful amounts of the product. 33 34 35

Richter et al. 2000. Ma et al. 2003. Twyman et al. 2003.

22

2 The technology of pharming

2.2.3.2 Cereals, legume seeds and oilseeds

Numerous crops have been investigated for seed-based production, including the cereals rice, wheat oilseed rape and maize; the legumes pea and soybean36 and the oilseed sallower37. The advantage of the expression of proteins in seeds over leafy crops is the possibility to long-term storage, even at room temperature, because seeds have the appropriate biochemical environment to promote stable protein accumulation. Cereal seeds also lack the phenolic compounds present in tobacco leaves, thus improving the efficiency of downstream processing. However the overall yields of recombinant proteins in seed crops are much lower than in tobacco, and the most appropriate system must be determined on a case-by-case basis. 2.2.3.3 Fruits and vegetables

A major aim of protein expression in fruit and vegetable crops is that the edible organs can be consumed as uncooked, unprocessed or partiallyprocessed material, making them suitable for the production of recombinant vaccines and antibodies designed for topical applications. Potatoes are the major system for vaccine production. Tomatoes are more palatable than potatoes and have other advantages including high biomass yield and the use of greenhouses increases containment38. One of the main aims to develop edible vaccines is the fact that they could be distributed without refrigeration, which could be of importance – especially in developing countries. Bananas are feasible vehicles for edible vaccine distribution – but to date problems of inconsistent expression levels in the fruits due to different growth conditions and the resulting inconsistent doses are not yet solved39. 2.2.3.4 Plant cell cultures and hairy root systems

In some cases fermenter grown large-scale plant cell cultures or hairy root cultures might offer an alternative route for recombinant protein production40. These production platforms can be chosen in situations when the production of biopharmaceuticals in field grown plants comes along with too many obstacles: In cases where the foreign protein is toxic, for example to soil microorganisms or to wild-life capable of consuming the plants, issues of environmental safety might be too hard to solve. In addition agricultural production of delicate proteins might fail to provide adequate assurance of product safety and quality, for example when weather and soil 36 37 38 39 40

Burkhardt et al. 1997; Stöger et al. 2000; Azzoni et al. 2002. Cory et al. 2006. Conrad and Hain 2005. Lal et al. 2007. Shadwick and Doran 2007.

2.2 Plants as a production platform for recombinant biopharmaceuticals

23

conditions are hard to control or the fields are subject to contamination with pesticides, herbicides and mycotoxins. Even when a very important advantage of agricultural plant pharming – the cheap and faster scale up procedure – is lost in this production platform, plant tissue cultures and root cultures still offer a number of advantages over currently utilized mammalian cell cultures: One important factor is that plant cell culture media are relatively simple in composition and less susceptible to contamination with undesirable organisms, which leads to reduced material and process costs. Another advantage is that plant culture media do not contain proteins – thus the recovery of the desired protein is easier and cheaper than mammalian cell cultures. In addition as most plant pathogens are unable to infect humans or animals, the risk of pathogenic infection being transferred from cell culture via the product is minimized. Plant cell cultures comprise small aggregates of undifferentiated plant cells in liquid nutrient medium. Dedifferentiation and the promotion of growth are achieved by the addition of plant hormones to the medium. Tobacco has been the species most studied but rice (Oryza sativa) has also been utilized by several groups41. The predominance of tobacco in plant cell cultures differ from the situation of transgene expression in whole plants, where advantages associated with producing the pharmaceutical protein in edible species or in different storage organs such as seeds have resulted in a variety of species being transformed (see table 2.1). Obstacles in plant cell culture are that they are often subject to genetic instability that might cause significant reduction in yield of the foreign protein over time and that to date few plant cell cultures have shown to accumulate or secrete biopharmaceuticals at concentrations sufficient for commercial viability42. Hairy root cultures are a second system for protein production in culture (illustrated in table 2.4). They comprise roots that can grow independently of the plant in liquid nutrient medium due to action of phytohormones. They are called “hairy” because their morphology is more branched and thinner than roots adherent to the plants. In contrast to plant cell cultures exogenous plant hormones are not applied to the medium, but are synthesized by the roots themselves after they have been transformed with the bacterium Agrobacterium rhizogenes. These bacteria transmit the genes that code for the phytohormones to the host plant that afterwards produce hairy roots. Biopharmaceutical expressing hairy roots can be obtained by the infection of transgenic plants with Agrobacterium rhizogenes or can be produced by performing root initiation and transformation with the recombinant gene at the same time using genetically modified Agrobacterium rhizogenes with the transgene inserted into the plasmid construct, that is as described transmitted to the host plant. When cultured in liquid medium, hairy roots often 41 42

Kim et al. 2008; Lee et al. 2007; Jung et al. 2006. Shadwick and Doran 2004.

24

2 The technology of pharming

Table 2.4: Hairy root cultures: establishing and manufacturing Hairy root formation, after infection with Agrobacterium rhizogenes. The plant pathogen induces tumour-like hairy root formation, by transforming the plant with genes for tumour inducing plant hormones. Transformed roots are selected by the addition of antibiotics, to liberate the roots from the plant pathogen. They therefore become sterile and are able to grow in liquid media.

After the transformation of the hairy roots with the desired gene for biopharmaceutical expression, the hairy root cultures are scaled up for the growth in production vessels.

Source (pictures): ROOTec, Basel

exhibit rapid growth relative to untransformed roots, due to plant hormones. They can be propagated indefinitely in liquid medium and have been found to have significantly greater long-term stability than plant cell cultures.

2.2.4 Cultivation While plant cell cultures or hairy root systems are propagated in contained environments (vessels) the large-scale cultivation of transgenic plants is carried out in greenhouses or, for reasons of economy and scalability, in the field. A series of varying environmental factors must be taken into account at this step. They include light intensity, temperature, water regime, soil quality, the kind of fertilization, the presence of pests and the substances utilized to treat them. These factors are important on the level of the quantity of the yield but also on the level of its quality (See section 2.4). For containment strategies see chapter 3.

2.2.5 Purification of biopharmaceuticals from transgenic plants 2.2.5.1 Puriﬁcation of biopharmaceuticals from whole plants

Besides the level of expression of a recombinant protein the subsequent purification steps are crucial for the resulting yields. The process includes the initial processing of the source material (i.e. grinding of the grain), extraction, capture, intermediate purification and polishing (see also figure 2.2). The

2.2 Plants as a production platform for recombinant biopharmaceuticals

25

key issues in these steps are removal of host cell proteins and nucleic acids, as well as other product or process related or adventitious contaminants43. After harvesting, leaves and other soft tissue begin to degrade. Therefore the conditions and duration of storage before initial processing are critical. In seed based production systems this step is less important, since long-term storage is possible. Therefore in seed based systems it is possible to separate the place and time of harvesting and the initial processing of the material. Once the material is harvested and introduced to further processing, all further steps must be performed according to the regulations of pharmaceutical GMP (good manufacturing practice). That includes the need for equipment qualification and process validation. Since the expression level and activity will depend on the environmental conditions and plant health for field-grown plants a validation of the process will include the definition and separation of batches. The maintenance of a batch to batch consistency can not always be achieved fully. The situation is different for plant cell cultures cultivated in bioreactors, since growth conditions are controlled in this case and batch variance can therefore be minimized. The initial processing of the plant material is routine, because machinery, for example corn mills, leaf shredders etc., are already available from suppliers from food and feed industry. The design of the initial processing step is largely dependent on the source material. The most commonly used target tissues for the expression of recombinant proteins are leaves and seeds. Leaf material contains a lot of water. For the extraction of the desired protein from leaf material it can be homogenized and extracted in its own juice. The addition of extraction buffer is mainly needed to control pH and to introduce protein-stabilising agents. Seeds, on the other hand, must be subjected to dry milling followed by extraction in a large proportion of buffer. Before purifying the target protein from the crude plant extract, a clarification step is required to separate particulate material from liquid phase. In the case of leaves as row material, leaves are disrupted by shredding. Therefore fibres and other fine materials are inevitably generated. Even after removal of bulk cell for example by centrifugation or material depth filtration a portion of fine green material will remain that could in the following chromatographic purification block the columns. Thus an additional step of cross-flow microfiltration has to be incorporated in the purification process. After extraction the following steps of capture, intermediate purification and polishing of the target protein are basically performed by chromatographic purification. All supplementary procedures like extraction, centrifugation and filtration ultimately serve to condition the protein for chromatography. A series of chromatographic steps, usually termed capture, intermediate purification and polishing, making use of different intrinsic features of proteins, is usually required to archive sufficient separation 43

Drossard 2004.

26

2 The technology of pharming

of the target from contaminants44. Common modes of chromatography include ion exchange chromatography, affinity chromatography, hydrophobic interaction chromatography, gel filtration and to a limited extend, reverse phase chromatography45. The whole extraction process contributes significantly to the overall costs of the production of biopharmaceuticals. A majority of product-specific requirements for the purification from the raw material are not associated with the particular expression system, leading to the conclusion that potential economic advantages of plant production systems lie in the upstream rather than the downstream (purification) part of the process. The large contribution of the downstream costs to the overall cost of the process will always put pressure on the this side of the process In the whole process of purification, expression specific risks for product quality and safety must be adequately taken into account. In contrast to established host systems like Escherichia coli and mammalian cells, the regulatory requirements for plant based pharmaceuticals are not fully defined. However, both the US Food and Drug Administration (FDA) and the European Agency for the Evaluation of Medicinal Products (EMEA) have recently published draft guidance documents addressing this issue46. 2.2.5.2 Purification of biopharmaceuticals from plant cell cultures and hairy root cultures

Purification of biopharmaceuticals from plant cell cultures follows the same pattern as the purification of recombinant proteins from field grown plants: initial processing of the source material, extraction, capture, intermediate purification and polishing. However often in cell cultures many of these steps can be circumvented by an ingenious molecular strategy: Proteins produced in plant cells can remain within the cell or are secreted into the apoplast via the secretory pathway. Foreign proteins are generally not directed for secretion but can be added with a small piece of signal sequence that initiates their secretion. In plant cell cultures an excretion of the recombinant protein to the culture medium can be achieved by this strategy47. The biopharmaceutical can than be harvested directly from the culture medium and purification to remove the host cell proteins and nucleic acids is not required. Thus protein purification from whole plant biomass is potentially much more difficult and expensive than protein recovery from culture medium and protein secretion is considered an advantage in plant cell culture systems. However not all biopharmaceuticals are suitable for protein excretion since i.e. their stability after secretion in the culture medium can not always be ensured. Also the size of the protein might be crucial: Secre44 45 46 47

Fahrner et al. 2001. Further reading: Drossard 2004. FDA 2002; EMEA 2002. Sharp and Doran 2001.

2.3 Animals as a production platform for recombinant biopharmaceuticals

27

tion of recombinant proteins into the medium requires that they pass pores in the cell wall of the plant cell. The pores in plant cell walls are thought to allow passage of proteins of maximum size around 20 kDa. However a small number of wider pores may serve as channels for relatively slow secretion of larger molecules. Also protein composition and structure may affect the extracellular availability of recombinant protein: in many expression systems it has been described that despite an appropriate signal sequence an considerable amount of foreign protein remain within the plant cell or remain associated with the plant cell wall48. Depending on the recombinant protein plant cell cultures and hairy root cultures can also differ in their ability to secrete the recombinant protein, which could mean that the one or the other system is the preferable choice.

2.3 Animals as a production platform for recombinant biopharmaceuticals Animal pharming involves the expression of protein products in whole animals, principally livestock. Production in mammalian cell culture is not included under this definition. This section 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 purification of protein products from transgenic animals.

2.3.1 Transgene constructs used for animal pharming In animal pharming, the basic intentions underlying the design of all transgene constructs are that the construct should integrate into the host genome, be inherited in classic Mendelian fashion and direct the abundant expression and secretion of a desired protein without affecting the health and well-being of the producing animal. The first step in expressing any protein in a transgenic animal is to obtain a region of cloned DNA that encodes the amino acid sequence. This may be a cloned fragment of genomic DNA, complementary DNA (cDNA) or chemically synthesized DNA. In some cases, it may be necessary to alter this coding sequence to make it suitable for expression in a transgenic animal. For example, if the protein is from an evolutionarily distant species, it may be necessary to alter codons to accord with those most frequently used in the host. Also, if the protein is not normally secreted, a signal peptide may be added at the N-terminal to direct secretion from the producing cell, for example into milk. Furthermore, one needs a gene that is expressed specifically and preferably abundantly in the tissue to be used as the site of production. For example, for the lactating mammary gland milk genes such as beta-casein or 48

Matsumoto et al. 1995.

28

2 The technology of pharming

beta-lactoglobin would be appropriate, or for production in chicken eggs the ovalbumin gene. These genes provide control elements that can be used to confer tissue-specific expression on foreign gene sequences. Typically, these are the gene promoter, where gene transcription initiates, one or more enhancers that positively regulate transcription activity, and inhibitory elements with negative influence. Animal pharming has been heavily reliant on genes whose regulation has been well studied; genetic control elements can sometimes be distant, dispersed and ill-defined. Many researchers have found that transgenes based on genomic sequences are expressed more consistently and more abundantly than those based on cDNA. However, many genes are too large to be conveniently inserted into bacterial plasmid cloning vectors. As a compromise, cDNA and genomic sequences may be combined in a “minigene” carrying one or two introns, and 5’ or 3’ flanking regions known, or suspected, to contain regulatory elements. It has, however, been found that mixing and matching in this way can sometimes lead to aberrant RNA splicing due to the presence of cryptic splice sites. Such effects are often difficult to predict in advance. Despite considerable knowledge of the control of gene expression, there is at present no standardized method of combining disparate genetic elements that is guaranteed to achieve successful transgene expression. Transgene design is a matter of informed trial and error, usually requiring test and refinement of successive generations of constructs. An important factor determining the success of transgene expression lies not in the construct itself, but in the location at which it integrates into the host genome. Transgenes integrated at different sites can exhibit wide variations in expression, due to the influence of different chromatin environments. Heterochromatic regions tend to suppress expression of an adjacent transgene, while a transcriptionally active chromatin domain may support expression. The proximity of endogenous enhancers, promoters, silencers and activation sequences may also influence the level and pattern of transgene expression. This “position effect” has been a long-standing source of inefficiency in the use of transgenic animals. A review49 provides further details of the effect. Several strategies are available to either minimize or circumvent the position effect. Improvements in the level, specificity and consistency of transgene expression can be gained by flanking the transgene construct with insulator elements. Insulators of various types have been demonstrated to block the spread of heterochromatin into the transgene, and isolate transgenic promoters from the effects of adjacent endogenous enhancers and other regulatory elements. Alternatively, the uncertainties of random integration can be avoided by placing the transgene construct at a well-characterized permissive site in the host genome using gene targeting, as described 49

De Laat and Grosveld 2003.

2.3 Animals as a production platform for recombinant biopharmaceuticals

29

in section 2.3.2.4. The Rosa26 locus in the mouse is one example of a commonly used transgene locus that is not subject to gene silencing. Placing recognition sites for either a recombinase, for example bacteriophage P1 Cre recombinase, or an integrase, for example phage φC31 integrase, at the locus allows transgene cassettes to be inserted at very high efficiency by socalled recombinase-mediated cassette exchange50. A human homologue to the mouse Rosa26 locus has recently been identified51 and it is probable that similar sites will also be identified in livestock. A quite different approach is to avoid transgene integration altogether. The problem, then, is to ensure that the foreign DNA replicates autonomously and is stably maintained after multiple cell divisions. This can be achieved using so-called artificial chromosome vectors that carry DNA regions responsible for stable chromosome behaviour. These are: a centromere that mediates chromosome segregation during mitosis, and two telomeres that stabilize the ends of the DNA molecule. Artificial chromosome vectors have the advantage that they can carry very large regions of DNA; however, they are still in early stage development. There has been one report of “transchromosomic” cattle, designed to express human immunoglobulins in blood52. A second form of non-integrating, or episomal, vector has been developed that incorporates an attachment site for the system of structural proteins within the nucleus, termed the nuclear matrix or scaffold53. Such vectors are thought to achieve mitotic stability by “piggybacking” on chromosomes. Episomal vectors are not widely used for generating transgenic livestock, but there has been one report of the production of pigs expressing a fluorescent reporter gene54.

2.3.2 Methods of producing transgenic livestock Mammalian oocytes and early stage embryos are tiny, self-contained, freefloating structures that can be obtained relatively easily by flushing the female reproductive tract with fluid. They can be kept in culture and then reintroduced into recipient females to continue gestation. This accessibility has facilitated the development of a variety of micromanipulation, culture and embryo transfer procedures which support the production of transgenic mammals. Techniques for avian transgenesis are less well-developed. Early stage avian embryos are attached to a large fragile yolk and, once explanted, cannot easily be transferred back into the oviduct, necessitating ex vivo embryo culture55. 50 51 52 53 54 55

Hitz et al. 2007. Irion et al. 2007. Kuroiwa et al. 2002. Jenke et al. 2004. Manzini et al. 2006. Love et al. 1994.

30

2 The technology of pharming

The first viable transgenic mammal was produced in 1974 at the University of California, San Diego, when Rudolf Jaenisch and Beatrice Mintz injected simian virus 40 particles into the cavity of a mouse blastocyst-stage embryo. Two years later, Jaenisch reported that mouse embryos could be infected with a retrovirus which stably integrated the proviral DNA into their genome. These animals incorporated the retroviral DNA into their somatic tissues and germ line and passed it on to their progeny in Mendelian fashion. Since then, transgenic animals have proved to be valuable research tools and are gradually finding practical applications, for example as models of human disease, in xenotransplantation and animal pharming. The term ‘transgenic’ is now generally used not only to describe animals bearing additional exogenous DNA, such as an expression construct, but also those carrying other forms of engineered genetic change, such as gene replacement, deletion, inactivation or alteration. In this section, we provide an overview of various methods of generating transgenic mammals and birds, including those in routine use and some still being developed. The methods divide into two broad categories: nucleic acid transfer directly into embryos (sections 2.3.2.1 to 2.3.2.3) and cell-mediated transgenesis (sections 2.3.2.4 to 2.3.2.8). Nucleic acid transfer into embryos is more straightforward than cell-mediated transgenesis. However, it is at present limited to transgene addition and allows no control over where the transgene integrates into the genome. Definitive analysis of the integrated transgene must be carried out in resultant animals, which in livestock is costly and restricts the number of independent integrations that can feasibly be generated and investigated. The various methods of cell-mediated transgenesis each have in common the feature that genetic manipulation and analysis of the transgenic genotype is carried out in cells in the laboratory, rather than in animals “on the farm”. These cells are then used to transfer the modified genotype to whole animals. While cell-mediated transgenesis is more labour intensive than direct transgenesis, in vitro genetic manipulation of cells followed by detailed genome analysis offers significant advantages. Firstly, it reduces the total number of animals required to generate a useful transgenic. Secondly, it increases dramatically the number of independent transgene integration events that can be screened and investigated. Thirdly, it facilitates the engineering of precisely controlled genetic alterations (gene targeting) by allowing selection and isolation of rare integration events resulting from homologous recombination (see section 2.3.2.4). 2.3.2.1 Pronuclear DNA microinjection

In 1980, Frank Ruddle of Yale University reported that naked DNA, mechanically microinjected into the pronuclei of fertilized mouse eggs, could stably integrate into the host genome56. Microinjection by the same basic pro56

Gordon et al. 1980.

2.3 Animals as a production platform for recombinant biopharmaceuticals

31

cedure was successfully extended to livestock shortly after57. DNA microinjection is rarely used in birds, but in mammals it offers a straightforward method that has been used to produce transgenic pigs, cattle, rabbits, goats and sheep. A wide variety of transgenes have been microinjected, including DNA fragments as large as 0.5 mb and combinations of up to four different transgenes in mice, and three in sheep. The procedure may be divided into a number of stages as summarized below. Several of these stages are common to other methods. Collection of fertilized eggs. Oocytes fertilized in vivo can be collected from animals induced to ovulate by hormonal treatment or from spontaneous ovulation. Alternatively, oocytes can be withdrawn from follicles of ovaries collected from slaughterhouses; in the United States companies have been established to provide such oocytes by post. Embryos are then produced by in vitro maturation and fertilization (IVM/IVF), often termed in vitro production (IVP). IVP procedures are well established for cattle and have recently been developed for pigs. This technique avoids the need for donor animals and has become the method of choice in cattle and is increasingly being used in pigs. Using current techniques, IVP fertilized oocytes are slightly less viable than those derived in vivo, but this is offset by the greater numbers available. Preparation of DNA. A recombinant DNA construct used to generate a transgenic animal will generally have two broad components, the transgene portion and the bacterial plasmid “backbone” which comprises sequences required for selection and DNA replication in bacteria, usually Escherichia coli. Preparation of a DNA construct for microinjection entails purifying linear fragments of the transgene portion away from the plasmid DNA backbone, residual bacterial material and chemical reagents. These procedures are well developed standard molecular biology practice. Exogenous DNA is toxic to early embryos, therefore the quality and concentration of microinjected DNA is critical for survival. Purified linear fragments are dissolved in microinjection buffer at a concentration between 1–6 ng/µl. Within this range, lower concentrations result in higher embryo viability but a lower proportion of transgenic embryos, higher concentrations produce more transgenic embryos but viability is reduced. Injection of DNA. Morphologically normal fertilized oocytes with visible pronuclei are selected for microinjection. In some species, for example pig and cattle, oocytes are relatively opaque due to abundant lipid vesicles, but transparency can be improved by localising the lipids using gentle centrifugation. Microinjection is carried out on the stage of an inverted microscope, with the fertilized oocytes contained within microdrops into which are inserted a blunt-ended, glass holding pipette and a sharp, glass microin57

Hammer et al. 1985.

32

2 The technology of pharming

jection needle containing the DNA solution. Movements of the pipettes are controlled by micromanipulation arms (see figure 2.4). Individual oocytes are held in place on the holding pipette by gentle suction and oriented so that one of the pronuclei is adjacent to the needle. The needle is inserted into one of the pronuclei and DNA solution injected (figure 2.4b). Injected oocytes can then be either immediately transferred to recipients, or cultured in vitro, sometimes as far as blastocyst stage, to reveal those that sustain further development. Transfer and gestation in recipients. Viable embryos are selected for embryo transfer and introduced into the oviducts of hormonally primed (or in mice, pseudopregnant) recipients to complete gestation. Transgenic animals that develop from manipulated embryos are conventionally referred to as “founders”, because they may be used to found genetically modified lines by conventional breeding. a)

b)

Instrument holder Holding pipette

Embryo in droplet of culture medium Oil

Injection needle

Figure 2.4: a) Typical micromanipulator stage (photographic image provided

courtesy of Eppendorf) b) Microinjection of a mammalian oocyte

2.3 Animals as a production platform for recombinant biopharmaceuticals

33

Identification of founders and subsequent breeding. Microinjection is an inefficient process. Between 5–20 % of mice and 1–5 % of large animals born after microinjection carry a transgene, and a smaller proportion of these are likely to express the transgenic product. In mice, this inefficiency can readily be compensated for by injecting and transferring large numbers of embryos. However, this is frequently impractical in livestock because gestation times are longer and maintenance costs far higher. There has therefore been a strong incentive to reduce the number of animals gestating non-transgenic foetuses. One approach has been to screen embryos for the presence of a transgene before transfer. This can be achieved by extracting a portion of the embryo, often a single blastomere, and detecting the transgene by polymerase chain reaction (PCR) amplification. This procedure is however labour-intensive, can reduce embryo viability, and the presence of non-integrated DNA may result in false positives. Alternatively, a gene encoding a non-toxic fluorescent protein can be either co-injected, or incorporated into the transgene construct to identify intact living transgenic embryos. However, expression of a non-integrated reporter construct may again produce a false positive signal, and the presence of additional DNA may be undesirable. Transgenic foetuses can be identified in utero by analysis of cells shed by the developing foetus obtained by amniocentesis or allantocentesis; however, these procedures carry a significant risk of inducing abortion. Some efforts have also been made to detect and analyse foetal cells or DNA in the maternal circulation, with limited success. The most common practice is therefore to screen animals shortly after birth by either PCR or Southern hybridization, using small samples taken from blood, tail or ear tips. Transgenic animals are then analysed in more detail to identify those most suitable for further breeding. Where the intention is to express a protein, researchers will wish to determine the amount and properties of the protein expressed, and whether the physiology of the animal is affected in any way. Usually, some preliminary data on these parameters will already have been gained by mouse experiments before transgenic livestock are generated. However, prediction across quite distantly related species is necessarily imperfect. Furthermore, the position effect means that each founder is potentially different and must be analyzed independently. Collection of milk expression data requires that female animals must be derived, attain sexual maturity, breed and lactate. This is necessarily time-consuming in livestock, especially where a founder is male. Protocols for artificial hormone-induced lactation in virgin females and even males have therefore been developed. This can accelerate the process of identifying the most suitable founders, but the quantity of milk obtained is often low, and expression data from induced lactation may differ from natural lactation.

34

2 The technology of pharming

Subsequent breeding from founder animals is carried out by conventional animal husbandry. However, it is not uncommon for founders produced by microinjection to be mosaic for the presence of the transgene. This is believed to be a consequence of delayed DNA integration. Mosaicism in the germ line reduces the frequency of transgene inheritance in the first generation. Mosaic components of a founder animal may possibly contain independent and different integrations of the transgene that segregate in the first generation. Transgene segregation due to founder mosaicism does not indicate transgene instability. In summary, DNA microinjection is a straightforward method of generating transgenics, and has been the most commonly used over the past two decades. However, the limitations of random transgene addition and the inefficient use of animals are significant drawbacks, and radical improvements to the technique are not expected in the immediate future. We anticipate that it will soon be superseded by lentiviral transduction (section 2.3.2.2) and nuclear transfer (section 2.3.2.6). 2.3.2.2 Viral gene transfer

A number of viruses (for example SV40, AAV) have been used to generate transgenic animals with varying degrees of success. By far the most commonly used are retroviruses, see also review58. Retroviruses are a diverse group of viruses that share a basic common structure and replicative strategy. Each viral particle is approximately 100nm in diameter and consists of a protein “capsid” containing a genome of two single-stranded 8-11kb RNA molecules, surrounded by a lipid envelope through which glycoprotein spikes protrude. Retroviruses infect their hosts by binding the spike proteins to specific receptors on the cell surface. The virus enters the cell, converts the RNA genome into DNA and this integrates randomly into a host chromosome as a provirus. The transcriptional apparatus of the host cell is subverted to produce RNA from the provirus. Viral RNA transcripts include smaller species that encode viral protein components, and fulllength genomic RNAs that are packaged into virus particles and bud off from the cell surface. Retroviruses infect susceptible cells with very high efficiency and integrate as single copies into the host genome. This has led to the development of retroviral vectors capable of introducing foreign genes into cells, termed viral transduction. These vectors generally retain the outermost parts of the viral genome, termed long terminal repeats (LTRs) necessary for the production of genomic RNA, but lack viral genes and carry other disabling mutations. Retroviral vectors are therefore “replication incompetent”, being unable to produce viral particles on their own.

58

Buchschacher 2001.

2.3 Animals as a production platform for recombinant biopharmaceuticals

Simplified structure of a retrovirus genome Structural genes gag

pol

env

35

Vector integrates as provirus into host genome Transgene

Retroviral vector Transgene

Vector DNA transfected into packaging cell

Virus introduced into perivitelline space of oocyte or early embryo

Vector RNA transcribed and packaged into virus

Viral proteins ex- gag pressed from sepapol rate constructs

Infectious viral particles shed into culture medium

env

Figure 2.5: Transduction of a transgene using a replication-defective retroviral

vector

Retroviral transgenesis is summarized in figure 2.5. The transgene is first inserted into retroviral vector DNA. This construct is transfected into an artificial “packaging cell line” designed to provide the viral components deleted from the vector. Typically, a packaging cell contains two or three stably transfected separate plasmids expressing viral genes: env which determines host cell specificity, gag encoding structural proteins and pol encoding the reverse transcriptase enzyme necessary to convert viral RNA into DNA. The transfected vector generates genomic RNA transcripts within the packaging cell. These associate with the other viral proteins, become incorporated into infectious virus particles and shed into the culture medium. An embryo may be infected either by removing its zona pellucida and culturing in culture medium containing virus, or a small volume of viral supernatant may be microinjected into the perivitelline space. The virus vector can infect and incorporate into the host genome, but is incapable of further replication. Procedures for embryo explantation and transfer are basically those described for DNA microinjection. Until recently, the majority of retroviral vectors were based on the oncoretrovirus Moloney murine leukemia virus, as used by Jaenisch. Transgenic mice, pigs and cattle have all been produced using these vectors, but rates

36

2 The technology of pharming

of transgenesis have been low and transgenes are almost always silenced by epigenetic modification during development. However, in 1996 scientists at the Salk Institute, San Diego, developed vectors based on lentiviruses. Lentiviruses are a subclass of retroviruses that include the sheep maedi-visna virus, equine infectious anaemia virus and the bovine, feline, simian and human immunodeficiency viruses. These offer two significant advantages over oncoretroviruses. Firstly, they are not subject to epigenetic inactivation in transgenic animals. This is probably because, unlike oncoretroviruses, lentiviruses are usually transmitted horizontally between individuals rather than vertically through the germ line. Host animals have, therefore, not adapted mechanisms to silence their expression. Secondly, unlike oncoretroviruses they do not require breakdown of the nuclear envelope to gain access to the host genome. Lentiviral integration can therefore occur at all stages from the unfertilized oocyte onwards, in both replicating and non-replicating cells. Lentiviruses have been used successfully in a number of mammalian species: mice, rat, pig and cattle. For further background see a review by Pfeifer59. Two publications, describing the production of pigs carrying a green fluorescent protein reporter transgene, illustrate the high rate of transgenesis obtained using lentivirus vectors. Alexander Pfeiffer and Eckhard Wolf of the Ludwig-Maximilians University, Munich used a vector based on human immunodeficiency virus and report 70 % of live born pigs carried a fluorescent protein transgene, of which 94 % expressed the transgene60. Bruce Whitelaw and colleagues of the Roslin Institute, Scotland obtained similarly high results in pigs using a vector based on equine infectious anaemia virus. 31 % of zygotes injected with virus developed to transgenic animals, of which 95 % expressed the transgene61. However, on the downside, delayed retroviral integration and independent integrations into different embryonic blastomeres are both quite common, and lead to a relatively high incidence of mosaicism in founders. This can extend the time required to establish stable transgenic lines; firstly, because mosaicism in the germ line reduces the frequency of transgene inheritance and secondly, because several generations may be required to segregate independent transgene loci. Retrovirus-mediated gene transfer has been the principle means of producing transgenic chickens, and has also been used to produce transgenic quail. Infection is most commonly carried out on embryos within freshly laid eggs. At this stage the chick embryo comprises a multilayered plate, or blastoderm, of about 60,000 cells overlying the large yolk. The shell of the egg is removed and retroviral particles are injected into a fluid-filled space, 59 60 61

Pfeifer 2004. Hofmann et al. 2003. Whitelaw et al. 2004.

2.3 Animals as a production platform for recombinant biopharmaceuticals

37

Injection needle Membrane Blastoderm disc Subgerminal cavity

Window in shell

Yolk

White

Figure 2.6: Injection of retrovirus particles into a hen’s egg (shown in cross

section)

the subgerminal cavity, that lies beneath the embryo, see figure 2.6. The shell is then resealed and the embryo allowed to incubate, sometimes being transferred to a surrogate shell after two or three days incubation. Because infection takes place at a relatively late stage of development, founder birds are mosaic for retroviral integrations and must be bred on to segregate different proviruses. Early work used replication-defective vectors based on oncoretroviruses such as avian leucosis virus, or reticuloendotheliosis virus. The production of pharmaceutical proteins in bird eggs was first demonstrated using an avian leukosis virus vector to transduce a non-specific promoter directing the expression of interferon α-2b62. This resulted in extremely low frequencies of germ line integration and transgenes were frequently silenced over generations. More recent work using lentiviral vectors based on equine infectious anaemia virus has been more successful, and pharmaceutically relevant quantities of proteins useful for cancer treatment have now been expressed specifically in the egg white, with no evidence of ectopic expression or transgene silencing63. Two features of retroviral transduction may, however, make it difficult to achieve the high levels of expression desirable for animal pharming applications. The size of the transduced gene is limited to a maximum of approximately 8kb by the capacity of the retroviral particle. This effectively restricts their use to cDNA and small minigene constructs, and severely limits the amount of regulatory and enhancer elements that can be included. Retroviral vectors usually result in single or low copy transgenes, but high trans62 63

Rapp et al. 2003. Lillico et al. 2007.

38

2 The technology of pharming

gene copy numbers are often desirable because they tend to correlate with more abundant expression. Random integration of proviral DNA also carries the risk that genes adjacent to the integration site may be inappropriately activated, or that integration disrupts an endogenous gene with possible deleterious effects. Retroviruses have, in fact, been successfully used as experimental insertional mutagens. There is also the possibility that, despite extensive precautions taken to disable the retroviral vector, an integrated proviral transgene could somehow recombine with endogenous retroviral elements resulting in a replication-competent virus. While such events are likely to be extremely uncommon, they cannot be absolutely excluded. In summary, lentiviral transduction offers an efficient means of transgenesis that is likely to become more widely used. Its usefulness in animal pharming will however depend on the levels of expression that can be achieved, a deeper understanding of any possible risks and public acceptance of viral-based vectors. 2.3.2.3 Sperm-mediated gene transfer

This technique could formally be categorized as cell-mediated transgenesis, because a sperm cell is used to introduce exogenous DNA into an oocyte. However it is best regarded as a means of direct nucleic acid transfer into an embryo, because no culture, selection or analysis of cells is carried out before animals are produced. Sperm-mediated gene transfer has been recently reviewed64. Since 1971, it has been known that rabbit spermatozoa can associate in vitro with exogenous DNA and transfer it to an oocyte by fertilization. Subsequently the same has been shown for sperm cells of other species. In 1989 Marialuisa Lavitrano of the University La Sapienza, Rome, claimed that mouse spermatozoa exposed to exogenous DNA could be used as a vector to generate transgenic mice by artificial insemination. This report stimulated considerable interest because it offered a simple approach to the production of transgenic animals. However, the method has suffered considerable problems of reproducibility and, while there have been reports of transgenic calves and pigs, transgenes introduced in this way frequently undergo rearrangement. This is consistent with findings that DNA which penetrates sperm nuclei becomes fragmented. There have also been reports that DNA introduced directly into the epididymis is transferred to offspring by natural ejaculation and fertilization (reviewed by Sato65). The reproducibility and usefulness of this method has yet to be confirmed. A variation of this technique, based on intracytoplasmic injection of DNA complexed with frozen-thawed or detergent-treated sperm, was devel64 65

Lavitrano et al. 2006. Sato 2006.

2.3 Animals as a production platform for recombinant biopharmaceuticals

39

oped in 1999 by Ryuzo Yanagimachi of the University of Hawaii. This has revived interest in the field and several groups have used it to generate transgenic mice, rats and pigs. An outline of the method is given in a recent review66. In mice it offers a lower incidence of founder mosaicism and a greater efficiency of transgenesis than standard DNA microinjection, particularly for large transgenes. In pigs, however, the efficiency of transgenesis and number of live born piglets obtained has so far been very low. One problem is failure of the injected sperm head to properly decondense and form a male pronucleus. In summary, sperm-mediated gene transfer by natural fertilization has not yet proved a reliable method of producing transgenic animals. Variations based on intracytoplasmic sperm injection may offer greater success, but their potential in large animal transgenesis has yet to be fully explored. 2.3.2.4 Embryonic stem cells

Embryonic stem (ES) cells are pluripotent cells derived from early mammalian embryos, first isolated from mice in 1981. ES cells have three defining functional characteristics. Firstly, they can proliferate undifferentiated for extended periods in vitro. Secondly, they can differentiate in vitro, or as tumours in vivo, into cells of the three embryonic germ lineages: ectoderm, mesoderm and endoderm. Thirdly, when reintroduced into a host embryo they are able to participate in development and contribute to all tissues of the animal, including gametes. ES cell research is a large and fast-moving field and the reader is referred to two reviews67 for further background. ES cells made possible the engineering of genetic modifications in culture and then study of the effects in whole animals. In the past two decades ES cells have been a powerful tool for the experimental manipulation of the mammalian genome. This work has, however, been largely restricted to mice, and only in mice has an ES genotype been transmitted through the germ line. Efforts have been made to derive ES lines from other mammals and there are reports of ES-like cells in hamster, mink, sheep, cattle, pig, monkey, rabbit, rat and human, but none so far have been germ line competent. Human ES cells remain untestable in this respect for obvious ethical reasons. ES-like cells have also been derived from chickens, but without germ line contribution. Because of their potential advantages, work is ongoing to derive definitive ES cells from livestock species. Recent reports that mouse and human somatic cells can be converted to germ line competent ES cells by the expression of defined genes 68 are particularly interesting, and will aid the extension of ES cell technology to other species. 66 67 68

Moreira et al. 2007. Wobus and Boheler 2005; Prelle et al. 2002. Okita et al. 2007; Meissner et al. 2007; Takahashi et al. 2007; Yu et al. 2007.

40

2 The technology of pharming Cloned gene Gene targeting vector carrying engineered alteration and selectable markers Transfection Homologous recombination Drug selection

Cell clone expansion DNA analysis

Target gene in chromosome

Targeted alteration introduced into chromosome

Figure 2.7: Gene targeting by homologous recombination in cultured cells

DNA can be introduced into cultured mammalian cells by a wide range of chemical, electrical, mechanical and viral methods, often collectively referred to as “transfection”. In ES cells, transfection can be used to randomly add DNA sequences to the genome, but the most potent application has been the precise modification of genes in situ, termed gene targeting, see figure 2.7. Gene targeting exploits the ability of cells to support recombination between exogenous DNA molecules and their cognate chromosomal sequences at regions of shared homology. Typically, ES cells are transfected with a DNA construct carrying an engineered modification flanked by 2–15 kb “arms” of DNA homologous to the target locus. At a certain frequency, transfected cells undergo homologous recombination with the construct and seamlessly incorporate the engineered modification at the target locus. Targeted cell clones can be selected and identified amongst a background of random integrants by a variety of methods. Isolated targeted ES cell clones can be characterized and then used to generate animals. Gene targeting in mice has been used to inactivate individual endogenous genes by insertion or deletion, replace whole genes, precisely place transgenes in the host genome, introduce subtle gene modifications and to delete megabase-size DNA fragments rendering large regions hemizygous. There are many variations and refinements of the technique, and these continue to produce a wealth of information about many aspects of mammalian biology, see review69. 69

Capecchi 2005.

2.3 Animals as a production platform for recombinant biopharmaceuticals Indefinite growth in culture

41

DNA transfection, selection, clonal isolation, analysis

ES cell isolation from blastocyst

Genetically modified line

Chimeric mouse

Introduction of manipulated ES cells into host embryo

Figure 2.8: Production of genetically modified mice using embryonic stem

cells

There are now several possible methods of generating mice from ES cells. The most common exploits the ability of an early embryo to incorporate exogenous ES cells. A small number of ES cells may be aggregated with a pre-morula stage embryo by co-culture in a microwell, or microinjected into the cavity of a blastocyst. Embryos containing ES cells are then transferred to the reproductive tract of recipients to complete development. Animals produced in this way are chimeric, composed of a patchwork of ES and host-derived cells. In mice, chimerism can be readily visualized by marking the ES cells and host embryos with different coat colour genotypes. Chimeras are bred to derive offspring from ES-derived germ cells. See figure 2.8. Mice derived entirely from ES cells can also be produced by aggregating ES cells with an unviable tetraploid host embryo, generated by prior electrofusion of a two-cell stage embryo. This method is more successful with either early passage, or F1 hybrid ES lines. ES cells have also been used to produce whole animals by nuclear transfer, as described below. Chicken ES-like cells derived from blastoderm stage embryos have been cultured and transfected in a similar way to mouse ES cells. When injected into the subgerminal cavity of freshly laid eggs, some cells become incorporated into the embryo and contribute to the body. Transfected chicken ES-like cells have been used to generate chimeric hens with a substantial ES contribution, including the egg white producing cells of the oviduct70, but germ line transmission has not yet been demonstrated. 70

Zhu et al. 2005.

42

2 The technology of pharming

Possible new routes to the production of whole animals using ES cells are opening with reports that gametes can be produced from ES cells in vitro. Oocytes and follicular structures have been reported from differentiating mouse ES cells, but it remains unknown whether they can be fertilized71. Mouse ES cells have also been shown to form sperm capable of producing viable offspring by ICSI72. However, the efficiency of the procedure was low and the offspring had a reduced lifespan, possibly indicating an imprinting problem. In summary, the importance and usefulness of ES cells are widely acknowledged and considerable research efforts are underway to further understand and manipulate them. This rapidly increasing body of knowledge will likely enable practical ES technology to extend to livestock in the future. 2.3.2.5 Embryonic germ cells

Embryonic germ (EG) cells are undifferentiated cells that resemble ES cells, in that they can be cultured and transfected in vitro and then contribute to somatic and germ cells of a chimera. EG cells are derived from isolated primordial germ cells. These are the progenitors of the gametes, and arise outside the embryo and migrate through the body during mid-stage development to their eventual site in the developing gonads, the urogenital ridges. Primordial germ cells cultured with a combination of growth factors: leukemia inhibitory factor, steel factor, and basic fibroblast growth factor, grow and convert to an undifferentiated cell type resembling ES cells. EG lines were first derived from mice in 1992 and have since been isolated from rat, pig, cattle and chicken. Reintroduction of livestock EG cells into embryos in the same way as ES cells has not been very successful. Porcine chimeras have been produced from both genetically manipulated and normal EG cells. However, the efficiency of chimera formation was poor, EG contribution to each chimera was consistently low and restricted to somatic cells, probably due to imprinting problems73. This has restricted their practical usefulness and mammalian EG cells are now rarely used. Avian EG cells have been more successful, and produced transgenic chickens on several occasions, see review74. The typical technique has been to isolate primordial germ cells from the blood of chick foetuses two to three days after laying and convert them to EG cell. Cultured EG cells can be reintroduced into the blood stream of foetuses at the same stage, where they migrate to the gonads and form functional gametes. The efficiency of chimera production and incidence of germ line transmission is increased if 71 72 73 74

Hubner et al. 2003. Nayernia et al. 2006a. Labosky et al. 1994. Petitte et al. 2004.

2.3 Animals as a production platform for recombinant biopharmaceuticals

43

Recipient depleted of endogenous primordial germ cells

Primordial germ cells from fetal blood

EG cell culture

Genetic manipulation

EG cells injected into fetal circulation – migrate to gonads

Chimeric chicken

Figure 2.9: Production of genetically modified chickens using embryonic germ

cells

recipient eggs are chemically pre-treated to deplete endogenous primordial germ cells (see figure 2.9). 2.3.2.6 Nuclear transfer

The reader is referred to recent reviews75 for more details of this subject. The replacement of an egg nucleus with that of another cell was first suggested in 1938 by Hans Spemann of the University of Freiburg as a means of determining whether nuclei of differentiated and undifferentiated cells have equivalent development potential. In the 1950s, Robert Briggs and Thomas King in Philadelphia showed that cell nuclei from frog blastocysts could be transplanted to enucleated eggs and direct normal development to feeding-stage larvae, whereas nuclei from mesoderm or endoderm of lategastrulation stage embryos were unable to do so. This led to the view that development is accompanied by determination of cell fate, a position that was not disproved until three decades later. During the 1980s, culture conditions and micromanipulation techniques for mammalian embryos improved significantly and nuclear transfer in mammals became a practical proposition. Somewhat unusually, the major developments and breakthroughs were made using livestock rather than mice. Nuclear transfer in livestock is generally carried out by removing the genomic DNA from an unfertilized oocyte, usually by microsurgical withdrawal of a portion of ooplasm containing the second metaphase plate. Enucleation of zygotes has also been used but with less success. The donor nucleus is then introduced into the cytoplast by microinjection or electrofusion. Reconstructed embryos are activated to simulate fertilization, then cultured if possible to identify viable embryos and transferred to the oviduct of foster mothers to complete gestation, see figure 2.10. 75

Campbell et al. 2005; Wells 2005; Vajta and Gjerris 2006.

44

2 The technology of pharming

In 1986, the first viable mammal was produced by nuclear transfer, a sheep produced by fusion of a blastomere of an eight-cell embryo into an enucleated egg76. For a decade after, nuclear transfer in mammals was limited to cells obtained directly from early embryos or cultured for very short periods. Nuclear transfer in mice proved to be particularly difficult. One feature that emerged during this time was the importance of matching the cell-cycle stage of the donor nucleus and the oocyte cytoplasm. Oocytes of most mammalian species pause twice during meiosis, once before the first meiotic metaphase and again at the second, at which stage the oocyte is mature and can be fertilized. Oocyte maturation and arrest are induced by a high level of maturation (or mitosis) promoting factor (MPF). Fertilization causes a chain of events that result in proteolytic cleavage of MPF, breaking the arrested state and allowing the fertilized oocyte to complete meiotic division. The level of MPF in the oocyte has a profound effect on the outcome of nuclear transfer. If a nucleus is transferred into oocyte cytoplasm with high MPF, the nuclear envelope breaks down and chromatin undergoes chromosome condensation, followed by nuclear reformation and DNA replication. A nucleus from a cell in G1 phase of the cell cycle will undergo normal DNA replication and can support normal development. However, a donor nucleus in S, or G2 phase undergoes aberrant re-replication of DNA, causing aneuploidy or chromosomal damage and consequent failure of development. Nuclear transfer efficiency into unfertilized oocytes can therefore be improved by synchronising donor nuclei in G1. In contrast, if a nucleus is transferred to a fertilized oocyte in which MPF levels have declined, nuclear envelope breakdown does not occur and the cell cycle of an incoming nucleus in either G1, S, or G2 phase will be completed normally. MPF can also be induced to decline by experimentally activating the oocyte by exposure to ionomycin, ethanol, strontium, or by an electrical pulse and these methods have been used to produce live offspring by transfer of S or G2 nuclei. Cell-mediated transgenesis using nuclear transfer became possible when in 1996, Keith Campbell and Ian Wilmut of the Roslin Institute, Edinburgh reported live sheep produced from embryonic cells cultured for several weeks77. They later proposed that the key to successful nuclear transfer was the induction of a quiescent state in the donor cell by serum starvation. Subsequent work revealed that quiescence per se is not critical and the effect of serum starvation is to synchronize donor cells in G1 phase. Nuclear transfer was then extended to adult mammary cells. The sheep “Dolly” provided unequivocal evidence that differentiated cells have the capacity to form whole animals, overturning the concept of irreversible determination78. 76 77 78

Willadsen 1986. Campbell et al. 1996. Wilmut et al. 1997.

2.3 Animals as a production platform for recombinant biopharmaceuticals

45

Nuclear transfer from somatic cells opened a new route to produce transgenic livestock, lifting the requirement for embryonic stem cells for cell-mediated transgenesis. “Ordinary” cells such as primary foetal fibroblasts can be obtained in large quantities, manipulated in culture and then converted into whole animals by nuclear transfer. This could potentially generate large numbers of genetically modified animals without conventional breeding. Initial work in the area was inspired by the possibility of producing such “instant flocks” of animals for pharming. In 1997, Angelika Schnieke of PPL Therapeutics, Edinburgh reported the production of a sheep “Polly” that carried a human clotting factor IX transgene, randomly introduced into the genome by in vitro transfection of foetal fibroblasts79. Shortly after, this was extended to gene targeting. An α1-antitrypsin transgene was placed into a site chosen as favourable for expression. In both cases transgene expression was directed into milk80. Somatic cell nuclear transfer has now been used to generate transgenic and gene targeted animals in several other species, demonstrating that sophisticated genetic manipulations are possible in livestock. However, so far relatively few gene-targeted animals have been generated. Genes that have so far been targeted in livestock are: α1 type1 collagen, α1-3 galactosyltransferase, immunoglobulin-µ, and PrP and CFTR. The main difficulty is that primary somatic cells are short-lived in culture, allowing little time for cell transfection, selection and clonal expansion. This obstacle can in part be overcome by “rejuvenating” cells by nuclear transfer and rederivation from resulting foetuses. Such a procedure also allows successive rounds of in vitro genetic manipulation to be carried out relatively quickly, but does introduce the risk that genetic aberrations occur in the cultured cells, but remain undetected until animals are born. Foetal and perinatal mortality and morbidity is the most serious issue that faces nuclear transfer technology. The severity of the problem varies between species, cell types and experimental regimes and is unrelated to genetic manipulation of the cultured cells. Rather, it is believed to be a consequence of defective epigenetic reprogramming of the donor nucleus and possibly incompatibility between the cell-derived nuclear genome and the oocyte-derived mitochondria. Cumulative data for cattle collected up to 2005 indicate that more than 1,500 cloned calves were born, of these 60–70 % survived normally to adulthood. The performance of these, including reproduction, was similar to non-cloned animals. Evidence is also accumulating that ill effects are limited to the first generation. Offspring, including those from two nuclear transfer parents, exhibit no increased morbidity or mortality. 79 80

Schnieke et al. 1997. McCreath et al. 2000.

46

2 The technology of pharming

Oocyte donor

Cell donor

Genetic manipulation and analysis

Enucleation

Cell donor Fusion and activation in vitro culture Embryo transfer to surrogate mother

Rederivation of foetal cells

Genetically modified animal

Figure 2.10: Generation of transgenic animals by nuclear transfer

Extension of nuclear transfer to mice81 was an important development, because it allowed ES cells to be used as nuclear donors. ES cells were found to be significantly superior to somatic cells, both in the efficiency with which animals are derived and also the viability and health of cloned animals. One explanation is that the pattern of gene expression in an ES cell nucleus resembles an embryonic blastomere more closely than does that of a highly differentiated cell such as a fibroblast, and consequently requires less extensive reprogramming. However, recent analysis has questioned whether the differentiated state of the donor cell correlates with success of nuclear transfer82. In summary, nuclear transfer from somatic cells has yet to fulfil its initial promise of producing instant flocks or herds, although efficiencies are 81 82

Wakayama et al. 1998. Oback and Wells 2006.

2.3 Animals as a production platform for recombinant biopharmaceuticals

47

continually improving. At present, nuclear transfer offers the only practical method of gene targeting in livestock. Clone viability remains a problem, but technical refinements and the prospect of livestock embryonic stem cells as nuclear donors may yet resolve this. 2.3.2.7 Spermatogonial stem cells

Spermatogonial stem cells (SSCs) are a self-renewing population of germ cells within the adult testes that form the spermatogonia and differentiate to spermatozoa. In the early 1990s, Ralph Brinster of the University of Pennsylvania showed that SSCs from one mouse could be transferred to the testes of another sterile or sub-fertile mouse, where they locate to a stem cell niche in the seminiferous tubules and go on to produce functional sperm. Subsequently Brinster and others showed that SSCs could be transferred across species. Rat SSCs could be transplanted into mouse testes and produce sperm and vice versa; hamster SSCs also produce sperm in mouse testes. More distantly related species, such as rabbit, pig, baboon, and human, also populate mouse testes, but do not complete spermatogenesis. It is possible that this barrier could be overcome by co-transplantation of sertoli cells, or testis tissue. Transfer of germ cells between livestock species is now being developed. Ina Dobrinksi of the University of Pennsylvania reported in 2002 the transfer of germ cells between pigs, although sperm were not produced, and in 2003 the transfer of germ cells marked with a transgene between goats and transmission of the transgene through sperm to offspring, see review83. Germ cell transplantation offers an exciting new method of producing genetically modified animals, providing culture conditions are developed under which SSCs can be expanded, undergo DNA transfection and selection, while remaining genetically and epigenetically intact for transplantation. Because modifications are introduced directly into the male germ line, without embryo manipulation, the time required for gestation and maturing founder animals is avoided, a significant factor in large animals. In early work, Brinster demonstrated that SSCs could be cultured in vitro for several months while retaining the ability to repopulate a recipient testis, but they proliferated poorly due to inhibitory factors in serum. This problem severely limited their usefulness for several years. In 2004, glial cell line-derived neurotrophic factor (GDNF) was identified as promoting SSC self renewal in vivo. Subsequently, serum-free conditions including GDNF that support SSC self-renewal in vitro were developed. In 2006, a group led by Takahashi Shinohara of the University of Kyoto reported targeted inactivation of a gene in cultured mouse SSCs, their transplant into testes and transmission of the targeted allele through sperm to offspring84. Researchers differ in their findings regarding the most suitable SSC culture media. 83 84

Dobrinski 2005. Kanatsu-Shinohara et al. 2006.

48

2 The technology of pharming

Interestingly, SSCs themselves offer a source of primitive pluripotent cells, apparently functionally equivalent to ES cells. In 2004, Shinohara used specific culture conditions to establish cells from neonatal mouse testis that were phenotypically similar to ES cells85. These differentiated into various types of somatic cells in vitro and, most importantly, formed germ line chimeras when injected into blastocysts. Other researchers have made similar findings, reporting that a subset of cultured SSCs converted to an ES-like morphology in the presence of particular growth factors86. Again, these differentiated in vitro and early stage derivatives could form germline chimeras. In summary, this is a promising new field. SSC technology is, however, at an early stage and the techniques required for transgenesis, such as conditions for extended culture, are still being developed. 2.3.2.8 Adult stem cells

The nature and differentiative abilities of adult stem cells, for example haematopoietic, amniotic and mesenchymal stem cells, are the subject of active research with frequently contentious results. Most of the effort in this field is directed towards generating human tissue for regenerative medicine. However, the development of methods that support in vitro culture and transfection of adult stem cells may also facilitate their future use in cellmediated transgenesis. This could be in combination with nuclear transfer as has already been shown for mesenchymal stem cells87, or as a source of gametes, if recent reports that male germ cells can be derived from bone marrow stem cells88 are confirmed. 2.3.2.9 Overview

Figure 2.11 (p. 49) outlines the stages and procedures involved in the establishment of a transgenic production herd. Table 2.5 summarizes the various methods of transgenesis.

2.3.3 Choice of species and site of production In mammals, research and development of animal pharming projects has overwhelmingly been carried out in mice. Species used for production are: rabbit, sheep, goat, pig and cattle. The choice of production species is made on the basis of a combination of factors, such as ease of husbandry, generational interval, fecundity and the potential yield of recombinant protein. Reproduction data and production capacity of different mammalian species in milk are shown in table 2.6 below (p. 50–51). 85 86 87 88

Kanatsu-Shinohara et al. 2004. Guan et al. 2006. Kato et al. 2004; Bosch et al. 2006. Nayernia et al. 2006.

2.3 Animals as a production platform for recombinant biopharmaceuticals Number of experimental animals required for production of a transgenic founder animal Oocyte donors 0

Slaughterhouse

(

(or 10–20)

49

Procedures Oocytes from slaughterhouse (or in vivo derived: super-ovulation, transvaginal oocyte recovery)

)

Tg method, e.g. DNA microinjection, lentiviral vector, nuclear transfer

Foster mothers: 5–50

Experimental phase

In vitro culture to blastocyst stage embryo Hormone treatment, embryo transfer, conventional animal husbandry Pregnancy and parturition (Caesarean section common with nuclear transfer) Tg assessment: e.g. blood sample, ear clip, biopsy, induced lactation, milking Cull of non-tg male offspring Breeding, semen collection (conventional animal husbandry)

Further generations: G1, G2

Tg evaluation and assessment: e.g. blood sample, ear clip, biopsy, induced lactation, milking Cull of non-tg male offspring Breeding, semen collection (conventional animal husbandry)

Production animals

Collection of protein: e.g. milking Ongoing assessment: ear clip, blood sample, etc. Cull of non-tg male offspring (and most tg males, depending on collection method) Breeding (conventional animal husbandry) tg: transgenic

Production phase

Offspring: G0 founders 1–5

Figure 2.11: Establishing a transgenic production herd: animals and proce-

dures required

50

Table 2.5: Methods of producing transgenic animals Transfer route

Additional DNA

Gene targeting

Founder anima I

Germline transmission

Inadvertent health effects

Demonstrated in mammalian species species

Efficiency in livestock (oocyte/live offspring4)

DNA Microinjection

DNA => 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

Very rare, insertional mutagenesis possible

Mouse,

Not done

Embryonic Stem Cells (ES)

DNA=> ES cells => incorporate into tetraploid embryo

Selectable marker

Yes

Completely ESderived

Yes

Very rare, insertional mutagenesis possible

Mouse

Not done

Continued on next page

2 The technology of pharming

Method

Method

Transfer route

Additional DNA

Gene targeting

Founder animal

Embryonic Stem Cells (ES)

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

2 The technology of pharming

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

2.3 Animals as a production platform for recombinant biopharmaceuticals

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

2 The technology of pharming

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.

2.3 Animals as a production platform for recombinant biopharmaceuticals

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

2 The technology of pharming

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.

2.3 Animals as a production platform for recombinant biopharmaceuticals

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

2 The technology of pharming

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.

2.3 Animals as a production platform for recombinant biopharmaceuticals

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

2 The technology of pharming

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),

2.3 Animals as a production platform for recombinant biopharmaceuticals

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

2 The technology of pharming

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

2 The technology of pharming

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 The technology of pharming

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

2 The technology of pharming

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

2 The technology of pharming

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

3 Risk assessment of plant pharming and animal pharming

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

3 Risk assessment of plant pharming and animal pharming

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

3 Risk assessment of plant pharming and animal pharming

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.

3.1 Environmental risks and co-existence of plants genetically modified

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

3 Risk assessment of plant pharming and animal pharming

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

3.1 Environmental risks and co-existence of plants genetically modified

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)

3 Risk assessment of plant pharming and animal pharming

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

Modes of reproduction

82

Crop

3.1 Environmental risks and co-existence of plants genetically modified

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

3 Risk assessment of plant pharming and animal pharming

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.

3.1 Environmental risks and co-existence of plants genetically modified

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

3 Risk assessment of plant pharming and animal pharming

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.

3.1 Environmental risks and co-existence of plants genetically modified

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

3 Risk assessment of plant pharming and animal pharming

– 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.

3.1 Environmental risks and co-existence of plants genetically modified

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

3 Risk assessment of plant pharming and animal pharming

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.

3.1 Environmental risks and co-existence of plants genetically modified

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

3 Risk assessment of plant pharming and animal pharming

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

3 Risk assessment of plant pharming and animal pharming

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

3 Risk assessment of plant pharming and animal pharming

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

3 Risk assessment of plant pharming and animal pharming

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

4 The welfare of pharming animals

– 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

4 The welfare of pharming animals

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

4 The welfare of pharming animals

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

4 The welfare of pharming animals

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

4 The welfare of pharming animals

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.

4.5 Animal welfare considerations in the development phase

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

4 The welfare of pharming animals

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.

4.5 Animal welfare considerations in the development phase

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

4 The welfare of pharming animals

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

4 The welfare of pharming animals

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

4 The welfare of pharming animals

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

4 The welfare of pharming animals

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

5 Public views and attitudes to pharming

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

5 Public views and attitudes to pharming

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

5 Public views and attitudes to pharming

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

5 Public views and attitudes to pharming

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

5 Public views and attitudes to pharming

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 Public views and attitudes to pharming

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).

5.3 Attitudes to pharming in advanced societies

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

5 Public views and attitudes to pharming

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

5.3 Attitudes to pharming in advanced societies

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

5 Public views and attitudes to pharming

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

5 Public views and attitudes to pharming

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-

5.4 A differentiated landscape of perceptions of pharming

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 Public views and attitudes to pharming

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).

5.4 A differentiated landscape of perceptions of pharming

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

5 Public views and attitudes to pharming

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

5 Public views and attitudes to pharming

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<.01, and mean values (on a scale from 0 to 10) for the three groups of 4.87, 5.35 and 5.82 respectively. In the case of animal pharming, the equivalent results were F(2, 17436)=55.01, p<.01, and mean values 3.36, 3.82 and 4.20. Knowledge counts in supporting pharming, even if the size of its contribution is small. Another potentially important variable is the worldview people hold about the effects of scientific and technological developments on our lives. In the analysis to be presented here, we use an indicator that captures a positive view of science. This indicator, which we can call “expectations 37 38

The following analyses refer only to the consolidated sample of the European countries included in our study. The knowledge summated scale was integrated by the following question and items: “Please tell me, for each of the following sentences, the extent to which you believe they are true or false? Please use the following scale: Absolutely true (1), Probably true (2), Probably false (3), Absolutely false (4)”. Read list. Show card. Plants and animals are made up of cells; 2. The gene is the basic unit of inheritance of all living creatures; 3. The mother’s gene determines whether a newborn will be a boy or a girl; 4. DNA is the genetic material contained in the cell nucleus; 5. It is DNA that gives the instructions for the construction, maintenance, reproduction and repair of the organism; 6. DNA has a circular form; 7. Humans evolved from earlier animal species; 8. It is not possible to transfer genes from humans to animals; 9. It is possible to know from a test taken in about the third month of pregnancy whether or not the baby will have Down syndrome (Mongolism); 10. Cloning is a form of reproduction in which offspring result from the union of egg and sperm; 11. Nowadays, stem cells are usually extracted from human embryos without destroying them; 12. In a fifteen-day-old embryo you can already see the organs starting to form; 13. Stem cells can be transformed into different types of cells that can then be grown into specialized tissues like muscles or nerves.

146

5 Public views and attitudes to pharming

about techno-scientific developments”, measures perceptions of the future impacts of techno-scientific advances. “Expectations” measures perceptions of how a series of specific, present-day techno-scientific advances (computers, Internet, genetic engineering, solar energy, among others) will impact on “the quality of life” in a defined future term (the next 25 years). Using a 10-item battery designed to capture expectations about the capacity of different techno-scientific developments to improve our quality of life over the next 25 years, we can construct an aggregate measure of general predispositions toward scientific and technological advances.39 The first step is to assign a score to individuals based on the number of “improve” responses (range 0–10) then divide them into three groups: those with high expectations (7 or more “improve” responses, representing 39.6 % of the total sample), those with medium expectations (4 to 6 “improve” responses, accounting for 39.7 %) and those with low expectations (3 or fewer points, the remaining 20.7 %). The assumption is that individuals with a generally or consistently optimistic view of techno-scientific progress in the future would also have a more favourable view of a particular, even if controversial, new scientific area such as pharming; an assumption which is open to empirical corroboration or falsification. The mean scores of each of these groups on the level of support for plant and animal pharming (range 0–10), evidences a marked differentiation between the high expectations group and the low expectations group. For plant pharming the mean values are 4.46 and 6.31, with the medium expectations group in between with a mean of 5.13 points, F (2, 17497)=595.45, p<.01). In the case of animal pharming, the mean values are 2.86, 3.50 and 4.83 for the low, medium and high expectations groups respectively, F(2, 17436)=708.972, p<.01). Another important worldview is the image or representation people in a society may have about nature and the role of science and technology in its transformation. Cultural historian Leo Marx has observed that the belief in progress, which characterized modern Euro-American culture, has eroded over the last three decades. He argues that the main factor in its decline has been the growing pessimism about the role of human beings in nature; that is, the awareness that our industrial production system and modernity in general, sustained by science and technology, are having serious adverse 39

The question used to capture expectations about science is drawn from one used in the Eurobarometer series on perceptions of biotechnology, and is worded: “I will read a list of new technologies to you. Please tell me, in each case, if you think this technology will improve our quality of life in the next 25 years, if it will have no effect, or if it will worsen the quality of life”. 1. Solar energy 2. Computers 3. Biotechnology 4. Telecommunications 5. New materials 6. Genetic engineering 7. Space exploration 8. Animal cloning 9. Internet 10. Nuclear energy

5.7 Elements of an explanatory model

147

effects on the global ecosystem40. This judgment is based on ample and converging evidence: the public in late modern societies is aware of some aspects of environmental degradation such as urban and industrial pollution, global warming, the greenhouse effect and loss of plant and animal biodiversity.41 An indicator of perceptions of nature’s balance and the valence of the impacts on it of science and technology was built by aggregating the scores of 5 items which were shown through a principal component analysis to form a highly intercorrelated dimension. 42 The range of the indicator is 0-50, with higher values indicating perceptions of nature’s fragility and a negative influence of science (a sort of romantic view of nature scale). After collapsing the scores into three groups, as done before with other variables, the following distribution was obtained for the consolidated European sample: low “romantic” vision of nature (0–16, represented by a mere 6 % of the European population), medium “romantic” vision (17–34, represented by 64 %) and high “romantic” vision (scores 35–50, with 29.5 % of Europeans falling within this subgroup). A one-way ANOVA analysis shows that, as expected, a more romantic vision of nature and the role played by science and technology in shaping it, translates as less support for both types of pharming. In the case of plant pharming, the mean values on the variable “support” are 6.31, 5.70 and 4.86 respectively, F(2, 17841)= 196.365, p<.01) and in the case of animal pharming, the corresponding values follow the same pattern: 4.98, 4.25, 3.0, F(2, 17416)=415.750, p<.01). If we focus now on animal pharming – which meets with strong rejection in most of the countries analysed – it seems clear that the vision held of an animal will affect our willingness to support its genetic modification to obtain biopharmaceutical drugs. The survey on which this chapter is based looked at various dimensions of perceptions of the animal world with results as follows, referring once more to the set of European countries analysed. A first point of interest is the similarity perceived between animals in general (not just the primates) and human beings. To this end, respondents were asked about a number of facets in which they might resemble each other or, alternatively, differ; namely: 1. Animals think in a similar way to humans, 2. Animals feel pleasure in a similar way to humans, 3. Animals feel physical pain in a similar way to humans, 4. Animals have family bonds similar to those of humans. The mean values (agreement with 40 41 42

Marx 1998. Worcester 1993; Mertig and Dunlap 1995. The wordings of the items is as follows: “Changes made to nature with the aid of science and technology make things worse”, “Science destroys nature’s charm and mystery”, “Nature’s balance is extremely delicate, and can easily be altered by human activities”, “Science and technology have upset the balance of nature” and “Nature would be in peace and balance if it wasn’t for the action of human beings”.

148

5 Public views and attitudes to pharming

the view expressed by each statement on a scale from 0, total disagreement, to 10, total agreement) are 4.0, 5.6, 7.3 and 6.0, respectively, suggestive of a fairly close positioning of animals and humans. A summated scale of animal closeness to humans, adding up the scores on the four questions with a range of 0–40, was collapsed into the three following groups: group with a perception of low closeness animals-humans (scores 0–13, representing 14 % of the population), intermediate group (scores 14–27, with 57 % of the population falling into this category) and high closeness (scores 28–40, 28 % of the population). A one-way ANOVA analysis shows that, aligned with the expectation, a vision of greater animal-human closeness tends to be associated with less support for animal pharming: the mean values on the variable “support” are 4.39 (in the low closeness group), 4.07 (in the intermediate group) and 3.36 (in the high closeness group), F(2, 17377)= 140.510, p<.01. Of growing importance in advanced societies is the conception held about animals’ right to life; a question broached in the following manner: “Please tell me which of the following three sentences you agree with most [read and show card]. 1. It makes no sense to talk about the recognition of animals’ right to life. 2. We should recognize that animals have a right to life but not in the same way as humans. 3. We should recognize that animals have a right to life in a similar way to humans. 4. Don’t know”, with the response distribution as follows: 4.1 %, 55.2 %, 38.0 % and 2.7 %. The difference in mean values of support for animal pharming according to views on animals’ right to life is of considerable magnitude: 5.2, 4.3 and 3.2 respectively (one-way ANOVA with F(2, 17058)=339.749, p<.01). Views on animals’ closeness to humans and their relative rights exert a significant influence on willingness to accept animal pharming to some or other degree, though they are by no means the only predictor. As discussed in the first part of this chapter, most people are prepared to be more accommodating about the use of means which they would abstractly reject in the presence of ends or goals they believe to be important. It is accordingly interesting to know how far the use of animals is seen as acceptable for different purposes and, in the specific framework of scientific research, which practices are regarded as acceptable and which not. Respondents were asked with respect to a range of uses of animals for diverse purposes: “To what extent do you see it as acceptable or not to use animals for each of the things I am going to read out? Please use a scale from 0 to 10, in which 0 means you see it as totally unacceptable and 10 that you see it as totally acceptable [show card]: 1. In scientific research to improve our knowledge about life. 2. For sport hunting. 3. As food for humans. 4. In the circus. 5. In research to produce cosmetics. 6. In scientific research to improve human health. 7. To make fur coats for people to wear”. The levels of acceptance found for each (mean values on a scale from 0 to 10) in the European countries analysed are 5.4, 3.0, 7.2, 4.1, 2.3, 5.9 and 1.8 respectively. Food and, at some distance, biomedical scientific research and research for gaining new

5.7 Elements of an explanatory model

149

knowledge about life are the goals accepted, while all others are strongly rejected (including research on cosmetics). Not all forms or practices of research using animals are equally accepted, as the responses to the following question show: “Thinking more specifically about the use of animals in scientific research to improve human health, would you consider each of the uses I am going to read out acceptable or unacceptable? (for clarity here responses are embedded in the questions) [read]: 1. To observe their behaviour in the laboratory (72.8 % acceptable vs. 21.5 unacceptable and 5.7 % don’t know). 2. To take blood samples from them (81.3 vs. 14.1, plus 4.6 % don’t know). 3. To test the effects of medicines for use in humans (62.9 vs. 29.1, plus 8.0 % don’t know). 4. To conduct tests involving surgery (48.6 vs. 41.6, and 9.8 % don’t know). 5. To conduct tests involving genetic modification (32.4 acceptable vs. 54.4 unacceptable and 13.2 % don’t know”). It is revealing that of all forms or practices in the context of research to improve human health, genetic modification is the only form that is rejected by a large majority, with only a third declaring their support (mean values for genetic modification acceptance in Austria and Germany, countries particularly reticent about animal pharming, are 23.7 acceptable vs. 63.4 unacceptable and 12.9 % don’t know”, and 28.5, 59.0 and 11.2 % respectively). Risk perception is another general variable potentially related to people’s willingness or otherwise to support pharming. Some analysts have labelled the society of the late twentieth century “the risk society”43. The hard-wiring of science and technology into a wide array of systems that are fundamental for the late modern way of life – such as transportation, communication, food and energy production – has generated systems so tightly coupled that local problems or accidents can spread rapidly and affect numerous system elements, thereby magnifying risks and complicating their management44. At the same time, greater knowledge of many processes that were previously opaque to science has contributed to an increased awareness of risk factors. As the ever-lengthening list of potential hazards has become part of the cognitive schemas of lay people, it has led to a “culture of fear” and “zero tolerance” of risk. The latter attitude has flourished especially when the risk is involuntary, causes long-term effects, remains invisible to the non-scientific eye, and might contribute to diseases that are dreaded in our culture, as well as other properties that lie outside expert definition or assessment, as has been pointed out in the “psychometric paradigm” investigation of risk perceptions45. The combined effect of systemic risk (in contrast to the risk characteristic of previously only loosely connected components) and a much greater knowledge about risks (both 43 44 45

Beck 1992, 1999. Perrow 1999. Slovic 1987, 2000.

150

5 Public views and attitudes to pharming

in scope and depth) has lead to a distorted image of advanced societies as more at risk than those of the past, and to the idea that science and technology are largely to blame for this alleged increase in risk. The correlation matrixes of tables 5.4 and 5.7 shows that the perception of science and technology-related risk, and specifically the perception of science and technology risk with regard to pharming, attains mean levels matching those of preparedness to support this techno-scientific development. But it is also worth examining whether a general psychological predisposition – that of feeling personally threatened by potential risks – finds an echo in predispositions towards pharming. An aggregate indicator or summated scale of this last general dimension of risk perception was built from responses to the following question: “I am going to read a description of different situations. For each one, tell me how likely you think it is that it will happen to you in the next few years. Please use this scale from 0 to 10, in which 0 means you think it is not at all likely to happen and 10 that it is very likely to happen. You may, of course, choose any score between 0 and 10. [read and show card]. 1. Being the victim of a terrorist attack. 2. Your home being burgled. 3. Being the victim of a large-scale epidemic. 4. Global warming seriously harming your health. 5. Being physically assaulted in the street. 6. Being the victim of a natural disaster (like an earthquake or hurricane)”. This indicator has a range from 0–60. After collapsing the scores in four groups (0–15 or very low risk perception, representing 18.9 % of the European population included in our study; 16–30, 41.7 % of the population; 31–45, 31.1 %, and 46–60, 7.3 %), a one-way ANOVA analysis was conducted. The mean differences among the four groups were not significant as regards their willingness to support plant pharming (mean values of 5.47, 5.53, 5.47 and 5.33 respectively; F(3, 17393)=1.864, p=.133), and modest, though significant, for animal pharming (mean values of 3.90, 4.0, 3.94 and 3.38; F(3, 17330)=16.915, p<.01). So, rather than this general risk perception construct, the specific lay assessment of the risks posed by pharming should be part of an explanatory model of attitudes to pharming. Finally, trust in regulators is a key variable to account for positive or negative predispositions to a potentially controversial area such as pharming. The broad multidisciplinary literature concerning social capital has shown the key role played by trust in interpersonal relations, the functioning of society’s institutional fabric and even economic growth. Specifically, the economic literature contends that trust reduces the transactional costs in economic relations arising from uncertainty about contract performance (each party to a transaction lacks complete information about the other party or parties and their true intentions, especially in non-recurrent operations). The existence of trust between the parties, the argument goes, will significantly reduce the cognitive or informational requirements for a deal to be successfully concluded. One way to define the stock of trust in a soci-

5.7 Elements of an explanatory model

151

ety is the total time that agents save by not having to check what others are doing46. From a sociological standpoint, it has been argued that trust comes into its own as a core component of social conduct in situations of high complexity, as is generally the case with those involving science and technology, where it operates as a cognitive shortcut or cue for decision-making by reducing information needs. At the same time, numerous risk perception experts and other analysts of public perceptions of science claim that trust in the scientific community and the institutions regulating science or technology matters is an important predictor of attitudes to potentially conflictive scientific areas, and may even have more explanatory power than the core PUoS variable of scientific literacy47. Hence, for a large subset of the public, strong trust in the regulatory agencies may substitute or compensate for a modest level of scientific literacy when evaluating research areas like the one that concerns us. Trust can be understood as an expectation of “fair play” by individuals, social groups and organizations who recognize and respect the interests of others. This facet of trust in the strictest sense, or Trust (Trust, Vertrauen), is when we can say along with the philosopher Otto Friedrich Bollnow, well before the emergence of social capital literature, “I trust a man when I believe I can put myself in his hands”48. In our own analysis, this hypothesis or expectation about the standard behaviour of trusting social agents would imply a positive association between trust in the regulator and attitudes toward pharming. In the case of techno-scientific applications, trust in the regulator is of particular importance. Regardless of how much the public actually knows about regulatory agencies and their procedures, the belief that they are reliable may play an important role in determining acceptance or support for novel and complex applications like those of pharming. Two questions from the survey presented here measured this variable as follows: “Please tell me how much you agree or disagree with the following statement, using a scale from 0 to 10, in which 0 means you totally disagree and 10 that you totally agree. [show card]: “I trust the laws and controls that the [nationality] government has established to regulate the growing of genetically modified plants in order to obtain pharmaceutical drugs” and “I trust the laws and controls that the [nationality] government has established to regulate the production of genetically modified animals in order to obtain pharmaceutical drugs”. The resulting scores were then aggregated in five groups with a practically identical distribution in both cases (plants and animals, though only the plant distribution is given 46 47 48

Zack and Knack 2001. Sjoberg 2004; Siegrist 2000; Priest et al. 2003; Pardo and Calvo 2008. Bollnow 1958. This work of Bollnow’s is an invaluable source for the social analyst interested in tracing the evolution of certain core values and virtues through the changes taking place in the everyday language employed to refer to them in a given society.

152

5 Public views and attitudes to pharming

here): 0–2 (very low level of trust), 16.8 % of the population; 3–4 (low level of trust), 11.4 %; 5 (medium level), 15.3 %; 6–7 (high trust), 21.8 %; 8–10 (very high level of trust), 26.4 % of the population, and an additional 8.4 % of don’t know responses. In both cases (plant and animal pharming), and in consonance with the expectation, the mean differences in level of support for pharming by level of trust in national regulatory agencies are of considerable magnitude and statistically significant. For plants, the mean scores for pharming support are 3.28, 4.79, 5.28, 5.88 and 6.96 respectively, F(4, 16500)=998.335, p<.01) while the same scores for animals are 2.10, 3.52, 3.83, 4.48 and 4.93 respectively, F(4, 16420)=556.694, p<.01).

5.8 Conclusions Pharming is at present an attitudinal object embedded in a complex space of high expectations for reaching many desirable goals, and also of strong reservations of varied nature about the means to be applied, i.e., the genetic modification of plants and animals. For reasons beyond the scope of this chapter, in many advanced societies, particularly in Europe, there is suspicion, if not a stigma, regarding anything connected with genetic engineering, even if in the last few years the climate for the development of this area has improved. Biological knowledge among the general public is not only limited, it is also rife with plain misunderstandings of basic genetic concepts. But lack of knowledge is not the main barrier to a more supportive cultural and regulatory environment for applications deriving from biotechnological advances. A number of overarching views (or, as they are known in the literature, worldviews) also play a significant role in shaping attitudes. In the case of pharming, the position occupied by specific goals in a rank of preferences has a counterbalancing influence on accepting the genetic modification of plants, and more weakly, of animals. Among the worldviews, general expectations about the net effect of science at large on society and nature, images of nature and views of animals are of particular interest. Perceptions of the morality of these applications and the risks entailed are also significant influences. Trust in the regulator is another critical component of the public’s predispositions to pharming. The profile of the public in most advanced societies is characterized by low biological knowledge, reservations about the morality of genetic modification of plants and particularly animals, a dominant romantic vision of nature and a medium level of risk perception, but also of high expectations about science, a considerable level of trust in regulators and sensitivity to the specifics of the means and the trade-offs between means and ends. In this context, it is of paramount importance that the scientific community working in this area does not confine itself to the simple recipe of “more knowledge” for the public (even if this is urgently needed), but is willing to

153

5.9 Tables

adopt a more complex view of the current situation and engage in an open dialogue with citizens taking into account most of the dimensions covered here, without assuming that critical views on the desirability of particular developments are just a function of ignorance. Attitudes to pharming at this early stage are very much in flux, and a transparent and unbiased regulatory framework could play a major role in avoiding mishaps and alienating the public.

5.9 Tables Table 5.1: Awareness about pharming “Recently items have appeared in the news about the production of pharmaceutical drugs obtained through the genetic modification of plants and animals to treat diseases in animals and humans. Have you heard or read anything about this method of producing pharmaceutical drugs?” Yes No Don’t know

DK DE ES FR IE IT NL AU SE UK CZ PL IL US JP 33.1 31.8 32.1 22.1 20.9 21.2 21.0 21.7 41.9 30.8 23.6 31.4 25.5 29.7 33.6 64.7 62.7 61.6 76.5 68.9 69.9 77.3 71.8 54.4 66.4 72.8 65.1 64.7 65.5 56.3 2.1 5.4 6.3 1.5 10.2 8.9 1.8 6.5 3.7 2.8 3.6 3.5 9.8 4.7 10.2

DK: Denmark / DE: Germany / ES: Spain / FR: France / IR: Ireland / IT: Italy / NL: Netherlands / AU: Austria / SE: Sweden / UK: United Kingdom / CZ: Czech Republic / PL: Poland / IL: Israel / US: United States / JP: Japan

154

5 Public views and attitudes to pharming

Table 5.2: Evaluation of plant pharming “I am going to read out some sentences about the genetic modification of plants in order to produce pharmaceutical drugs for treating human diseases. Please tell me how much you agree or disagree with each, using a scale from 0 to 10, where 0 means you totally disagree and 10 means you totally agree.” “The genetic modification of plants in order to produce pharmaceutical drugs:” Poses no serious risks DK From 0 to 2 16.2 From 3 to 4 14.3 5 13.1 From 6 to 7 17.5 From 8 to 10 35.3 Don’t know 3.6 Total disagree (0–4) 30.4 Total agree (6–10) 52.8 Mean 5.8 Is immoral DK From 0 to 2 50.8 From 3 to 4 17.5 5 9.0 From 6 to 7 8.5 From 8 to 10 12.3 Don’t know 2.0 Total disagree (0–4) 68.3 Total agree (6–10) 20.7 Mean 3.2 Is very useful DK From 0 to 2 4.5 From 3 to 4 4.9 5 9.9 From 6 to 7 20.3 From 8 to 10 55.1 Don’t know 5.4 Total disagree (0–4) 9.3 Total agree (6–10) 75.4 Mean 7.4 Is reckless DK From 0 to 2 52.8 From 3 to 4 16.9 5 9.5 From 6 to 7 7.8 From 8 to 10 10.3 Don’t know 2.7 Total disagree (0–4) 69.8 Total agree (6–10) 18.1 Mean 3.0

DE 26.3 16.1 18.0 16.0 17.2 6.5 42.4 33.1 4.5

ES 13.4 9.6 19.4 18.5 21.7 17.4 23.0 40.3 5.4

FR 22.0 12.9 17.7 16.0 25.8 5.7 34.9 41.8 5.2

IE 20.2 14.3 13.8 14.2 15.9 21.5 34.6 30.1 4.8

IT 17.4 19.3 16.0 22.0 9.7 15.7 36.6 31.6 4.6

NL 11.3 11.7 12.3 29.7 30.8 4.1 23.1 60.5 6.0

AU 31.1 19.2 14.2 12.1 14.3 9.1 50.3 26.4 4.0

SE 22.0 14.7 16.5 17.4 18.7 10.8 36.6 36.1 4.9

UK 19.1 14.4 18.0 14.9 22.9 10.8 33.5 37.7 5.2

CZ 17.1 12.2 16.9 15.8 31.3 6.6 29.3 47.1 5.7

PL 17.4 12.1 15.1 17.5 24.8 13.0 29.6 42.3 5.4

IL 12.7 10.5 11.2 18.1 39.7 7.9 23.2 57.7 6.3

US 16.1 9.8 15.3 15.3 37.8 5.7 25.9 53.1 6.0

JP 19.0 26.7 22.2 15.1 10.5 6.4 45.7 25.6 4.4

DE 25.3 17.7 16.0 13.1 22.1 5.8 43.0 35.2 4.8

ES 31.7 19.6 18.0 12.2 9.8 8.6 51.3 22.0 3.8

FR 41.8 14.3 14.1 9.7 17.1 3.1 56.0 26.8 3.9

IE 27.3 17.2 12.5 13.6 13.3 16.1 44.5 26.9 4.2

IT 20.0 21.5 17.1 18.7 14.2 8.6 41.5 32.9 4.7

NL 38.1 28.3 11.6 12.0 7.6 2.4 66.4 19.6 3.5

AU 18.8 17.0 17.0 15.9 26.7 4.7 35.7 42.6 5.4

SE 35.4 17.0 14.9 10.7 15.8 6.2 52.4 26.4 4.0

UK 42.5 15.8 13.8 7.7 12.3 7.8 58.3 20.1 3.5

CZ 41.7 18.1 16.0 7.6 13.1 3.4 59.8 20.7 3.6

PL 32.4 18.6 15.3 8.8 15.4 9.5 51.0 24.2 4.1

IL 41.4 14.3 9.6 12.1 17.7 4.9 55.6 29.9 3.9

US 55.4 12.1 10.0 6.2 13.0 3.3 67.5 19.2 2.9

JP 8.5 18.8 29.9 20.9 17.3 4.7 27.3 38.2 5.4

DE 14.0 9.6 20.1 23.7 26.7 5.8 23.7 50.4 5.7

ES 5.0 6.7 16.1 27.3 32.0 12.9 11.7 59.3 6.5

FR 14.8 7.1 16.3 22.0 35.8 4.1 21.9 57.8 6.0

IE 8.3 8.8 12.4 25.5 28.6 16.4 17.2 54.1 6.3

IT 11.2 13.5 15.0 33.9 16.5 10.0 24.6 50.4 5.5

NL 4.3 5.6 12.1 36.9 37.8 3.4 9.9 74.7 6.8

AU 19.1 16.7 16.4 19.3 20.9 7.5 35.8 40.2 5.1

SE 6.2 4.3 15.1 24.1 39.4 10.9 10.5 63.5 6.8

UK 9.9 5.3 14.7 24.9 38.7 6.6 15.2 63.5 6.5

CZ 8.3 7.5 16.7 20.7 42.8 3.9 15.8 63.5 6.7

PL 8.1 7.5 16.7 22.8 36.1 8.8 15.6 58.9 6.5

IL 7.7 8.9 10.4 22.3 43.4 7.3 16.6 65.7 6.8

US 9.7 4.8 11.9 17.8 52.0 3.8 14.5 69.8 7.1

JP 6.8 11.6 31.6 29.3 16.1 4.7 18.4 45.4 5.6

DE 9.8 11.3 16.7 22.3 34.5 5.3 21.1 56.8 6.2

ES 25.0 17.2 19.5 16.0 11.6 10.7 42.2 27.6 4.3

FR 20.2 12.3 19.2 18.7 25.5 4.0 32.5 44.3 5.3

IE 24.9 20.0 13.5 11.5 12.5 17.6 44.9 24.0 4.2

IT 13.7 17.6 16.3 24.4 18.7 9.2 31.3 43.2 5.3

NL 36.7 28.9 14.1 10.9 6.7 2.8 65.6 17.5 3.5

AU 12.6 15.1 14.9 19.5 33.4 4.6 27.6 52.9 6.0

SE 33.3 18.0 15.6 10.7 15.1 7.3 51.3 25.8 4.1

UK 38.2 18.5 13.9 10.3 11.9 7.2 56.7 22.2 3.7

CZ 40.4 18.5 14.9 8.6 13.6 4.0 59.0 22.1 3.7

PL 29.6 19.7 17.8 9.3 13.3 10.2 49.3 22.7 4.1

IL 36.2 17.5 10.9 14.1 14.3 7.0 53.7 28.4 3.9

US 49.0 12.2 13.2 6.4 15.0 4.2 61.2 21.4 3.3

JP 12.3 23.8 31.7 15.3 12.0 5.0 36.1 27.3 4.9

DK: Denmark / DE: Germany / ES: Spain / FR: France / IR: Ireland / IT: Italy / NL: Netherlands / AU: Austria / SE: Sweden / UK: United Kingdom / CZ: Czech Republic / PL: Poland / IL: Israel / US: United States / JP: Japan Continued on next page

155

5.9 Tables Goes against nature

DK DE ES FR IE IT NL AU SE UK CZ From 0 to 2 20.8 9.4 16.1 15.0 14.3 10.5 16.2 11.5 15.5 21.8 23.2 From 3 to 4 9.3 9.0 12.5 8.1 12.3 13.3 12.9 12.2 8.3 10.1 12.0 5 11.1 14.8 19.4 14.0 14.5 16.2 12.3 12.8 12.5 13.8 17.5 From 6 to 7 13.8 19.8 22.1 14.8 18.1 28.1 29.7 19.9 16.9 17.3 16.6 From 8 to 10 43.0 43.6 22.4 45.8 27.3 26.4 27.0 40.9 43.0 31.8 28.2 Don’t know 2.1 3.5 7.5 2.3 13.5 5.4 2.0 2.7 3.9 5.3 2.5 Total disagree (0–4) 30.1 18.4 28.6 23.1 26.6 23.8 29.0 23.7 23.8 31.9 35.2 Total agree (6–10) 56.8 63.4 44.6 60.6 45.4 54.6 56.6 60.8 59.8 49.1 44.8 Mean 6.0 6.7 5.4 6.4 5.7 5.9 5.7 6.5 6.4 5.6 5.3 Is a product of the arrogance of scientists DK DE ES FR IE IT NL AU SE UK CZ From 0 to 2 49.1 23.9 29.3 34.3 22.3 19.8 35.9 18.8 38.2 35.3 48.4 From 3 to 4 12.6 15.8 12.1 11.5 14.9 19.4 23.9 18.9 12.7 13.2 14.7 5 12.5 18.4 19.6 16.5 13.9 16.3 13.8 15.4 18.3 15.2 14.4 From 6 to 7 9.2 15.8 15.2 11.2 14.9 21.9 15.1 18.2 8.2 11.8 7.6 From 8 to 10 12.7 18.0 12.9 20.5 15.9 13.5 8.4 20.2 11.0 15.9 9.3 Don’t know 3.9 8.1 10.9 6.0 18.0 9.1 2.8 8.5 11.6 8.8 5.6 Total disagree (0–4) 61.7 39.8 41.4 45.8 37.2 39.2 59.8 37.6 50.9 48.4 63.2 Total agree (6–10) 21.9 33.8 28.1 31.7 30.8 35.4 23.5 38.4 19.2 27.6 16.9 Mean 3.3 4.7 4.2 4.3 4.7 4.7 3.7 5.1 3.6 4.1 3.1 Is playing God DK DE ES FR IE IT NL AU SE UK CZ From 0 to 2 50.3 20.8 35.4 46.6 21.5 23.9 39.7 16.9 35.6 32.0 45.7 From 3 to 4 7.7 11.2 11.1 7.7 12.0 17.1 19.0 15.3 10.2 9.1 12.4 5 8.5 13.4 15.5 11.9 11.3 13.9 11.7 11.5 11.5 11.1 10.4 From 6 to 7 7.6 15.2 13.5 6.7 13.2 19.2 12.9 16.3 11.5 12.0 8.1 From 8 to 10 22.5 31.3 14.4 20.5 23.5 16.7 12.6 34.3 23.3 28.2 16.3 Don’t know 3.3 8.2 10.1 6.6 18.5 9.3 4.1 5.8 8.0 7.5 7.1 Total disagree (0–4) 58.0 32.0 46.5 54.3 33.6 40.9 58.7 32.2 45.8 41.2 58.1 Total agree (6–10) 30.1 46.5 27.9 27.2 36.7 35.9 25.5 50.5 34.7 40.2 24.3 Mean 3.7 5.5 3.9 3.7 5.1 4.7 3.7 5.8 4.4 4.9 3.5 Is a product of the interests of a small number of pharmaceutical multinationals DK DE ES FR IE IT NL AU SE UK CZ From 0 to 2 15.3 9.0 7.9 6.6 10.0 7.6 16.1 9.1 9.8 10.2 13.3 From 3 to 4 10.8 13.3 9.8 6.1 9.6 14.1 18.7 15.4 9.5 10.4 10.6 5 16.8 15.1 24.2 15.1 14.4 15.3 18.5 14.4 18.6 19.4 20.7 From 6 to 7 19.6 22.8 22.7 22.0 20.7 29.3 21.8 20.1 18.3 18.9 16.8 From 8 to 10 30.2 32.1 22.2 42.2 25.6 18.6 15.8 31.8 28.8 25.1 21.9 Don’t know 7.4 7.8 13.3 8.0 19.7 15.1 9.2 9.1 15.0 15.9 16.8 Total disagree (0–4) 26.0 22.3 17.7 12.7 19.6 21.7 34.7 24.5 19.3 20.7 23.9 Total agree (6–10) 49.8 54.8 44.9 64.2 46.3 47.8 37.6 52.0 47.2 44.0 38.6 Mean 5.8 6.2 5.8 6.8 6.0 5.7 5.0 6.2 6.1 5.8 5.6 Should be supported DK DE ES FR IE IT NL AU SE UK CZ From 0 to 2 13.6 25.3 7.2 23.4 15.6 14.7 9.8 29.6 21.3 16.8 11.8 From 3 to 4 9.1 14.1 5.9 11.9 9.2 13.3 9.4 19.9 10.9 8.7 7.4 5 13.8 18.2 19.7 21.7 15.1 17.8 19.4 15.0 17.9 17.6 18.7 From 6 to 7 20.9 19.8 27.8 17.6 20.8 29.7 34.3 14.8 19.0 20.4 20.3 From 8 to 10 38.7 17.4 28.4 21.0 18.7 12.9 24.8 14.0 23.9 29.5 37.8 Don’t know 3.9 5.2 11.1 4.3 20.5 11.6 2.3 6.7 7.1 7.0 4.0 Total disagree (0–4) 22.7 39.4 13.1 35.3 24.9 28.0 19.2 49.5 32.2 25.6 19.2 Total agree (6–10) 59.6 37.2 56.1 38.6 39.5 42.6 59.1 28.8 42.9 49.9 58.1 Mean 6.2 4.6 6.2 4.9 5.3 5.1 5.9 4.2 5.2 5.7 6.3

PL 16.4 13.1 16.9 18.1 28.4 7.1 29.5 46.5 5.7

IL 22.5 10.9 11.7 14.8 35.7 4.4 33.4 50.5 5.6

US 34.0 10.9 16.3 10.1 25.4 3.3 44.9 35.5 4.6

JP 6.9 14.7 26.0 25.9 22.7 3.7 21.6 48.6 5.8

PL 30.4 16.7 15.9 12.2 13.1 11.7 47.0 25.3 4.1

IL 31.8 14.7 11.2 14.5 20.7 7.0 46.6 35.2 4.4

US 44.0 10.5 16.1 7.5 16.3 5.7 54.5 23.8 3.6

JP 12.8 21.7 33.0 15.1 12.7 4.7 34.5 27.8 4.9

PL 30.9 13.3 11.9 10.6 19.7 13.6 44.2 30.3 4.4

IL 30.6 8.7 9.5 13.7 29.6 6.5 39.4 43.2 5.0

US 45.5 8.7 9.5 6.9 25.3 4.1 54.2 32.2 4.0

JP 23.6 18.1 31.3 10.1 9.6 7.3 41.7 19.7 4.3

PL 6.6 8.9 16.1 23.3 28.7 16.4 15.5 52.0 6.4

IL 12.6 10.8 13.4 18.4 37.5 7.4 23.4 55.9 6.2

US 19.8 9.7 22.5 13.3 21.4 13.2 29.5 34.7 5.1

JP 10.1 18.8 23.4 17.1 10.1 20.6 28.9 27.2 5.0

PL 11.5 9.9 17.1 23.1 28.9 9.5 21.4 52.0 6.0

IL 11.8 10.0 12.0 19.3 39.4 7.5 21.8 58.6 6.4

US 14.8 7.2 14.4 16.4 43.3 3.8 22.0 59.7 6.4

JP 11.5 18.4 39.4 17.8 7.7 5.2 29.9 25.5 4.8

DK: Denmark / DE: Germany / ES: Spain / FR: France / IR: Ireland / IT: Italy / NL: Netherlands / AU: Austria / SE: Sweden / UK: United Kingdom / CZ: Czech Republic / PL: Poland / IL: Israel / US: United States / JP: Japan

156

5 Public views and attitudes to pharming

Table 5.3: Trust in the regulatory agencies dealing with plant pharming and disposition to take

plant biopharmaceuticals

“Please tell me how much you agree or disagree with the following statements, using a scale from 0 to 10, in which 0 means you totally disagree and 10 that you totally agree.” I trust the laws and controls that the (NATIONALITY) government has established to regulate the growing of genetically modified plants in order to obtain pharmaceutical drugs DK DE ES FR IE IT NL AU SE UK CZ PL IL US JP From 0 to 2 10.5 24.3 9.9 21.3 17.7 14.9 8.3 22.2 12.3 22.8 15.5 21.1 14.1 23.8 19.4 From 3 to 4 8.2 15.9 5.5 8.9 9.9 16.0 9.1 16.6 12.5 8.9 9.7 15.5 10.9 9.0 19.8 5 10.3 15.9 18.3 16.3 11.5 19.3 11.5 16.2 13.5 16.2 16.0 18.1 12.3 14.0 26.1 From 6 to 7 17.5 18.9 27.9 17.4 21.2 28.8 35.8 19.2 20.2 18.7 20.5 15.7 19.0 19.9 14.4 From 8 to 10 49.1 21.0 26.1 28.0 23.1 11.4 32.7 21.0 37.4 23.4 30.8 14.1 37.1 29.1 9.2 Don’t know 4.4 4.0 12.3 8.3 16.6 9.6 2.5 4.7 4.1 10.1 7.4 15.5 6.5 4.2 11.1 Total disagree (0–4) 18.7 40.2 15.4 30.1 27.7 30.9 17.4 38.8 24.7 31.7 25.2 36.6 25.1 32.8 39.2 Total agree (6–10) 66.6 39.9 54.0 45.4 44.2 40.2 68.5 40.2 57.7 42.1 51.3 29.8 56.1 49.0 23.6 Mean 6.8 4.8 6.0 5.3 5.4 5.0 6.3 5.0 6.1 5.1 5.8 4.6 6.1 5.3 4.4

“Please tell me how much you agree or disagree with the following statements, using a scale from 0 to 10, in which 0 means you totally disagree and 10 that you totally agree” I am quite prepared to take pharmaceutical drugs produced from genetically modified plants DK DE ES FR IE IT NL AU SE UK CZ PL IL From 0 to 2 15.0 29.5 11.6 32.4 20.1 23.0 9.0 38.8 18.0 16.8 14.3 16.5 15.6 From 3 to 4 5.1 7.9 4.5 7.9 8.4 10.6 4.5 12.5 7.7 4.6 6.4 6.2 6.3 5 5.9 13.0 15.8 12.1 10.4 12.2 8.5 10.7 12.7 11.1 12.6 13.4 8.6 From 6 to 7 14.4 14.5 24.4 14.7 17.6 25.9 29.3 12.8 13.3 17.2 17.1 18.1 18.6 From 8 to 10 57.3 27.3 29.6 28.4 26.5 15.8 44.6 17.4 42.4 42.4 43.7 33.6 44.8 Don’t know 2.3 7.8 14.1 4.5 16.9 12.5 4.0 7.8 6.0 7.8 5.9 12.2 5.9 Total disagree (0–4) 20.1 37.4 16.2 40.3 28.6 33.6 13.5 51.3 25.7 21.5 20.7 22.7 22.0 Total agree (6–10) 71.7 41.8 54.0 43.1 44.1 41.8 74.0 30.3 55.7 59.6 60.8 51.7 63.4 Mean 7.0 4.9 6.0 4.8 5.4 4.8 6.8 3.8 6.1 6.3 6.5 6.0 6.5

US 20.5 5.3 11.4 14.9 43.2 4.7 25.8 58.1 6.1

JP 30.9 18.8 20.0 14.6 10.3 5.5 49.7 24.9 4.0

DK: Denmark / DE: Germany / ES: Spain / FR: France / IR: Ireland / IT: Italy / NL: Netherlands / AU: Austria / SE: Sweden / UK: United Kingdom / CZ: Czech Republic / PL: Poland / IL: Israel / US: United States / JP: Japan

5.9 Tables

Table 5.4: Pearson’s r correlation between evaluative criteria and support for plant pharming Correlations between “support for plant pharming” and “evaluative criteria” DK Poses no serious risks Is immoral Is very useful

DE

ES

FR

IE

IT

NL

AU

SE

UK

CZ

PL

IL

US

JP

0.5

0.5

0.4

0.5

0.6

0.6

0.6

0.6

0.5

0.5

0.6

0.5

0.5

0.5

0.4

-0.5

-0.6

-0.5

-0.5

-0.4

-0.5

-0.5

-0.6

-0.5

-0.6

-0.7

-0.4

-0.3

-0.5

-0.4

0.6

0.7

0.7

0.7

0.7

0.8

0.7

0.8

0.6

0.7

0.8

0.7

0.6

0.7

0.6

Is reckless

-0.6

-0.5

-0.5

-0.6

-0.5

-0.5

-0.6

-0.5

-0.6

-0.6

-0.7

-0.6

-0.4

-0.6

-0.5

Goes against nature

-0.3

-0.6

-0.4

-0.4

-0.3

-0.4

-0.4

-0.5

-0.4

-0.5

-0.6

-0.4

-0.2

-0.5

-0.4

Is a product of the arrogance of scientists

-0.5

-0.6

-0.4

-0.5

-0.4

-0.4

-0.5

-0.5

-0.5

-0.6

-0.6

-0.5

-0.3

-0.5

-0.5

Is a product of the interests of a small number of pharmaceutical multinationals

-0.3

-0.4

-0.2

-0.3

-0.2

-0.3

-0.3

-0.3

-0.3

-0.3

-0.3

-0.1

-0.1

-0.2

-0.2

Is playing God

-0.4

-0.5

-0.5

-0.4

-0.5

-0.5

-0.4

-0.5

-0.5

-0.6

-0.6

-0.5

-0.3

-0.6

-0.3

DK: Denmark / DE: Germany / ES: Spain / FR: France / IR: Ireland / IT: Italy / NL: Netherlands / AU: Austria / SE: Sweden / UK: United Kingdom / CZ: Czech Republic / PL: Poland / IL: Israel / US: United States / JP: Japan

157

158

5 Public views and attitudes to pharming

Table 5.5: Evaluation of animal pharming “Can you please tell me how much you agree or disagree with each of the statements I am going to read out? Please use a scale from 0 to 10, where 0 means you totally disagree, and 10 means you totally agree. You may, of course, choose any score between 0 and 10.” “The genetic modification of animals in order to produce pharmaceutical drugs:” Poses no serious risks DK From 0 to 2 31.0 From 3 to 4 19.5 5 12.2 From 6 to 7 14.2 From 8 to 10 18.3 Don’t know 4.8 Total disagree (0–4) 50.5 Total agree (6–10) 32.5 Mean 4.4 Is immoral DK From 0 to 2 25.6 From 3 to 4 14.1 5 10.5 From 6 to 7 13.5 From 8 to 10 33.8 Don’t know 2.5 Total disagree (0–4) 39.8 Total agree (6–10) 47.3 Mean 5.4 Is very useful DK From 0 to 2 9.0 From 3 to 4 8.1 5 14.1 From 6 to 7 23.7 From 8 to 10 37.9 Don’t know 7.2 Total disagree (0–4) 17.1 Total agree (6–10) 61.6 Mean 6.4 Is reckless DK From 0 to 2 27.0 From 3 to 4 14.6 5 10.7 From 6 to 7 15.8 From 8 to 10 29.1 Don’t know 2.9 Total disagree (0–4) 41.6 Total agree (6–10) 44.9 Mean 5.2

DE 39.5 17.2 14.3 12.7 10.6 5.7 56.7 23.3 3.5

ES 22.7 11.5 22.6 13.2 9.8 20.1 34.2 23.0 4.2

FR 41.9 15.5 15.1 9.8 10.8 6.9 57.4 20.6 3.5

IE 32.6 18.9 11.3 10.4 10.0 16.7 51.5 20.4 3.6

IT 29.3 19.3 16.6 15.8 4.7 14.3 48.5 20.6 3.8

NL 29.1 26.8 14.8 16.2 8.9 4.2 55.9 25.0 4.0

AU 48.3 18.3 9.9 7.1 10.4 6.0 66.6 17.5 3.0

SE 45.4 20.0 10.5 8.0 8.4 7.8 65.4 16.4 3.1

UK 37.2 17.3 15.2 9.6 10.0 10.7 54.5 19.6 3.5

CZ 29.9 16.6 17.6 11.6 15.4 9.0 46.5 27.0 4.2

PL 32.1 16.7 14.7 11.1 13.1 12.3 48.7 24.3 3.9

IL 30.8 18.1 12.8 15.1 14.3 8.9 48.8 29.4 4.2

US 41.5 13.7 13.6 10.9 13.8 6.6 55.2 24.7 3.5

JP 29.6 26.3 18.5 11.6 8.7 5.3 55.9 20.3 3.9

DE 11.6 11.1 15.1 19.3 39.3 3.7 22.7 58.5 6.4

ES 16.8 16.5 21.0 19.6 16.5 9.6 33.2 36.1 5.0

FR 15.3 10.9 13.7 14.9 41.2 3.9 26.2 56.1 6.3

IE 16.2 15.3 11.8 17.0 28.2 11.4 31.5 45.3 5.6

IT 11.5 16.8 15.2 25.1 24.7 6.7 28.4 49.8 5.7

NL 14.4 18.4 14.4 23.0 27.7 2.0 32.9 50.7 5.6

AU 11.2 10.3 10.6 20.2 45.2 2.4 21.5 65.4 6.7

SE 15.9 11.0 9.6 17.4 43.0 3.1 26.9 60.4 6.3

UK 17.8 11.9 13.6 14.7 35.3 6.7 29.6 50.1 5.9

CZ 21.5 15.3 16.4 14.0 28.6 4.2 36.8 42.6 5.4

PL 15.3 13.5 14.6 14.5 32.8 9.3 28.8 47.3 5.9

IL 20.4 12.9 10.9 17.9 34.8 3.1 33.3 52.7 5.7

US 23.7 11.7 13.0 14.3 34.8 2.7 35.4 49.1 5.6

JP 7.4 14.4 24.6 23.2 26.3 3.9 21.8 49.5 5.9

DE 23.9 11.8 21.0 20.8 15.4 7.0 35.7 36.3 4.7

ES 8.2 7.9 21.4 28.1 19.4 15.2 16.0 47.4 5.8

FR 30.0 11.7 20.9 19.3 11.3 6.8 41.7 30.6 4.2

IE 15.0 11.4 14.4 24.1 21.2 13.8 26.4 45.3 5.4

IT 17.7 14.7 16.8 30.2 10.4 10.2 32.4 40.5 4.9

NL 18.4 14.0 19.5 30.8 14.1 3.3 32.3 44.9 5.0

AU 30.6 20.3 13.3 14.5 12.5 8.8 50.9 27.0 4.0

SE 13.3 7.3 17.8 22.3 26.7 12.7 20.6 49.0 5.8

UK 21.6 11.6 17.2 22.0 18.9 8.7 33.2 40.9 4.9

CZ 16.8 12.5 20.5 19.1 24.9 6.3 29.3 44.0 5.4

PL 19.0 12.7 20.6 17.8 18.9 11.1 31.6 36.6 5.0

IL 17.2 12.1 12.4 19.2 29.2 9.9 29.4 48.4 5.6

US 23.1 9.1 18.7 18.0 26.2 4.9 32.2 44.2 5.2

JP 13.2 16.0 34.6 21.3 9.5 5.5 29.2 30.8 4.9

DE 6.4 8.2 12.0 24.4 43.9 5.1 14.6 68.3 6.9

ES 12.7 15.3 21.9 22.3 18.3 9.5 28.0 40.6 5.3

FR 6.9 9.3 14.0 22.0 42.5 5.3 16.2 64.5 6.8

IE 16.3 16.5 12.0 16.9 24.1 14.2 32.8 41.0 5.4

IT 8.0 13.7 16.0 28.4 26.3 7.5 21.8 54.7 6.0

NL 18.1 22.0 16.9 19.6 21.0 2.4 40.1 40.6 5.1

AU 7.6 10.8 10.5 22.7 43.8 4.6 18.4 66.5 6.9

SE 17.7 13.8 14.0 15.3 32.8 6.4 31.4 48.2 5.7

UK 19.1 14.3 14.6 15.2 29.2 7.6 33.4 44.3 5.5

CZ 24.4 19.5 14.5 12.7 24.4 4.5 43.9 37.1 4.9

PL 12.7 15.2 15.6 16.6 28.6 11.3 28.0 45.2 5.8

IL 20.1 16.1 14.5 16.9 26.6 5.7 36.2 43.5 5.3

US 23.7 14.5 16.5 12.1 30.4 2.8 38.2 42.5 5.3

JP 7.9 17.6 30.7 18.9 20.3 4.6 25.5 39.2 5.5

DK: Denmark / DE: Germany / ES: Spain / FR: France / IR: Ireland / IT: Italy / NL: Netherlands / AU: Austria / SE: Sweden / UK: United Kingdom / CZ: Czech Republic / PL: Poland / IL: Israel / US: United States / JP: Japan Continued on next page

159

5.9 Tables Goes against nature

DK DE ES FR IE IT NL AU SE UK CZ From 0 to 2 12.9 6.8 8.4 5.3 11.3 6.4 8.0 6.9 8.6 12.0 14.4 From 3 to 4 7.5 7.2 12.0 6.3 9.4 12.4 8.8 10.0 5.4 6.5 11.1 5 8.1 10.0 17.0 9.5 10.1 13.7 8.6 8.4 7.6 9.7 14.3 From 6 to 7 14.5 18.0 23.9 18.5 18.8 28.3 28.8 19.3 15.7 17.8 17.4 From 8 to 10 55.0 55.2 32.1 57.6 40.5 34.2 44.4 53.7 59.9 48.1 40.0 Don’t know 2.1 2.7 6.8 2.9 9.9 5.0 1.4 1.7 2.9 5.8 2.8 Total disagree (0–4) 20.3 14.1 20.3 11.6 20.7 18.7 16.8 16.9 14.0 18.5 25.4 Total agree (6–10) 69.6 73.2 55.9 76.0 59.3 62.6 73.2 73.0 75.5 66.0 57.5 Mean 7.0 7.4 6.2 7.5 6.5 6.4 6.8 7.3 7.4 6.9 6.3 Is a product of the arrogance of scientists DK DE ES FR IE IT NL AU SE UK CZ From 0 to 2 41.4 19.1 22.4 27.9 16.6 14.4 25.7 13.5 30.7 23.7 36.7 From 3 to 4 11.7 12.3 11.4 10.3 13.8 19.1 20.4 15.8 12.0 11.3 16.3 5 13.1 16.6 20.6 15.7 14.5 15.7 14.5 14.6 15.9 15.0 15.4 From 6 to 7 11.4 17.4 19.3 12.8 16.7 23.7 20.0 18.5 13.9 13.7 10.8 From 8 to 10 18.5 26.7 15.7 27.0 24.4 18.4 16.6 30.2 18.2 26.6 14.4 Don’t know 3.9 7.9 10.6 6.3 14.0 8.6 2.8 7.5 9.3 9.7 6.3 Total disagree (0–4) 53.1 31.4 33.7 38.2 30.3 33.5 46.1 29.3 42.7 35.0 53.1 Total agree (6–10) 29.9 44.0 35.1 39.8 41.2 42.2 36.6 48.6 32.2 40.3 25.2 Mean 4.0 5.4 4.8 5.0 5.4 5.2 4.6 5.9 4.4 5.2 3.9 Is playing God DK DE ES FR IE IT NL AU SE UK CZ From 0 to 2 43.5 18.3 29.6 40.2 15.6 19.3 32.0 11.7 27.1 20.2 37.6 From 3 to 4 7.1 8.3 10.6 6.5 10.2 15.1 14.6 11.8 7.3 7.1 13.1 5 6.8 11.0 16.1 10.4 11.2 13.6 10.9 9.4 10.0 9.2 10.3 From 6 to 7 7.4 15.6 16.5 8.7 13.2 19.6 15.4 16.8 12.1 13.6 10.9 From 8 to 10 31.4 39.1 17.4 26.5 34.3 23.8 23.5 44.4 36.1 41.8 20.8 Don’t know 3.8 7.8 9.8 7.6 15.5 8.7 3.6 5.8 7.5 8.0 7.3 Total disagree (0–4) 50.6 26.5 40.2 46.7 25.7 34.3 46.6 23.6 34.4 27.3 50.7 Total agree (6–10) 38.8 54.7 33.9 35.3 47.5 43.3 38.9 61.3 48.2 55.5 31.7 Mean 4.4 6.1 4.4 4.4 6.1 5.3 4.6 6.6 5.5 6.2 4.1 Is a product of the interests of a small number of pharmaceutical multinationals DK DE ES FR IE IT NL AU SE UK CZ From 0 to 2 16.6 9.3 7.8 6.5 9.0 7.1 14.8 6.6 10.7 10.0 11.6 From 3 to 4 10.6 9.2 8.2 6.7 10.8 11.0 20.5 15.2 9.3 10.0 12.6 5 14.2 14.0 22.5 13.7 12.8 13.4 18.5 13.9 15.5 16.1 18.2 From 6 to 7 21.5 21.6 25.2 23.2 19.5 30.7 19.5 21.7 17.9 18.7 17.8 From 8 to 10 31.1 38.4 21.9 40.8 28.6 22.0 17.3 34.5 34.1 29.8 22.0 Don’t know 6.0 7.5 14.3 9.1 19.4 15.8 9.4 8.1 12.4 15.4 17.7 Total disagree (0–4) 27.2 18.5 16.1 13.2 19.8 18.2 35.3 21.8 20.1 20.0 24.2 Total agree (6–10) 52.6 60.0 47.1 64.0 48.1 52.7 36.8 56.2 52.1 48.5 39.8 Mean 5.8 6.5 5.9 6.9 6.2 6.1 5.1 6.4 6.3 6.2 5.6 Will transmit diseases from animals to humans DK DE ES FR IE IT NL AU SE UK CZ From 0 to 2 26.6 8.4 12.6 9.5 10.3 8.2 14.1 8.1 15.3 12.2 19.9 From 3 to 4 15.3 10.6 10.5 7.3 9.8 12.1 17.9 11.6 11.5 9.3 12.8 5 16.7 16.2 21.4 16.0 11.3 14.9 18.5 13.9 14.6 16.7 15.9 From 6 to 7 15.3 21.2 20.8 19.9 15.4 27.3 25.2 20.7 16.4 19.0 14.3 From 8 to 10 15.5 30.5 12.9 34.5 27.1 18.5 14.0 33.9 23.7 28.3 21.6 Don’t know 10.7 13.1 21.8 12.8 26.0 19.0 10.3 11.8 18.6 14.6 15.5 Total disagree (0–4) 41.9 18.9 23.2 16.8 20.1 20.3 32.0 19.7 26.8 21.4 32.7 Total agree (6–10) 30.8 51.7 33.6 54.4 42.6 45.8 39.2 54.6 40.0 47.2 35.9 Mean 4.5 6.2 5.1 6.4 6.1 5.8 5.1 6.5 5.5 6.0 5.1

PL 8.1 9.5 13.3 17.3 45.3 6.5 17.6 62.6 6.9

IL 15.3 10.5 9.3 13.6 47.7 3.7 25.7 61.3 6.5

US 17.3 8.9 13.5 14.1 44.0 2.2 26.2 58.1 6.4

JP 5.2 9.3 21.7 29.6 30.7 3.4 14.5 60.3 6.4

PL 21.5 13.2 15.9 15.0 21.0 13.5 34.6 36.0 5.1

IL 23.5 13.0 10.7 18.5 27.5 6.8 36.5 46.0 5.3

US 30.1 12.0 15.8 10.5 26.6 5.1 42.1 37.1 4.8

JP 9.0 16.2 30.7 20.2 19.3 4.6 25.2 39.5 5.5

PL 22.1 11.5 10.7 12.9 29.6 13.2 33.7 42.5 5.4

IL 23.3 8.4 8.9 12.9 39.2 7.3 31.8 52.1 5.8

US 27.0 8.9 11.5 7.4 40.8 4.5 35.9 48.2 5.7

JP 17.9 16.2 30.4 11.9 15.0 8.6 34.1 26.9 4.8

PL 6.1 7.7 17.6 21.4 30.6 16.5 13.9 52.0 6.5

IL 12.3 11.5 12.1 19.3 36.2 8.6 23.7 55.5 6.3

US 16.5 11.1 20.2 12.8 24.9 14.5 27.6 37.7 5.5

JP 4.7 11.8 35.1 18.6 20.2 9.5 16.5 38.8 5.8

PL 7.2 7.2 13.1 20.3 34.9 17.4 14.4 55.1 6.7

IL 11.5 10.3 13.1 16.6 31.4 17.0 21.9 48.0 6.1

US 16.2 10.6 15.1 13.5 29.6 15.0 26.8 43.1 5.8

JP 4.5 8.6 25.9 29.2 24.8 7.0 13.1 54.0 6.1

DK: Denmark / DE: Germany / ES: Spain / FR: France / IR: Ireland / IT: Italy / NL: Netherlands / AU: Austria / SE: Sweden / UK: United Kingdom / CZ: Czech Republic / PL: Poland / IL: Israel / US: United States / JP: Japan Continued on next page

160

5 Public views and attitudes to pharming

Is incompatible with the inherent dignity of animals DK DE ES FR IE IT NL AU SE UK CZ PL IL US JP From 0 to 2 20.5 6.6 7.8 8.9 10.5 6.0 9.5 7.4 10.5 12.5 15.9 6.5 13.9 17.1 4.5 From 3 to 4 11.3 8.7 9.1 7.9 12.2 11.5 14.4 10.4 6.5 8.4 13.5 9.0 11.3 9.1 8.0 5 13.1 17.8 28.4 16.1 13.0 16.4 14.4 11.5 12.1 15.7 20.5 16.1 12.0 15.0 23.2 From 6 to 7 14.9 18.3 23.7 17.1 19.2 29.6 25.8 20.1 17.6 16.6 14.8 20.7 17.9 16.0 31.1 From 8 to 10 36.2 44.1 19.2 45.0 33.2 29.9 33.0 47.8 50.5 37.8 30.3 37.7 39.1 37.5 29.4 Don’t know 4.1 4.5 11.8 5.0 11.8 6.6 2.8 2.9 2.9 9.0 5.0 10.0 5.8 5.3 3.8 Total disagree (0–4) 31.7 15.3 17.0 16.7 22.8 17.4 23.9 17.8 16.9 21.0 29.4 15.5 25.2 26.2 12.5 Total agree (6–10) 51.2 62.4 42.9 62.2 52.4 59.5 58.9 67.8 68.1 54.3 45.1 58.3 57.0 53.5 60.5 Mean 5.8 6.9 5.7 6.8 6.2 6.3 6.1 7.0 6.9 6.4 5.7 6.7 6.2 6.1 6.4 Is an unacceptable exploitation of animals (is making animals mere instruments for human purposes) DK DE ES FR IE IT NL AU SE UK CZ PL IL US JP From 0 to 2 20.5 7.0 7.8 8.5 11.4 6.9 10.3 7.4 10.9 13.9 17.6 10.9 14.8 18.6 5.5 From 3 to 4 13.1 11.0 10.9 8.9 11.1 12.0 17.8 11.2 9.6 10.8 15.5 11.8 11.1 10.7 11.1 5 10.5 14.2 25.2 12.6 11.4 17.3 12.5 11.2 11.1 12.8 16.3 18.1 12.7 12.4 24.4 From 6 to 7 17.0 19.0 25.0 19.0 19.4 28.2 24.3 21.0 18.6 17.0 16.2 17.2 16.9 15.5 29.5 From 8 to 10 36.6 45.6 22.2 48.1 35.8 29.1 32.8 46.7 47.0 38.6 31.1 32.3 40.5 40.0 26.1 Don’t know 2.3 3.1 9.0 2.9 11.0 6.5 2.4 2.5 2.9 6.8 3.4 9.7 3.9 2.9 3.4 Total disagree (0–4) 33.6 18.1 18.7 17.4 22.4 18.9 28.1 18.6 20.5 24.8 33.0 22.7 26.0 29.3 16.6 Total agree (6–10) 53.6 64.7 47.2 67.1 55.2 57.3 57.0 67.7 65.6 55.6 47.3 49.5 57.4 55.5 55.6 Mean 5.8 6.9 5.8 6.9 6.3 6.2 6.0 6.9 6.7 6.3 5.7 6.1 6.3 6.1 6.1 Will cause the animals a great deal of suffering DK DE ES FR IE IT NL AU SE UK CZ PL IL US JP From 0 to 2 21.9 4.1 5.1 5.9 10.0 5.1 11.3 6.7 9.0 9.9 14.3 5.2 7.7 14.3 2.8 From 3 to 4 13.3 7.5 9.0 6.3 10.9 7.9 15.6 11.2 7.4 7.7 11.2 7.1 10.3 8.4 7.6 5 15.7 12.8 20.9 14.5 10.8 14.1 17.7 8.3 13.1 14.6 15.5 11.2 10.7 13.0 21.0 From 6 to 7 14.7 21.0 25.7 16.3 15.7 32.5 21.6 19.9 18.6 17.1 15.1 22.0 18.5 17.2 32.2 From 8 to 10 25.6 48.2 22.4 47.3 33.6 30.4 26.0 48.6 42.5 38.1 30.5 42.6 44.3 39.3 33.2 Don’t know 8.7 6.4 16.8 9.6 19.0 10.0 7.8 5.3 9.5 12.6 13.4 11.8 8.5 7.8 3.3 Total disagree (0–4) 35.2 11.6 14.1 12.2 20.9 13.0 27.0 17.9 16.4 17.6 25.6 12.3 17.9 22.7 10.4 Total agree (6–10) 40.3 69.2 48.1 63.7 49.3 62.9 47.6 68.5 61.1 55.2 45.6 64.6 62.8 56.5 65.4 Mean 5.2 7.3 6.1 7.1 6.4 6.6 5.7 7.1 6.7 6.6 5.9 7.1 6.8 6.4 6.7 Should be supported DK DE ES FR IE IT NL AU SE UK CZ PL IL US JP From 0 to 2 30.5 40.3 18.1 45.5 27.9 24.8 30.7 48.3 42.3 37.0 24.7 27.2 21.1 36.2 20.3 From 3 to 4 11.6 17.8 9.6 16.3 16.5 16.5 20.0 18.8 15.3 13.9 14.4 14.3 14.1 10.5 20.0 5 15.4 15.2 23.9 18.6 12.9 16.5 19.2 11.7 15.2 16.3 17.7 18.0 14.8 16.0 38.5 From 6 to 7 17.7 13.8 21.9 9.9 13.3 23.7 18.8 9.2 11.4 13.2 17.3 15.4 17.4 13.1 12.0 From 8 to 10 20.9 8.3 13.6 4.4 10.7 7.9 8.0 6.9 10.3 12.1 18.4 13.1 22.5 18.1 4.7 Don’t know 4.0 4.6 12.8 5.3 18.8 10.6 3.2 5.1 5.6 7.5 7.5 12.1 10.1 6.2 4.3 Total disagree (0–4) 42.1 58.1 27.7 61.8 44.3 41.3 50.7 67.1 57.6 50.8 39.2 41.4 35.2 46.7 40.3 Total agree (6–10) 38.5 22.1 35.5 14.3 23.9 31.6 26.8 16.1 21.7 25.3 35.7 28.5 39.9 31.2 16.7 Mean 4.6 3.4 4.9 3.0 4.0 4.3 4.0 2.9 3.4 3.7 4.7 4.3 5.0 4.1 4.2

DK: Denmark / DE: Germany / ES: Spain / FR: France / IR: Ireland / IT: Italy / NL: Netherlands / AU: Austria / SE: Sweden / UK: United Kingdom / CZ: Czech Republic / PL: Poland / IL: Israel / US: United States / JP: Japan

161

5.9 Tables

Table 5.6: Trust in the regulatory agencies dealing with animal pharming and disposition to

take biopharmaceuticals from genetically modified animals

“Please tell me how much you agree or disagree with the following statements, using a scale from 0 to 10, in which 0 means you totally disagree and 10 that you totally agree.” I trust the laws and controls that the (NATIONALITY) government has established to regulate the production of genetically modified animals in order to obtain pharmaceutical drugs DK DE ES FR IE IT NL AU SE UK CZ PL IL US From 0 to 2 13.2 28.4 14.6 29.5 25.7 19.9 12.6 26.2 15.5 30.4 23.0 30.5 25.0 35.9 From 3 to 4 9.1 16.6 6.7 8.7 10.8 16.3 13.5 20.3 13.9 11.5 11.2 14.8 10.8 9.8 5 9.9 17.4 20.0 17.2 13.0 17.7 12.0 15.3 13.5 14.8 16.6 17.2 13.9 14.7 From 6 to 7 19.9 16.5 27.2 14.3 16.2 29.9 34.5 16.8 20.4 15.0 18.2 12.4 19.4 19.5 From 8 to 10 43.7 17.6 17.5 21.5 16.7 7.7 24.4 16.1 32.6 16.5 22.8 9.1 24.3 14.7 Don’t know 4.2 3.7 13.9 8.7 17.6 8.5 3.0 5.4 4.3 11.9 8.2 16.0 6.5 5.5 Total disagree (0–4) 22.3 44.9 21.4 38.3 36.6 36.2 26.1 46.4 29.3 41.9 34.2 45.3 35.8 45.7 Total agree (6–10) 63.6 34.0 44.7 35.8 32.9 37.6 58.9 32.9 53.0 31.4 41.0 21.5 43.7 34.2 Mean 6.4 4.4 5.3 4.5 4.5 4.6 5.7 4.5 5.7 4.3 5.0 3.9 5.0 4.0

JP 25.1 20.3 24.1 12.0 7.1 11.3 45.4 19.1 4.0

I am quite prepared to take pharmaceutical drugs produced from genetically modified animals From 0 to 2 From 3 to 4 5 From 6 to 7 From 8 to 10 Don’t know Total disagree (0–4) Total agree (6–10) Mean

DK 29.7 7.5 7.4 13.7 38.7 3.0 37.2 52.4 5.4

DE 43.6 11.7 10.6 10.8 15.4 7.9 55.3 26.1 3.5

ES 21.9 7.1 16.7 19.6 15.5 19.2 29.0 35.1 4.6

FR 59.5 7.4 9.4 8.3 11.3 4.1 66.9 19.5 2.6

IE 39.8 9.5 10.0 11.9 12.2 16.6 49.2 24.1 3.5

IT 35.7 11.7 12.1 20.2 7.1 13.2 47.4 27.2 3.7

NL 27.9 12.7 10.1 21.5 23.6 4.2 40.6 45.1 4.8

AU 57.4 9.3 8.3 8.6 10.1 6.2 66.7 18.7 2.6

SE 27.1 9.4 10.7 13.7 33.3 5.8 36.5 47.0 5.3

UK 39.6 8.7 9.6 13.0 20.5 8.6 48.3 33.5 4.0

CZ 29.3 10.7 12.7 13.4 25.1 8.7 40.0 38.6 4.8

PL 33.9 10.4 11.3 11.8 18.6 14.1 44.3 30.4 4.2

IL 35.0 10.5 9.9 13.2 23.6 7.8 45.5 36.8 4.4

US 42.8 8.1 11.6 14.9 16.4 6.2 50.9 31.3 3.7

JP 42.5 19.2 16.9 10.3 5.8 5.5 61.7 16.1 3.2

DK: Denmark / DE: Germany / ES: Spain / FR: France / IR: Ireland / IT: Italy / NL: Netherlands / AU: Austria / SE: Sweden / UK: United Kingdom / CZ: Czech Republic / PL: Poland / IL: Israel / US: United States / JP: Japan

162

Table 5.7: Pearson’s r correlation between evaluative criteria and support for animal pharming Correlations between “support for animal pharming” and “evaluative criteria” DK Poses no serious risks Is immoral Is very useful

DE

ES

FR

IE

IT

NL

AU

SE

UK

CZ

PL

IL

US

JP

0.5

0.5

0.5

0.5

0.5

0.6

0.6

0.5

0.5

0.4

0.7

0.5

0.4

0.5

0.3

-0.5

-0.6

-0.5

-0.6

-0.4

-0.5

-0.6

-0.5

-0.5

-0.6

-0.7

-0.5

-0.4

-0.5

-0.4

0.5

0.7

0.6

0.7

0.6

0.7

0.7

0.7

0.5

0.7

0.8

0.7

0.6

0.7

0.6

-0.6

-0.4

-0.5

-0.5

-0.4

-0.5

-0.6

-0.5

-0.5

-0.7

-0.7

-0.6

-0.4

-0.5

-0.5

Goes against nature

-0.5

-0.5

-0.5

-0.5

-0.3

-0.5

-0.5

-0.6

-0.4

-0.6

-0.6

-0.5

-0.3

-0.5

-0.5

Is incompatible with the inherent dignity of animals

-0.6

-0.6

-0.4

-0.6

-0.4

-0.4

-0.6

-0.6

-0.5

-0.6

-0.6

-0.5

-0.3

-0.5

-0.4

Is a product of the arrogance of scientists

-0.4

-0.5

-0.5

-0.4

-0.4

-0.4

-0.5

-0.5

-0.4

-0.6

-0.6

-0.5

-0.3

-0.4

-0.5

Is an unacceptable exploitation of animals (is making animals mere instruments for human purposes)

-0.6

-0.7

-0.4

-0.6

-0.5

-0.5

-0.7

-0.6

-0.6

-0.7

-0.7

-0.6

-0.4

-0.5

-0.5

Will cause the animals a great deal of suffering

-0.5

-0.6

-0.4

-0.5

-0.5

-0.4

-0.6

-0.6

-0.5

-0.6

-0.6

-0.5

-0.3

-0.5

-0.4

Will transmit diseases from animals to humans

-0.4

-0.5

-0.4

-0.5

-0.4

-0.3

-0.4

-0.5

-0.4

-0.6

-0.5

-0.5

-0.3

-0.5

-0.3

Is a product of the interests of a small number of pharmaceutical multinationals

-0.4

-0.4

-0.3

-0.3

-0.3

-0.3

-0.4

-0.3

-0.4

-0.4

-0.4

-0.2

-0.2

-0.3

-0.3

Is playing God

-0.5

-0.5

-0.4

-0.3

-0.4

-0.4

-0.4

-0.5

-0.5

-0.6

-0.5

-0.5

-0.3

-0.5

-0.3

DK: Denmark / DE: Germany / ES: Spain / FR: France / IR: Ireland / IT: Italy / NL: Netherlands / AU: Austria / SE: Sweden / UK: United Kingdom / CZ: Czech Republic / PL: Poland / IL: Israel / US: United States / JP: Japan

5 Public views and attitudes to pharming

Is reckless

163

5.9 Tables Table 5.8: Acceptance of plant pharming for different biomedical purposes

“To what extent do you think it is acceptable or not to genetically modify plants to obtain drugs and treatments for the following purposes? Please use a scale from 0 to 10, where 0 means you find it totally unacceptable, and 10 means you find it totally acceptable. You may, of course, give any number between 0 and 10.”

To treat lifethreatening diseases

DK DE ES FR IE IT NL AU SE UK CZ PL IL US JP

7.8 6.0 7.4 6.4 6.8 6.1 7.5 4.8 7.4 7.1 7.5 7.2 7.9 7.4 6.4

To treat diseases in children

7.1 5.0 7.2 6.0 6.6 5.9 7.1 4.0 7.0 7.0 7.4 6.8 7.7 7.2 5.7

For vaccinating adults beTo obfore they tain anti- travel dotes or to armedicines eas where to counthere is ter the ef- a risk fects of conof biotracting logical certain weapons diseases

6.9 5.1 6.6 5.5 5.9 5.3 6.5 4.2 6.1 6.0 7.2 7.0 7.7 7.0 5.6

6.6 4.5 6.6 5.5 5.9 5.1 6.4 3.8 5.7 6.4 6.8 6.5 7.1 6.9 6.1

To obtain food products with beneficial properties for health

5.9 3.5 5.7 3.8 5.0 4.5 5.9 3.3 4.4 5.5 5.5 5.4 6.0 6.4 5.3

To treat minor ailments

5.4 3.4 5.9 3.8 4.9 4.5 5.7 3.3 4.7 5.3 5.6 5.4 6.5 6.2 4.8

To treat non serious diseases

5.2 3.9 6.4 4.1 5.4 4.5 5.9 3.3 4.4 5.4 5.9 5.7 5.7 6.4 4.6

To help people live longer

4.1 3.4 5.4 3.6 4.8 4.6 4.6 3.2 3.4 4.4 5.8 6.0 6.5 6.1 4.2

To obtain To delay the effects cosmetic of ageing products

3.5 3.3 5.4 3.4 4.2 4.2 4.7 3.0 3.4 4.2 5.5 5.6 6.3 5.8 4.5

3.2 2.5 4.7 2.6 3.3 3.4 3.9 2.6 2.3 3.3 4.3 4.9 4.9 4.9 4.3

DK: Denmark / DE: Germany / ES: Spain / FR: France / IR: Ireland / IT: Italy / NL: Netherlands / AU: Austria / SE: Sweden / UK: United Kingdom / CZ: Czech Republic / PL: Poland / IL: Israel / US: United States / JP: Japan

164

5 Public views and attitudes to pharming

Table 5.9: Acceptance of plant pharming for different socio-economic purposes “To what extent do you think it is acceptable or not to genetically modify plants for each of the purposes I am going to read out to you? Please use a scale from 0 to 10, where 0 means you find it totally unacceptable, and 10 means you find it totally acceptable. You can, of course, give any score you like between 0 and 10.”

Country Denmark Germany Spain France Ireland Italy Netherlands Austria Sweden Great Britain Czech Republic Poland Israel United States Japan

To obtain cheaper drugs for people in less developed countries 6.8 4.4 6.9 5.5 6.1 4.9 6.1 3.8 5.8 6.0 6.4 6.0 6.8 6.5 5.6

To overcome problems due to shortages of certain kinds of drugs where current production methods cannot cope with the scale of demand 6.5 4.4 6.7 5.6 5.9 4.9 6.1 3.7 5.6 5.9 6.2 6.0 7.0 6.8 5.8

To obtain cheaper drugs for people in advanced countries 5.3 3.6 5.9 4.4 4.9 4.5 5.4 3.1 4.0 4.9 6.1 5.4 6.1 6.1 4.8

5.9 Tables

165

Table 5.10: Acceptability of growing plants for pharming in open fields and in closed precincts “Thinking about the growing of genetically modified plants to obtain pharmaceutical drugs, I would like you to tell me to what extent you think the following options are acceptable or not. Please use a scale from 0 to 10, where 0 means it is totally unacceptable and 10 that it is totally acceptable.” The growing of genetically modified plants to obtain pharmaceutical drugs in open fields DK DE ES FR IE IT NL AU SE UK CZ PL IL US JP From 0 to 2 54.8 47.5 29.5 53.5 39.6 28.1 28.1 56.8 44.4 39.2 30.8 38.3 38.5 35.0 30.3 From 3 to 4 14.8 18.7 13.4 13.5 13.5 18.7 23.2 15.2 13.9 13.7 16.1 17.5 13.5 11.7 23.8 5 7.8 11.7 19.5 10.8 9.8 15.2 12.3 9.6 10.5 13.2 15.2 12.5 10.5 12.9 21.2 From 6 to 7 8.2 9.3 15.4 9.1 12.3 21.1 19.8 8.5 11.8 12.1 14.1 10.6 14.6 14.9 13.4 From 8 to 10 11.7 8.7 10.5 8.7 8.5 8.4 12.8 6.3 13.3 13.5 19.4 12.3 16.5 20.3 5.7 Don’t know 2.8 4.2 11.8 4.4 16.3 8.3 3.8 3.6 6.1 8.3 4.4 8.9 6.4 5.2 5.5 Total unacceptable (0–4) 69.6 66.2 42.9 67.1 53.1 46.9 51.3 72.0 58.3 52.9 46.9 55.7 52.0 46.7 54.1 Total acceptable (6–10) 19.8 17.9 25.8 17.7 20.8 29.6 32.6 14.8 25.0 25.6 33.5 22.9 31.0 35.2 19.1 Mean 3.0 3.0 4.0 2.8 3.3 4.1 4.3 2.5 3.5 3.7 4.4 3.6 3.9 4.3 3.7 The growing of genetically modified plants to obtain pharmaceutical drugs in open fields at a distance of several kilometres from other areas of crops or plants that have not been genetically modified DK DE ES FR IE IT NL AU SE UK CZ PL IL US JP From 0 to 2 43.6 37.0 23.0 47.7 29.8 23.5 30.2 50.8 35.5 31.3 22.8 31.2 29.0 31.0 29.1 From 3 to 4 15.6 15.5 12.7 13.5 11.9 17.0 23.2 17.1 12.3 13.8 14.3 15.1 14.5 11.5 21.9 5 7.8 14.2 21.7 10.7 11.3 12.5 13.1 10.0 12.0 13.6 15.4 15.6 12.3 13.8 23.1 From 6 to 7 12.3 14.5 18.8 11.9 17.7 25.8 17.4 9.5 15.3 16.4 18.5 13.9 15.4 16.3 14.6 From 8 to 10 18.4 14.3 11.6 11.7 14.3 13.1 11.2 9.1 18.5 15.6 24.5 15.5 22.5 22.8 5.4 Don’t know 2.3 4.5 12.1 4.6 15.0 8.1 4.9 3.5 6.4 9.2 4.6 8.7 6.3 4.7 5.9 Total unacceptable (0–4) 59.2 52.5 35.7 61.1 41.7 40.5 53.4 67.9 47.8 45.1 37.1 46.2 43.4 42.5 51.0 Total acceptable (6–10) 30.7 28.8 30.5 23.6 32.0 38.9 28.6 18.6 33.8 32.0 43.0 29.5 38.0 39.1 20.0 Mean 3.8 3.8 4.4 3.2 4.2 4.6 4.1 2.9 4.2 4.3 5.1 4.2 4.6 4.6 3.8 The growing of genetically modified plants to obtain drugs in closed precincts (for example greenhouses) DK DE ES FR IE IT NL AU SE UK CZ PL IL US JP From 0 to 2 9.9 13.6 7.1 13.4 12.1 10.2 8.7 26.2 15.1 12.1 6.8 7.5 6.9 10.7 11.2 From 3 to 4 5.7 7.8 3.8 6.4 8.3 9.4 9.1 15.6 8.0 7.5 4.8 5.2 5.9 4.6 13.1 5 6.7 12.8 15.1 9.5 11.5 13.2 9.6 10.8 9.7 12.1 9.2 12.1 4.9 9.4 24.7 From 6 to 7 17.9 21.6 27.3 18.3 18.7 29.9 34.9 16.2 16.1 22.4 17.3 21.0 15.9 16.0 27.6 From 8 to 10 57.8 39.8 35.7 48.8 34.9 31.3 34.9 28.2 46.4 38.0 59.0 48.3 60.8 55.8 17.6 Don’t know 2.1 4.4 11.0 3.6 14.6 6.0 2.8 3.1 4.7 7.9 3.0 5.9 5.7 3.6 5.8 Total unacceptable (0–4) 15.5 21.3 10.9 19.8 20.3 19.6 17.8 41.7 23.1 19.6 11.5 12.7 12.7 15.3 24.3 Total acceptable (6–10) 75.7 61.4 63.0 67.1 53.6 61.2 69.9 44.4 62.5 60.4 76.3 69.3 76.7 71.8 45.2 Mean 7.2 6.3 6.6 6.6 6.3 6.1 6.4 5.0 6.5 6.4 7.5 7.1 7.6 7.2 5.5

DK: Denmark / DE: Germany / ES: Spain / FR: France / IR: Ireland / IT: Italy / NL: Netherlands / AU: Austria / SE: Sweden / UK: United Kingdom / CZ: Czech Republic / PL: Poland / IL: Israel / US: United States / JP: Japan

166

5 Public views and attitudes to pharming

Table 5.11: Perceived environmental and food risks associated with plant pharming “With the help of this scale (SHOW CARD) where 0 means a very low level of risk and 10 a very high one, please tell me: READ 1. What level of risk do you think there is that the growing of genetically modified plants to obtain drugs could pollute the environment? 2. And what level of risk do you think there is that the growing of genetically modified plants to obtain drugs could contaminate the non-genetically modified plants or seeds that we currently consume as food?” Pollute the environment DK DE ES FR IE IT NL AU SE UK CZ PL IL From 0 to 2 23.7 15.8 14.0 12.6 12.9 12.0 20.9 14.6 14.3 20.7 31.5 15.6 23.7 From 3 to 4 15.3 16.6 12.1 9.9 15.9 15.8 24.5 15.5 12.3 13.1 17.0 16.1 14.3 5 14.9 15.7 23.1 18.1 13.9 19.1 14.3 13.0 18.9 16.5 16.7 18.8 15.5 From 6 to 7 19.3 17.9 18.5 20.4 16.7 27.5 23.4 19.8 20.2 19.2 13.8 17.5 17.5 From 8 to 10 22.1 26.8 10.9 30.4 19.6 13.5 10.4 27.8 22.1 18.3 13.5 17.0 16.9 Don’t know 4.7 7.2 21.5 8.5 21.1 12.1 6.5 9.3 12.3 12.3 7.4 14.9 12.1 Total no risk (0–4) 39.0 32.4 26.0 22.5 28.8 27.8 45.4 30.2 26.6 33.8 48.6 31.7 38.0 Total risk (6–10) 41.4 44.7 29.4 50.8 36.3 41.0 33.8 47.6 42.3 37.5 27.3 34.5 34.4 Mean 5.0 5.5 4.9 6.0 5.4 5.3 4.5 5.7 5.5 5.0 4.1 5.1 4.7 Contaminate the non-genetically modified plants or seeds that we currently consume as food DK DE ES FR IE IT NL AU SE UK CZ PL IL From 0 to 2 16.0 9.3 11.5 5.1 8.4 9.0 13.8 12.7 6.9 10.4 17.7 10.8 18.1 From 3 to 4 11.3 13.6 11.4 5.2 10.7 14.3 21.7 12.9 8.1 10.4 14.6 13.7 14.0 5 13.9 17.3 22.5 13.6 14.2 16.4 17.6 11.3 18.1 15.5 18.8 17.1 17.1 From 6 to 7 21.9 19.2 19.8 20.1 19.3 32.2 25.0 21.4 20.8 21.9 17.1 20.8 17.4 From 8 to 10 30.2 35.1 13.0 48.3 27.3 17.7 14.7 35.0 27.0 29.5 21.8 24.5 21.0 Don’t know 6.7 5.5 21.7 7.7 20.2 10.4 7.1 6.6 19.1 12.2 10.0 13.0 12.6 Total no risk (0–4) 27.3 22.9 23.0 10.3 19.1 23.3 35.6 25.7 15.0 20.8 32.3 24.5 32.0 Total risk (6–10) 52.1 54.4 32.8 68.4 46.6 49.8 39.7 56.4 47.8 51.5 38.9 45.4 38.3 Mean 5.8 6.3 5.2 7.2 6.2 5.7 5.1 6.2 6.3 6.1 5.3 5.8 5.2

US 21.8 14.0 18.5 18.7 18.4 8.7 35.8 37.1 4.9

JP 7.0 14.4 24.4 25.2 19.7 9.3 21.4 44.9 5.8

US 14.9 12.2 19.0 19.4 26.4 8.1 27.1 45.8 5.6

JP 4.8 9.5 19.3 28.4 29.6 8.4 14.3 58.0 6.4

Table 5.12: Acceptance of plant pharming by type of plant “Thinking about the genetic modification of plants to obtain and produce pharmaceutical drugs, do you think it is acceptable or not to use each of the following plants for this purpose: READ ITEMS” DK DE ES FR IE Leaves of genetically modified tobacco plants Yes 65.9 47.3 51.5 42.1 39.3 No 26.9 39.2 27.7 47.3 30.8 Don’t know 7.2 13.5 20.8 10.5 29.9 Genetically modified potatoes Yes 63.8 37.4 50.8 39.9 36.4 No 30.5 54.3 28.4 51.0 35.1 Don’t know 5.7 8.3 20.8 9.1 28.5 Genetically modified corn Yes 64.4 39.1 50.9 40.8 36.3 No 30.2 52.9 28.4 50.1 34.5 Don’t know 5.4 8.0 20.7 9.1 29.3

IT

NL

AU

SE

UK

CZ

PL

IL

US

JP

38.0 69.5 35.2 54.5 49.7 71.4 47.5 60.9 56.0 32.0 41.5 23.9 49.2 35.5 36.1 19.4 33.5 24.2 34.1 55.6 20.5 6.6 15.6 10.0 14.2 9.3 19.1 14.9 9.9 12.4 34.8 66.7 28.0 54.6 53.2 67.9 47.0 59.6 58.4 26.1 48.1 27.6 60.5 36.9 34.9 25.6 40.0 25.3 32.4 63.6 17.1 5.7 11.5 8.5 11.8 6.5 13.0 15.1 9.3 10.4 34.5 67.2 27.3 54.5 53.5 69.4 49.1 59.0 58.8 27.0 49.4 27.8 62.0 37.2 34.3 24.1 37.6 24.5 31.8 62.5 16.1 5.0 10.6 8.3 12.2 6.5 13.3 16.4 9.4 10.4

DK: Denmark / DE: Germany / ES: Spain / FR: France / IR: Ireland / IT: Italy / NL: Netherlands / AU: Austria / SE: Sweden / UK: United Kingdom / CZ: Czech Republic / PL: Poland / IL: Israel / US: United States / JP: Japan

167

5.9 Tables Table 5.13: Acceptance of animal pharming for different biomedical purposes

“And to what extent do you think it is acceptable or not to genetically modify animals to obtain drugs and treatments for the following purposes? Please use a scale from 0 to 10, where 0 means you find it totally unacceptable, and 10 means you find it totally acceptable. You may, of course, give any number between 0 and 10.”

To treat lifethreatening diseases

DK DE ES FR IE IT NL AU SE UK CZ PL IL US JP

6.2 5.0 6.6 4.5 5.1 5.1 5.6 3.7 5.3 5.2 6.2 6.0 6.5 5.1 5.6

To treat diseases in children

5.6 4.2 6.3 4.3 5.0 5.1 5.1 3.1 5.1 5.2 6.3 5.6 6.4 5.0 5.1

For vaccinating adults before they To obtravel tain anti- to ardotes or eas where medicines there is to couna risk ter the ef- of confects of tracting biological certain weapons diseases

5.2 4.0 5.7 3.6 4.3 4.3 4.5 3.0 4.1 4.2 5.9 5.7 6.3 4.9 5.0

4.7 3.4 5.5 3.6 4.1 4.3 4.0 2.7 3.4 4.2 5.3 5.2 5.4 4.7 5.3

To obtain food products with beneficial properties for health

3.9 3.0 5.4 2.6 3.8 3.7 3.7 2.5 2.7 3.6 4.4 4.4 4.1 4.4 3.9

To treat minor ailments

3.6 2.3 4.1 1.8 3.1 3.7 3.4 2.3 2.2 3.2 3.7 4.1 4.1 3.6 4.4

To treat non- serious diseases

3.5 2.4 4.3 1.8 3.1 3.7 3.3 2.4 2.0 3.1 4.0 4.0 4.5 3.6 4.1

To help people live longer

2.8 2.7 4.2 2.1 3.4 3.9 2.8 2.4 2.1 2.7 4.4 4.6 5.0 3.6 3.5

To obtain To delay the effects cosmetic of ageing products

2.3 2.6 4.1 1.9 2.8 3.5 2.8 2.3 2.1 2.5 4.2 4.2 4.8 3.4 3.7

1.8 1.7 3.1 1.3 2.2 2.6 1.7 1.7 1.0 1.5 2.7 3.3 3.1 2.3 3.6

DK: Denmark / DE: Germany / ES: Spain / FR: France / IR: Ireland / IT: Italy / NL: Netherlands / AU: Austria / SE: Sweden / UK: United Kingdom / CZ: Czech Republic / PL: Poland / IL: Israel / US: United States / JP: Japan

168

5 Public views and attitudes to pharming

Table 5.14: Acceptance of animal pharming for different socio-economic purposes “And to what extent do you think it is acceptable or not to genetically modify animals for each of these purposes? Please use the same scale from 0 to 10.”

Country Denmark Germany Spain France Ireland Italy Netherlands Austria Sweden Great Britain Czech Republic Poland Israel United States Japan

To obtain cheaper drugs for people in less developed countries 5.1 3.3 6.0 3.5 4.4 4.4 4.3 3.0 3.6 4.3 4.8 4.8 4.9 4.5 4.9

To overcome problems due to shortages of certain kinds of drugs where current production methods cannot cope with the scale of demand 4.8 3.3 5.8 3.5 4.2 4.3 4.0 2.7 3.5 4.1 4.6 4.8 5.1 4.5 5.1

To obtain cheaper drugs for people in advanced countries 3.7 2.7 4.9 2.5 3.3 3.9 3.6 2.4 2.2 3.4 4.6 4.4 4.3 4.2 4.2

Table 5.15: Acceptability of different types of animals for pharming “And to what extent do you think it is acceptable or not to genetically modify the following animals in order to obtain drugs? Please use a scale from 0 to 10, where 0 means you find it totally unacceptable, and 10 means you find it totally acceptable.” Fruit Country flies Mice Fish Hens Rabbits Sheep Cows Pigs Chimpanzees Denmark 6.0 5.2 5.2 4.8 4.7 4.5 4.4 4.5 3.8 Germany 4.6 3.7 3.5 3.0 2.8 2.7 2.6 2.6 2.2 Spain 5.8 6.1 5.2 5.1 5.2 4.9 4.8 4.9 4.6 France 4.3 3.9 3.3 2.9 2.9 2.7 2.6 2.8 2.4 Ireland 4.0 3.9 3.5 3.1 3.3 2.9 2.7 2.8 2.8 Italy 4.4 4.2 3.8 3.6 3.6 3.4 3.3 3.4 3.3 Netherlands 5.2 4.3 4.2 3.9 3.9 3.7 3.7 3.7 3.2 Austria 3.7 2.9 2.7 2.4 2.1 2.1 1.9 2.0 1.9 Sweden 5.2 3.8 3.7 3.1 3.1 2.7 2.6 2.8 2.2 Great Britain 4.3 3.8 3.7 3.4 3.4 3.3 3.2 3.3 2.7 Czech Republic 6.4 5.8 5.0 4.8 4.5 4.3 4.0 4.2 3.0 Poland 5.3 5.2 4.6 4.0 4.3 4.0 3.6 3.7 3.6 Israel 5.9 5.4 5.0 4.5 4.5 4.3 4.1 4.5 3.8 United States 5.3 5.0 4.4 3.9 4.0 3.7 3.6 3.7 3.3 Japan 5.7 5.0 4.9 4.4 4.2 4.1 4.1 4.1 3.6

169

5.9 Tables

Table 5.16: Segments of acceptability of different types of animals for pharming purposes “And to what extent do you think it is acceptable or not to genetically modify the following animals in order to obtain drugs? Please use a scale from 0 to 10, where 0 means you find it totally unacceptable, and 10 means you find it totally acceptable“. Fruit flies From 0 to 2 From 3 to 4 5 From 6 to 7 From 8 to 10 Don’t know Unacceptable (0–4) Acceptable (6–10) Mice From 0 to 2 From 3 to 4 5 From 6 to 7 From 8 to 10 Don’t know Unacceptable (0–4) Acceptable (6–10) Fish From 0 to 2 From 3 to 4 5 From 6 to 7 From 8 to 10 Don’t know Unacceptable (0–4) Acceptable (6–10) Hens From 0 to 2 From 3 to 4 5 From 6 to 7 From 8 to 10 Don’t know Unacceptable (0–4) Acceptable (6–10)

DK 26.4 6.8 6.3 9.7 48.4 2.3 33.2 58.1

DE 34.9 8.9 9.9 10.7 30.0 5.6 43.8 40.7

ES 19.2 7.9 13.2 17.3 35.8 6.6 27.1 53.1

FR 39.2 6.9 11.4 8.9 26.5 7.1 46.1 35.4

IE 40.2 7.8 8.3 8.7 21.4 13.6 48.0 30.0

IT 31.7 11.8 12.0 19.9 18.6 6.0 43.5 38.5

NL 30.4 9.6 7.7 17.4 33.5 1.4 40.0 50.9

AU 42.1 12.6 11.0 10.5 18.0 5.8 54.7 28.5

SE 36.3 6.7 7.6 6.5 40.2 2.7 43.1 46.7

UK 37.6 7.5 9.4 10.7 24.9 9.9 45.1 35.6

CZ 20.2 6.1 9.7 10.3 48.5 5.3 26.3 58.8

PL 28.4 6.8 11.9 10.9 34.8 7.2 35.2 45.7

IL 27.0 6.3 6.7 10.3 44.9 4.8 33.2 55.2

US 34.8 4.8 8.4 9.1 39.9 2.9 39.6 49.0

JP 18.8 11.1 17.5 16.5 33.1 3.0 29.9 49.6

DK 31.7 8.2 7.3 13.4 37.8 1.6 39.9 51.2

DE 44.3 9.8 9.9 13.8 19.2 2.9 54.1 33.0

ES 11.8 8.0 17.3 24.4 33.8 4.6 19.9 58.2

FR 42.8 8.7 13.1 11.3 22.5 1.7 51.5 33.8

IE 41.4 8.9 9.7 10.2 20.6 9.3 50.3 30.8

IT 34.1 12.2 11.7 22.4 14.5 5.1 46.3 36.9

NL 37.1 11.5 9.3 19.9 20.8 1.5 48.6 40.7

AU 53.4 12.7 10.2 8.4 12.5 2.7 66.1 20.9

SE 48.4 8.7 6.9 8.3 25.8 2.0 57.1 34.0

UK 44.7 8.8 10.1 12.0 20.0 4.4 53.5 32.0

CZ 24.2 8.2 10.5 13.7 41.1 2.3 32.3 54.8

PL 29.0 8.4 12.6 15.3 31.4 3.3 37.4 46.7

IL 28.7 8.0 9.6 12.2 38.4 3.1 36.7 50.6

US 35.2 6.3 8.7 13.7 35.0 1.1 41.5 48.7

JP 22.0 14.9 19.5 19.7 21.7 2.2 36.9 41.4

DK 32.1 8.5 6.9 12.7 37.9 1.8 40.6 50.6

DE 47.0 10.2 11.3 12.4 16.0 3.1 57.3 28.4

ES 19.1 10.2 21.4 22.5 21.7 5.2 29.3 44.2

FR 49.4 8.6 13.3 10.9 15.5 2.3 58.0 26.4

IE 45.1 9.9 9.3 9.8 15.7 10.2 55.0 25.5

IT 37.1 13.4 13.6 19.4 11.4 5.1 50.6 30.8

NL 36.7 13.3 10.2 19.9 18.7 1.3 50.0 38.6

AU 56.5 12.9 10.2 6.3 10.7 3.5 69.3 17.0

SE 48.7 9.1 8.0 8.6 23.4 2.2 57.8 32.0

UK 45.2 9.1 10.1 11.1 19.6 5.0 54.3 30.7

CZ 29.7 9.9 12.9 13.7 31.0 2.9 39.6 44.6

PL 31.6 10.5 15.2 15.0 23.0 4.6 42.2 38.0

IL 30.5 10.5 9.9 14.0 31.9 3.1 41.1 45.9

US 40.6 7.3 9.7 12.5 28.1 1.8 47.9 40.6

JP 22.3 15.9 21.1 19.1 19.2 2.4 38.2 38.3

DK 36.7 8.7 7.7 12.8 32.4 1.7 45.4 45.2

DE 52.4 11.3 10.2 10.6 12.6 2.7 63.7 23.3

ES 19.4 10.8 21.9 20.7 21.5 5.7 30.2 42.3

FR 54.9 9.2 12.7 7.8 13.1 2.3 64.1 20.9

IE 49.0 9.8 8.3 9.7 12.8 10.5 58.8 22.5

IT 40.0 14.8 12.1 18.8 9.2 5.1 54.8 28.0

NL 39.5 15.0 9.4 20.1 14.7 1.3 54.5 34.8

AU 62.1 11.9 8.6 6.4 8.7 2.3 74.0 15.1

SE 54.8 9.6 8.0 8.5 16.9 2.2 64.4 25.4

UK 47.2 10.6 10.9 10.5 16.0 4.8 57.8 26.6

CZ 32.2 11.5 11.6 12.6 29.5 2.5 43.7 42.1

PL 38.5 10.8 14.9 13.3 18.3 4.3 49.3 31.5

IL 37.9 9.3 8.3 11.9 28.7 3.9 47.2 40.5

US 44.0 9.1 9.7 13.4 22.1 1.7 53.1 35.5

JP 26.6 18.3 21.9 17.7 13.0 2.6 44.9 30.7

DK: Denmark / DE: Germany / ES: Spain / FR: France / IR: Ireland / IT: Italy / NL: Netherlands / AU: Austria / SE: Sweden / UK: United Kingdom / CZ: Czech Republic / PL: Poland / IL: Israel / US: United States / JP: Japan Continued on next page

170

5 Public views and attitudes to pharming

Rabbits From 0 to 2 From 3 to 4 5 From 6 to 7 From 8 to 10 Don’t know Unacceptable (0–4) Acceptable (6–10) Sheep From 0 to 2 From 3 to 4 5 From 6 to 7 From 8 to 10 Don’t know Unacceptable (0–4) Acceptable (6–10) Cows From 0 to 2 From 3 to 4 5 From 6 to 7 From 8 to 10 Don’t know Unacceptable (0–4) Acceptable (6–10) Pigs From 0 to 2 From 3 to 4 5 From 6 to 7 From 8 to 10 Don’t know Unacceptable (0–4) Acceptable (6–10) Chimpanzees From 0 to 2 From 3 to 4 5 From 6 to 7 From 8 to 10 Don’t know Unacceptable (0–4) Acceptable (6–10)

DK 36.1 10.1 9.6 13.2 29.0 2.0 46.2 42.2

DE 53.7 12.4 10.5 8.7 10.8 3.9 66.1 19.5

ES 16.8 10.2 23.8 22.7 21.4 5.1 26.9 44.1

FR 54.6 9.6 13.1 10.1 10.6 2.1 64.2 20.7

IE 46.3 9.8 10.4 10.0 13.9 9.7 56.1 23.9

IT 38.3 15.7 12.6 20.7 7.4 5.2 54.0 28.1

NL 40.3 13.5 9.7 21.3 13.6 1.6 53.8 34.8

AU 64.9 11.1 8.3 5.6 6.8 3.2 76.0 12.4

SE 55.1 9.8 7.8 8.4 16.7 2.2 64.9 25.1

UK 47.9 9.5 11.7 11.2 14.6 5.1 57.4 25.7

CZ 33.4 11.8 13.3 15.9 22.7 2.8 45.2 38.7

PL 34.6 12.1 14.7 13.9 19.5 5.1 46.7 33.5

IL 36.1 13.3 9.2 11.8 26.6 3.0 49.4 38.4

US 42.1 9.0 11.2 14.2 22.1 1.3 51.1 36.3

JP 26.4 20.3 23.2 16.7 11.1 2.3 46.7 27.8

DK 38.0 10.6 9.1 13.8 27.0 1.5 48.6 40.8

DE 55.9 13.0 9.4 9.0 9.6 3.1 68.9 18.6

ES 20.2 12.6 22.2 21.5 18.1 5.3 32.8 39.6

FR 56.6 10.6 12.6 8.1 10.1 1.9 67.2 18.2

IE 51.1 10.8 9.1 8.4 10.9 9.7 61.9 19.3

IT 42.4 14.5 13.0 17.7 7.3 5.0 56.9 25.0

NL 41.1 15.0 10.4 19.9 12.3 1.3 56.2 32.2

AU 66.2 11.6 7.7 5.3 6.8 2.4 77.8 12.1

SE 59.3 9.5 9.1 7.9 12.2 2.1 68.7 20.1

UK 49.3 9.5 11.1 11.5 14.0 4.6 58.8 25.5

CZ 36.6 12.1 12.4 15.3 21.1 2.5 48.7 36.4

PL 37.3 12.3 15.0 13.5 17.1 4.8 49.6 30.6

IL 38.3 9.9 10.4 14.5 23.3 3.5 48.2 37.8

US 46.1 9.0 11.5 13.4 18.7 1.4 55.1 32.1

JP 27.9 19.8 24.4 15.4 10.2 2.2 47.7 25.6

DK 40.0 10.4 9.0 12.8 26.2 1.5 50.4 39.0

DE 58.9 10.6 9.6 8.8 8.9 3.0 69.6 17.8

ES 22.6 12.1 20.3 20.6 19.0 5.4 34.6 39.7

FR 57.9 10.8 11.9 7.4 9.9 2.1 68.7 17.3

IE 53.4 10.6 7.7 8.4 10.1 9.8 64.1 18.4

IT 44.3 14.5 11.9 16.8 7.3 5.1 58.9 24.1

NL 42.5 15.1 9.7 18.8 12.6 1.3 57.6 31.4

AU 68.1 10.6 7.5 4.8 6.3 2.7 78.8 11.1

SE 62.4 8.9 8.2 7.3 11.2 2.1 71.2 18.5

UK 50.2 9.0 10.9 11.5 13.7 4.8 59.1 25.2

CZ 39.7 11.6 13.0 14.3 18.8 2.6 51.3 33.0

PL 42.8 12.4 14.9 11.2 14.6 4.1 55.2 25.8

IL 39.9 11.8 10.2 13.2 21.8 3.1 51.7 35.0

US 47.4 9.1 10.9 12.4 18.7 1.5 56.5 31.1

JP 29.4 19.4 23.4 15.2 10.2 2.3 48.8 25.4

DK 39.3 10.3 9.3 11.3 28.2 1.5 49.6 39.6

DE 58.4 10.5 9.1 9.2 9.8 2.9 68.9 19.0

ES 22.6 11.3 20.3 20.3 20.4 5.2 33.9 40.6

FR 56.0 10.9 12.2 7.5 11.3 2.1 66.8 18.9

IE 53.1 10.6 8.3 7.7 10.9 9.5 63.7 18.6

IT 43.0 14.3 12.6 17.7 7.1 5.4 57.2 24.8

NL 42.5 14.6 9.7 18.8 13.2 1.2 57.1 32.0

AU 66.3 11.4 8.0 5.4 6.6 2.4 77.7 12.0

SE 59.9 8.9 8.1 7.9 13.1 2.1 68.8 21.1

UK 49.5 9.8 10.2 11.5 14.3 4.6 59.4 25.8

CZ 38.2 11.0 13.1 14.2 20.7 2.8 49.2 34.9

PL 42.1 11.8 16.0 10.9 15.1 4.1 53.9 26.1

IL 38.3 9.7 8.5 11.3 28.8 3.2 48.1 40.2

US 47.0 8.1 10.9 12.5 20.0 1.6 55.1 32.5

JP 29.4 19.1 23.1 15.7 10.4 2.4 48.5 26.1

DK 46.7 11.8 8.1 10.0 21.7 1.7 58.5 31.6

DE 63.9 10.7 8.9 6.6 7.0 2.9 74.6 13.6

ES 26.0 11.7 18.6 20.0 18.8 4.9 37.7 38.8

FR 61.6 11.1 10.9 5.7 8.5 2.1 72.7 14.3

IE 54.7 9.4 7.4 7.5 11.2 9.8 64.0 18.8

IT 43.7 14.1 12.2 17.7 7.0 5.3 57.8 24.7

NL 49.1 16.0 9.0 15.3 9.4 1.3 65.0 24.7

AU 68.3 11.2 7.7 4.3 6.3 2.3 79.5 10.5

SE 67.3 8.2 6.9 5.5 9.7 2.3 75.5 15.3

UK 56.8 10.1 9.8 7.7 10.8 4.9 66.9 18.4

CZ 53.4 11.2 11.1 8.7 12.6 3.0 64.7 21.3

PL 44.3 11.6 12.5 10.8 15.6 5.2 55.9 26.4

IL 45.6 9.8 9.1 10.8 20.5 4.1 55.4 31.3

US 54.0 7.1 9.3 9.7 18.4 1.5 61.1 28.1

JP 36.5 21.3 20.5 11.0 8.4 2.2 57.8 19.4

171

5.9 Tables Table 5.17: Acceptance of pharming in different animals‘ mediums

“To what extent do you see it as acceptable or not to obtain pharmaceutical drugs from… Please use a scale from 0 to 10, where 0 means you find it totally unacceptable, and 10 means you find it totally acceptable.” Country Denmark Germany Spain France Ireland Italy Netherlands Austria Sweden Great Britain Czech Republic Poland Israel United States Japan

The eggs laid by genetically modified animals 5.0 3.7 5.0 3.5 3.6 3.9 4.8 2.9 3.8 4.0 5.6 4.6 5.6 4.3 4.2

The milk of genetically modified animals 5.0 3.7 5.0 3.4 3.6 3.9 4.8 2.9 3.9 4.0 5.5 4.5 5.4 4.2 4.2

The urine of genetically modified animals 5.1 3.3 4.0 3.0 2.8 3.4 4.8 2.4 3.8 3.7 4.8 4.2 4.1 3.5 3.7

The blood of genetically modified animals 4.9 3.3 4.0 3.0 2.9 3.4 4.5 2.4 3.5 3.6 4.9 4.2 4.1 3.6 3.6

172

5 Public views and attitudes to pharming

Table 5.18: Segments of acceptability of pharming in different animals‘ mediums “To what extent do you see it as acceptable or not to obtain pharmaceutical drugs from… Please use a scale from 0 to 10, where 0 means you find it totally unacceptable, and 10 means you find it totally acceptable.” The eggs laid by genetically modified animals DK DE ES FR IE From 0 to 2 33.0 43.2 19.3 43.6 40.3 From 3 to 4 9.3 12.0 9.8 12.5 10.7 5 8.1 12.4 20.3 12.3 11.1 From 6 to 7 13.2 10.6 20.5 13.3 13.7 From 8 to 10 34.6 17.1 18.3 13.2 11.4 Don’t know 1.9 4.6 11.8 5.1 12.8 Unacceptable (0–4) 42.3 55.2 29.1 56.1 51.0 Acceptable (6–10) 47.8 27.7 38.8 26.5 25.1 The milk of genetically modified animals DK DE ES FR IE From 0 to 2 33.2 43.0 17.8 44.4 38.6 From 3 to 4 10.1 12.2 10.1 12.1 11.7 5 8.3 12.0 21.4 12.9 10.2 From 6 to 7 13.6 11.1 19.0 13.5 14.6 From 8 to 10 32.9 17.0 19.5 11.9 11.2 Don’t know 2.1 4.7 12.2 5.1 13.6 Unacceptable (0–4) 43.2 55.2 27.9 56.5 50.3 Acceptable (6–10) 46.5 28.1 38.5 25.5 25.8 The urine of genetically modified animals DK DE ES FR IE From 0 to 2 32.6 47.2 31.8 51.3 49.8 From 3 to 4 9.1 12.1 13.4 10.3 10.7 5 7.5 11.1 15.3 10.3 7.2 From 6 to 7 12.3 9.3 14.8 10.5 9.9 From 8 to 10 35.7 14.9 13.1 11.0 9.1 Don’t know 2.8 5.4 11.6 6.6 13.3 Unacceptable (0–4) 41.7 59.3 45.1 61.5 60.5 Acceptable (6–10) 48.0 24.2 27.9 21.6 19.0 The blood of genetically modified animals DK DE ES FR IE From 0 to 2 33.8 47.4 31.1 51.1 48.5 From 3 to 4 9.7 11.5 13.4 10.4 10.9 5 7.4 11.4 14.8 11.2 8.0 From 6 to 7 14.4 10.5 15.2 11.4 9.8 From 8 to 10 32.4 14.3 13.6 10.3 9.5 Don’t know 2.3 4.9 11.9 5.5 13.3 Unacceptable (0–4) 43.5 58.9 44.5 61.5 59.4 Acceptable (6–10) 46.8 24.8 28.8 21.7 19.4

IT 34.9 14.3 13.6 20.3 9.1 7.8 49.2 29.4

NL 27.0 12.5 10.8 25.6 20.6 3.4 39.6 46.3

AU 52.0 13.2 11.0 9.6 10.3 3.9 65.2 19.9

SE 44.9 10.7 8.7 9.9 21.7 4.1 55.6 31.6

UK 38.5 10.9 10.8 12.9 20.7 6.3 49.4 33.5

CZ 22.2 10.0 12.2 16.5 36.2 2.9 32.3 52.7

PL 29.1 11.3 15.5 15.4 19.9 8.9 40.3 35.3

IL 23.4 11.1 9.4 13.8 35.5 6.7 34.6 49.3

US 37.3 9.5 12.6 15.1 22.5 3.0 46.8 37.6

JP 26.6 20.3 20.2 18.7 9.3 4.9 46.9 28.0

IT 34.8 14.6 13.8 19.2 9.3 8.3 49.4 28.5

NL 26.8 12.8 12.0 24.1 20.7 3.6 39.7 44.8

AU 53.2 11.9 9.5 10.4 9.9 5.0 65.2 20.3

SE 44.7 10.7 8.9 9.7 21.8 4.2 55.4 31.6

UK 38.6 10.2 11.2 12.6 20.4 7.0 48.7 33.0

CZ 22.6 10.8 12.5 16.3 34.1 3.8 33.4 50.4

PL 28.6 12.8 15.1 14.5 19.5 9.6 41.4 33.9

IL 24.1 11.1 9.2 16.0 32.8 6.7 35.3 48.8

US 38.4 8.9 12.9 14.1 22.7 3.0 47.3 36.8

JP 25.8 21.1 20.2 18.6 9.5 4.8 46.9 28.1

IT 42.0 13.7 11.2 17.1 7.8 8.2 55.7 24.8

NL 28.2 12.1 9.4 25.4 22.1 2.8 40.3 47.5

AU 59.8 11.9 9.0 8.0 7.0 4.3 71.7 15.0

SE 46.9 9.5 7.2 8.7 22.7 5.1 56.4 31.4

UK 43.4 9.7 9.6 10.6 19.2 7.6 53.0 29.8

CZ 32.2 9.8 11.5 12.0 29.3 5.2 42.1 41.3

PL 32.9 11.8 13.6 11.9 17.9 11.9 44.7 29.8

IL 38.2 12.1 8.9 10.1 21.1 9.4 50.4 31.3

US 47.4 9.2 10.7 10.8 18.0 3.9 56.6 28.8

JP 33.2 21.9 19.1 13.1 7.7 5.1 55.1 20.8

IT 41.1 15.3 11.6 17.6 6.9 7.5 56.4 24.5

NL 30.7 14.2 11.1 23.5 17.7 2.8 44.9 41.2

AU 59.8 12.9 9.1 7.3 6.9 3.8 72.7 14.3

SE 49.0 10.5 8.1 8.5 19.3 4.5 59.6 27.8

UK 44.8 8.9 10.4 10.9 18.1 6.9 53.7 29.0

CZ 29.4 11.5 12.1 13.5 29.5 4.0 41.0 43.0

PL 32.3 12.7 13.3 12.9 18.0 10.8 45.0 30.9

IL 38.9 11.0 9.1 10.1 22.6 8.3 49.9 32.7

US 45.9 8.5 10.9 12.1 18.8 3.8 54.4 30.9

JP 35.2 21.7 18.6 13.0 6.6 4.9 56.9 19.6

DK: Denmark / DE: Germany / ES: Spain / FR: France / IR: Ireland / IT: Italy / NL: Netherlands / AU: Austria / SE: Sweden / UK: United Kingdom / CZ: Czech Republic / PL: Poland / IL: Israel / US: United States / JP: Japan

173

5.9 Tables Table 5.19: Preferred method for producing a medicine with identical composition

“Please imagine that the same medicine, that is one with an identical composition, can be obtained by different production means or methods. The only difference is the way it is produced. Of the following ways of producing drugs, which would you rather was used to produce a particular drug, assuming its composition was identical in every case? Please list the methods starting with the one that you would most prefer. And in second place? And in third place?” 1st place Isolating the drug from species of wild plants Obtaining the drug through chemical synthesis Obtaining the drug through the genetic modification of animal cells in culture Obtaining the drug through the genetic modification of plants Obtaining the drug through the genetic modification of animals Don’t know 2nd place Isolating the drug from species of wild plants Obtaining the drug through chemical synthesis Obtaining the drug through the genetic modification of animal cells in culture Obtaining the drug through the genetic modification of plants Obtaining the drug through the genetic modification of animals Don’t know 3rd place Isolating the drug from species of wild plants Obtaining the drug through chemical synthesis Obtaining the drug through the genetic modification of animal cells in culture Obtaining the drug through the genetic modification of plants Obtaining the drug through the genetic modification of animals Don’t know Total (1st+ 2nd+ 3rd) Isolating the drug from species of wild plants Obtaining the drug through chemical synthesis Obtaining the drug through the genetic modification of plants Obtaining the drug through the genetic modification of animal cells in culture Obtaining the drug through the genetic modification of animals

DK DE ES FR IE

IT NL AU SE UK CZ PL

IL US JP

69.3 64.5 47.7 78.8 50.5 36.3 63.3 46.7 49.1 51.1 60.3 56.3 59.8 54.1 80.7 22.3 20.5 20.4 12.3 19.6 34.5 25.9 30.8 38.2 24.6 23.5 18.1 20.0 22.0 12.0 0.8 2.7 3.7 0.9 3.7 5.2 1.1 4.1 1.2 2.9 2.4 2.3 3.7 3.0 1.1 4.2 2.5 6.2 4.1 6.9 5.0 6.9 3.9 5.8 9.0 7.4 7.7 7.8 11.1 1.9 0.5 0.5 1.0 0.2 1.9 0.9 0.4 0.5 0.1 1.0 0.1 0.9 1.0 0.3 0.1 2.8 9.3 20.8 3.7 17.3 18.1 2.4 14.1 5.6 11.3 6.4 14.7 7.6 9.5 4.3 18.4 17.4 20.5 10.5 21.0 31.9 20.9 26.5 32.8 24.6 21.0 19.3 17.5 19.5 10.0 49.3 61.5 42.0 54.9 40.6 34.1 40.4 38.3 44.8 37.2 37.2 48.2 41.0 36.5 56.1 4.2 4.1 8.4 4.8 9.3 8.5 6.7 6.3 3.6 5.9 5.4 6.2 12.3 6.4 4.2 24.6 13.3 16.4 19.5 19.0 16.6 29.2 15.8 14.0 22.7 29.6 20.4 23.6 22.5 15.7 0.9 1.1 2.2 1.0 3.2 2.8 0.8 2.5 1.2 2.7 1.7 1.6 3.3 3.1 1.1 2.6 2.6 10.6 9.3 6.9 6.1 1.9 10.6 3.6 6.9 5.1 4.2 2.4 12.1 12.9 2.7 4.5 6.1 3.8 7.6 7.9 5.6 6.9 6.0 8.1 7.2 6.0 5.3 8.3 2.8 14.3 7.6 9.9 15.3 15.2 11.0 19.0 9.3 8.4 19.0 19.9 15.1 16.2 17.0 8.9 20.1 17.3 18.4 15.0 15.5 23.6 23.7 13.7 21.1 16.7 15.7 16.6 21.9 15.9 18.0 51.1 42.2 33.8 39.2 32.3 33.6 41.3 31.0 50.5 39.1 37.2 42.7 31.7 30.2 37.6 4.9 2.7 9.6 2.7 6.3 4.5 4.3 7.7 2.6 4.0 4.7 3.8 10.4 7.5 5.1 6.8 25.7 22.2 24.1 23.1 19.3 6.1 31.3 11.4 13.1 15.3 15.7 14.5 21.1 27.5 89.7 84.2 68.2 92.3 73.8 68.6 89.1 74.7 85.6 79.7 86.4 77.6 80.7 81.9 92.6 83.7 83.1 60.7 78.4 64.9 71.0 83.6 70.9 88.1 73.4 75.9 71.5 72.4 75.4 73.1 76.5 51.7 43.1 57.1 47.5 44.4 75.0 41.3 64.9 61.4 68.2 60.1 58.2 63.8 48.4 24.0 21.7 23.3 18.5 23.5 30.3 30.3 20.0 23.8 21.9 21.4 21.2 34.8 25.2 20.1 6.0 3.8 9.6 3.5 9.3 6.6 5.3 8.5 3.6 6.7 5.8 5.4 13.5 10.9 5.4

DK: Denmark / DE: Germany / ES: Spain / FR: France / IR: Ireland / IT: Italy / NL: Netherlands / AU: Austria / SE: Sweden / UK: United Kingdom / CZ: Czech Republic / PL: Poland / IL: Israel / US: United States / JP: Japan

174

5 Public views and attitudes to pharming

Table 5.20: Support for plant pharming by awareness “I’m going to read out some sentences about the genetic modification of plants in order to produce pharmaceutical drugs for treating human diseases. Please tell me how much you agree or disagree with each, using a scale from 0 to 10, where 0 means you totally disagree and 10 means you totally agree: Should be supported.”

Denmark Germany Spain France Ireland Italy Netherlands Austria Sweden Great Britain Czech Republic Poland Israel United States Japan

Have heard or read information 6.7 5.2 7.0 5.0 6.0 5.6 6.3 4.7 5.4 6.1 7.0 6.5 7.0 6.7 5.1

Have not heard or read information 5.9 4.3 5.8 4.9 5.1 5.0 5.9 4.0 5.0 5.5 6.1 5.8 6.2 6.3 4.6

Table 5.21: Support for animal pharming by awareness “Can you please tell me how much you agree or disagree with each of the statements I am going to read out? Please use a scale from 0 to 10, where 0 means you totally disagree, and 10 means you totally agree. You may, of course, choose any score between 0 and 10: Should be supported.” Have heard or read Have not heard or read information information Denmark 5.3 4.2 Germany 4.1 3.1 Spain 5.8 4.3 France 3.6 2.8 Ireland 4.7 3.7 Italy 4.9 4.1 Netherlands 4.5 3.9 Austria 3.3 2.7 Sweden 3.7 3.2 Great Britain 4.5 3.4 Czech Republic 5.6 4.4 Poland 4.8 4.0 Israel 5.6 4.7 United States 4.6 3.9 Japan 4.4 4.0

5.9 Tables

175

Table 5.22: Support for animal pharming by awareness of a specific development “British scientists have recently created genetically modified hens by inserting human genes in them. The eggs of these hens contain proteins that can be used for producing drugs to treat certain serious diseases. To what extent do you see it as acceptable or not to genetically modify hens in order to produce pharmaceutical drugs? Please use a scale from 0 to 10, where 0 means it is totally unacceptable and 10 that it is totally acceptable. You may, of course, choose any score between 0 and 10.” Have heard or read Have not heard or read information information Denmark 5.7 5.0 Germany 4.4 3.8 Spain 6.3 5.1 France 4.1 3.5 Ireland 5.5 4.0 Italy 4.7 4.2 Netherlands 5.2 4.6 Austria 4.7 2.9 Sweden 5.2 3.7 Great Britain 4.9 4.1 Czech Republic 6.4 5.3 Poland 5.0 4.2 Israel 6.6 5.8 United States 5.6 4.0 Japan 5.1 4.7

176

5 Public views and attitudes to pharming

5.10 References Ajzen I (2005) Attitudes, Personality and Behavior (2nd Edition). Open University Press, Maidenhead-Berkshire Allum N, Sturgis P, Tabourazi D, Brunton-Smith I (2008) Science knowledge and attitudes across cultures: a meta-analysis. Public Understanding of Science 17:35–54 Almond G (1950) The American People and Foreign Policy. Harcourt, Brace and Company, New York Bauer M (ed) (1995) Resistance to new technology. Cambridge University Press, Cambridge Bauer MW, Gaskell G (eds) (2002) Biotechnology. The Making of a Global Controversy. Cambridge University Press, Cambridge Bauer MW, Allum N, Miller S (2007) What can we learn from 25 years of PUS survey research? Liberating and expanding the agenda. Public Understanding of Science 16:79–95 Beck U (1992) The Risk Society: Toward a new Modernity. Sage, London Beck U (1999) World Risk Society. Blackwell Publishers, Oxford Ben-David J (1984) The Scientist’s Role in Society (2nd Edition). The University of Chicago Press, Chicago and London Bishop GF (2005) The Illusion of Public Opinion. Fact and Artifact in American Public Opinion Polls. Rowman & Littlefield Publishers, Inc., Lanham-BoulderNew York, Toronto, Oxford Bollnow OF (1958) Wesen und Wandel der Tugenden. Ullstein Taschenbücher-Verlag, Frankfurt/M, Berlin, Wien Campbell A, Converse PE, Miller WE, Stokes DE (1960) The American Voter. The University of Chicago Press, Chicago, IL Converse P (1964) The nature of belief systems in mass publics. In: Apter DE (ed) Ideology and Discontent. Free Press, New York, pp 206–261 Converse P (1970) Attitudes and non-attitudes: continuation of a dialogue. In: Tufte ER (ed) The Qualitative Analysis of Social Problems. Addison-Wesley, Reading, MA Crettaz von Rote F (2008) Mapping Perceptions of Animal Experimentation: Trend and Explanatory Factors. Social Science Quarterly 89(2):537–549 Delli Carpini MX, Keeter S (1996) What Americans Know about Politics and Why it Matters. Yale University Press, New Haven, CT Dietrich H, Schibeci R (2003) Beyond Public Perceptions of Gene Technology: Community Participation in Public Policy in Australia. Public Understanding of Science 12:381–401 Einsiedel EF, Jelsøe E, Breck T (2001) Publics at the technology table: The consensus conference in Denmark, Canada, and Australia. Public Understanding of Science 10:83–98 Frewer LJ (1999) Risk Perception, Social Trust, and Public Participation into Strategic Decision-making: Implications for Emerging Technologies. Ambio 28:569–74 Gaskell G (1997) Europe ambivalent on Biotechnology. Nature 387:845–847 Gaskell G, Bauer MW (eds) (2006) Genomics & Society. Legal, Ethical & Social Dimensions. Earthscan, London-Sterling, VA Gross P, Levitt N (1994) Higher Superstition. The Academic Left and Its Quarrell with Science. The Johns Hopkins University Press, Baltimore and London Gross P, Levitt N, Lewis MW (1996) The Flight from Science and Reason. The New York Academy of Sciences, New York, NY Haack S (2003) Defending Science –within Reason. Between Scientism and Cynicism. Prometheus Books, Amherst, NY

5.10 References

177

Hagelin J, Carlsson HE, Hau J (2003) An overview of surveys on how people view animal experimentation: some factors that may influence the outcome. Public Understanding of Science 12:67–81 Hard M, Jamison A (1998) The Intellectual Appropriation of Technology. The MIT Press, Cambridge, MA Hirschman A (1970) Exit, Voice, and Loyalty. Harvard University Press, Cambridge, Massachusetts and London, England Holton G (1992) How to think about the ‘anti-science’ phenomenon. Public Understanding of Science 1:103–128 House of Lords (Select Committee on Science and Technology) (2000) Science and Society, The Stationery Office, London Joss S, Durant J (eds) (1995) Public participation in science. The role of consensus conferences in Europe. Science Museum, London Kolakowski L (1967) Der Mensch ohne Alternative. R. Piper & Co Verlag, München Ladd EC, Bowman KH (1996) Public Opinion in America and Japan. The AEI Press, The Roper Center for Public Opinion Research, Washington, DC, Storrs, CT Levitt N (1999) Prometheus Bedeviled. Science and the Contradictions of Contemporary Culture. Rutgers University Press, New Brunswick, New Jersey, London Lévy-Leblond JM (1992) About misunderstandings about misunderstandings. Public Understanding of Science 1:17–21 Lyberg L, Biemer P, Collins M, de Leeuw E, Dippo C, Schwarz N, Trewin D (eds) (1997) Survey Measurement and Process Quality. John Wiley & Sons, Inc, New York Marx L (1988) The Neo-Romantic Critique of Science. In: Marx L, The Pilot and the Passenger. Essays on Literature, Technology, and Culture in the United States. Oxford University Press, New York-Oxford, pp 160–178 Marx L (1998) The Domination of Nature and the Redefinition of Progress. In: Marx L, Mazlish B (eds) Progress. Fact or Illusion? The University of Michigan Press, Ann Arbor Mertig A, Dunlap RE (1995) Public Approval of Environmental Protection and Other New Social Movement Goals in Western Europe and the United States. International Journal of Public Opinion Research 7(2):145–156 Miller JD, Suchner RW, Voelker AM (1980) Citizenship in an Age of Science. Changing Attitudes Among Young Adults. Pergamon Press, New York Miller JD (1983a) The American People and Science Policy. Pergamon Press, New York Miller JD (1983b) Scientific Literacy: A conceptual and empirical review. Daedalus 112(2):29–48 Miller S (2001) Public Understanding of Science at the Crossroads. Public Understanding of Science 10:115–120 Morrell JB (1990) Professionalization. In: Olby RC, Cantor GN, Christie JRR, Hodge MJS (eds) Companion to the History of Modern Science. Routledge, London and New York, pp 980–989 Nelkin D (1995) Forms of Intrusion: comparing resistance to information technology and biotechnology in the USA. In: Bauer M (ed) (1995) Resistance to new technology. Cambridge University Press, Cambridge, pp 379–390 Pardo R, Calvo F (2002) Attitudes toward science among the European public: A methodological analysis. Public Understanding of Science 11:155–195 Pardo R (2003) Attitudes toward Embryo Experimentation in Europe. In: Solter D et al. Embryo Research in Pluralistic Europe. Springer, Berlin Heidelberg New York, pp 157–203 Pardo R, Calvo F (2006a) Mapping Perceptions of Science in End-of-Century Europe. Science Communication 28(1):3–46

178

5.10 References

Pardo R, Calvo F (2006b) Are Europeans really antagonistic to biotech? Nature Biotechnology 24 (4): 393–395 Pardo R, Calvo F (2008) Attitudes to Embryo Research, Worldviews and the Moral Status of the Embryo Frame. Science Communication 30(1) (in press) Perrow C (1999) Normal Accidents. Princeton University Press, Princeton Prewitt K (1983) Scientific Illiteracy and Democratic Theory. Daedalus 112(2):49–64 Priest SH (2001) A Grain of Truth. The Media, the Public, and Biotechnology. Rowman & Littlefield Publishers, Lanham Boulder New York Oxford Priest SH, Bonfadelli H, Rusanen M (2003) The “Trust Gap” Hypothesis: Predicting Support for Biotechnology Across National Cultures as a Function of Trust in Actors. Risk Analysis 23(4):751–766 Ross A (ed) (1996) Science Wars. Duke University Press, Durham and London Schuman H, Presser S (1996) Questions & Answers in Attitude Surveys. Sage Publications, Thousand Oaks, London, New Delhi Siegrist M (2000) The influence of trust and perceptions of risks and benefits on the acceptance of gene technology. Risk Analysis 20(2):195–204 Simon HA (1982) Models of Bounded Rationality. Vol 1–2. The MIT Press, Cambridge, MA Sjoberg L (2004) Principles of risk perception applied to gene technology. EMBO Reports 5:47–51 Slovic P (1987) Perception of Risk. Science 236:280–285 Slovic P (2000) The Perception of Risk. Earthscan, London-Sterling, VA Sokal A, Bricmont J (1998) Fashionable Nonsense. Postmodeern Intellectual’s Abuse of Science. Picador USA, New York Solter D, Beyleveld D, Friele MB, Holòwka J, Lilie H, Lovell-Badge R, Mandla C, Martin U, Pardo Avellaneda R (2003) Embryo Research in Pluralistic Europe. Springer, Berlin, Heidelberg, New York Sturgis P, Cooper H, Fife-Schaw C, Shepherd R (2004) Genomic Science: Emerging Public Opinion. In: Park A, Curtie J, Thomson K, Bromley C, Phillips M (eds) British Social Attitudes. The 21st Report. Sage Publications-NatCen, London, Thousand Oaks, New Delhi Sturgis P, Cooper H, Fife-Schaw C (2005) Attitudes to biotechnology: estimating the opinions of a better-informed public. New Genetics and Society 24 (1):33–58 Thomas G, Durant J (1987) Why Should we Promote the Public Understanding of Science? In: Shortland M (ed) Scientific Literacy Papers. Oxford Department of External Studies, Oxford, pp 1–14 Thurstone LL, Chave EJ (1929) The Measurement of Attitude. The University of Chicago Press, Chicago, IL Tourangeau R, Rips LJ, Rasinski K (2000) The Psychology of Survey Response. Cambridge University Press, Cambridge Turner CF, Martin E (eds) (1984) Surveying Subjective Phenomena (two volumes). Russell Sage Foundation, New York Weinberg S (2001) Facing Up. Science and Its Cultural Adversaries. Harvard University Press, Cambridge, MA Winston PH (1984) Artificial Intelligence. Addison-Wesley, Reading, MA, Menlo Park, CA Worcester RM (1993) Public and Élite Attitudes to Environmental Issues. International Journal of Public Opinion Research 5(4):315–334 Wynne B (1996) May the Sheep Safely Graze? A Reflexive View of the Expert-Lay Knowledge Divide. In: Lash S, Szerszynski B, Wynne B (eds) Risk, Environment & Modernity. Towards a New Ecology. Sage Publications, London, Thousand Oaks, New Delhi, pp 44–83 Zack P, Knack S (2001) Trust and Growth. Economic Journal 111:295–321

6 The ethical evaluation of pharming

6.1 Introduction Typical questions concerning moral problems of pharming are, for example, ‘Is it morally acceptable to make animals suffer for the production of drugs for use in humans?’, ‘Does the intervention in the genetic makeup of animals and plants go against nature?’, ‘Is the environmental risk of growing genetically modified plants in open fields acceptable?’ These and other questions are discussed in this chapter with the first goal being to clarify how certain moral standpoints on pharming are structured. For this the discussion in the current chapter can draw on the analysis of public views on pharming presented in chapter 5. The analysis of what the main views on the use of plants and animals for drug production are, and how they are dependant on contextual factors (medical and social goals, type of plant and animals, locus of expression, and other specifics), can serve as a basis for a better understanding of what the concerns with pharming are. Given this map of moral arguments for and against pharming, a second goal of this chapter will be to develop recommendations for mastering moral controversies on pharming. Hardly any other scientific or technological development has caused such far-reaching controversies as gene technology. The status of these debates is, however, interpreted quite differently: some take the objections of concerned persons to be urgent and necessary. From this perspective “morality” and “ethics” are frequently seen as strongholds against the alleged lack of scruples and greed of advocates and users of biotechnology. Others assess this scepticism as a mere obstacle to research, as preventing the beneficial advancement of science. Sometimes this perspective is connected with the accusation that the sceptics suffer from scientific illiteracy and an unfounded fear of science. Nonetheless it seems that most, though unfortunately not all, disputants believe that the discussion should not be pursued as a kind of religious war. It is obvious that sustainable solutions to controversies are possible only if those solutions are not founded on power, violence, or deception, but on convincing one’s opponent. What counts are reasons favouring one’s own position and able to refute the opposing view. In the following section an approach to ethics will be developed that may prove helpful in mastering moral conflicts in an argumentative manner. In the subsequent sections this approach will then be applied to the moral evaluation of pharming.

180

6 The ethical evaluation of pharming

6.2 Foundations of moral reasoning If a moral controversy involves more than just two interest groups, and if the relevant interests are manifold and complex – as is the case with debates of societal relevance – it may be advisable to reconstruct and clarify the debate and the argumentative standards governing it in a systematic approach. It has always been one of the main tasks of ethics to work out rules for moral discourse and to test if these rules are adequate for mastering moral conflicts. From this perspective ethics is clearly not the right instrument for providing final, irrevocable solutions to moral conflicts: the ethicist is not in the possession of “higher” insights and has no privileged access to absolute moral values or principles that would confer such competency to him/ her. Rather, the specific role of the ethicist should be to counsel the concerned parties. He/she can advise the involved persons on suitable argumentation standards or give guidance for mastering conflicts. In the light of this, it becomes clear that the professional discourse between ethicists cannot replace societal procedures of decision-making. Nonetheless, it might be the case that society will benefit from the pool of suggestions and recommendations developed in professional debates and stored in the philosophical tradition. According to this view of the practical task of ethics sketched above and elaborated below, this chapter will not present final answers concerning the moral problems of pharming. Instead this chapter aims to analyse the moral arguments underlying conflicts concerning pharming, and thereby to outline some of the possible courses of action that may help to master these conflicts. The terms ‘moral’ and ‘ethics’ are frequently used synonymously in bioethical debates. Because of this, however, one loses the important differentiation between, on the one hand, a set of moral convictions, norms and recommendations to act that is subject to multiple changes and, on the other hand, the method that is concerned with the critical examination of those agglomerates. Everyone has a moral1, and for any grouping one can collect or rather reconstruct the set of convictions, norms and recommendations as the binding morals specific to this grouping.2 According to this understanding there exists not a single moral but a great variety, for instance professional morals that are thought to be endowed with binding force only for certain 1

2

We hope that using ‘moral’ in the singular does not give too much trouble to the native speaker. It appears that in English the plural ‘morals’ is used most often, where in German there is the singular ‘Moral’ as well as the plural ‘Moralen’ – similar in Latin: ‘mos’ and ‘mores’. To use the term ‘morality’ in this context does not seem to be adequate since ‘morality’ describes the fact that there are morals rather than the actual set of moral convictions a certain individual cherishes – a ‘moral’ as we call it. Cf. also Frankena 1973:4–9. This task is also part of social science analysis of the values in a particular society as is shown for pharming in chapter 5.

6.2 Foundations of moral reasoning

181

groups (like professional codes of conduct issued by medical associations), or religious morals, regulating social discourse among the members of a specific religious group. Morals are in many – concerning the everyday need for decision-making, probably in most – cases a sufficient basis from which to address questions on what to do and what not to do. Beyond the routine of decisionmaking guided by morals there are, however, numerous cases (particularly in some areas) where morals do not deliver satisfying ‘recommendations to act’. The rule, for instance, that ‘a doctor must always act in favour of his/ her patient’s well-being’ may well be widely accepted. But does a paediatrician act ‘in favour of his/her patient’s well-being’ when he/she performs lifesustaining measures on a severely disabled newborn? Questions like this demonstrate the limits of the efficacy of morals once more. A philosopher, wishing to contribute to the solution of bioethical issues, should aim to clarify the moral concepts involved and to reconstruct the arguments used in the discussion. It should be made clear, however, that these arguments are developed on the basis of an underlying ethical theory, and that a variety of different ethical theories has been developed during the history of ethics, so that the recommendations for solving a particular conflict might well differ depending on the background assumptions adopted.3 Though the plurality of ethical approaches is striking, a closer look shows that there is a decisive exception to this plurality. The participants in the bioethical debate have one goal in common: i.e. the wish to master conflicts with the help of moral argumentation. This claim is an empirical observation that would need further support if it is to serve as a normative foundation for an ethical theory. However, for the purpose of this study it is sufficient to point out that all those participating in the debate of moral problems in the life sciences subscribe to the view that ethics develops principles for the mastering of moral conflicts in an argumentative manner.4 A moral conflict arises when two actors perform actions or pursue goals that are incompatible with each other.5 Mastering conflicts is not restricted to the solution of conflicts that already exist; also the avoidance of future, but foreseeable conflicts is of importance, especially when bioethics ought 3

4

5

In the following sections the major ethical approaches relevant to the moral evaluation of pharming will be introduced. For a general introduction to environmental and animal ethics see for example the anthology of Light and Rolston 2003, and Krebs 1999. This position has been worked out in greater detail in Thiele 2004. Similar approaches can be found in Charlesworth 1993, Gethmann 1992, Hare 1997, and Hegselmann 1998. Since the aim of this chapter is to develop recommendations for the mastering of public conflicts concerning pharming, we do not deal here with the common problem that arises when an individual actor pursues incompatible goals – a situation that might be called an intra-individual conflict.

182

6 The ethical evaluation of pharming

to be used as a tool for the active shaping of science and technology policy. In order to successfully master conflicts independently from the specific situation in which the conflict arises, it is in principle sufficient to recommend rules of action that are justifiable not to everyone but only to the actual or potential conflict partners. The identification of those groups that can be conflict partners, therefore, is an essential part in mastering moral conflicts. The problems that are the object of professional bioethical investigation involve, as a rule, a great number of conflict-partners. Each party claims that the moral the party has adopted is or should be binding for all the other parties, too – with the result that the bioethicist is confronted with a multiplicity of moral convictions, norms and recommendations to act that are at least partly incompatible with each other. In a certain sense, mastering a conflict in an invariant manner, with respect to the involved parties, may be expressed as developing universally valid moral norms and norm-systems. It would be presumptuous to try to bring the ongoing debate on the most convincing arguments in environmental and animal ethics to a conclusion in a book such as this one. Moreover, an in depth discussion would demand not only a discussion of the different arguments separately but also in their different combinations, which makes the task even more demanding. To give one example: prima facie one might expect that an anthropocentrist, i.e. somebody claiming that only humans are bearers of rights, might not have much to say against pharming. But it could turn out that the same anthropocentrist, being at the same time a strong adherent of naturalness, becomes a fierce opponent of pharming because pharming animals and plants are created in a quite unnatural way. Thus, a combination of the different moral “parameters” creates a large number of moral positions on pharming.6 Certainly, in academic discussions aimed at the improvement of the method of ethics, sharpening and developing theories of ethics arises from the virtuous desire for intellectual advance. In this book, however, we pursue a different aim, namely to analyse the technical, legal, social, and moral factors that shape the further development and practical application of pharming, and to develop recommendations for societal conflicts caused by this technique. Exclusively searching for the “true” ethical position might not be the most promising strategy when one aims to master moral conflicts. Instead, in this book we take as a starting point for the moral evaluation of pharming the following assumption: as long as our societies allow, in principle, the use of animals and plants for purposes such as food production, they should also allow, in principle, the use of animals and plants for the improvement of human health. Claiming that similar actions in different areas of our life 6

For a more detailed discussion of these positions see below sections 6.3.1 to 6.3.4.

6.2 Foundations of moral reasoning

183

should be evaluated similarly – a claim that is sometimes called ‘the principle of consistency’ – obviously is contestable and needs further clarification and elaboration.7 From the perspective taken here, what should be consistent is the moral evaluation of actions. We are neither claiming that all the actions that individuals perform should be consistent with each other, nor do we claim that individuals should make their own moral systems consistent.8 Also, we are not claiming that actors having a moral conflict with each other should synchronize their respective moral systems, so that they eventually adopt the same moral system. The ‘principle of consistency’ as it is proposed here is no more and no less than a rule for organising moral debates. It should be recognized that the demand for consistency works in both ways: if we should conclude that the use of animals or plants for the purpose of pharming is morally unacceptable, this could also imply that we should rethink our use of animals and plants for food production and pleasure. For example, though poor animal welfare in order to produce cheap food is widespread, it remains controversial. There is public debate about it, and there are efforts at EU and at national level in some European countries to improve animal welfare standards, either by stricter legislation, information, or labelling. Because of this ongoing process, the current standard of treating animals cannot be used as an incontestable reference point for an evaluation of animal treatment in new areas such as pharming. Claiming that ‘similar actions should be evaluated similarly’ presupposes that one is able to show how and to what extent two actions actually are similar. The problem of comparability is a serious one and looms large in the rest of this chapter. Firstly, describing the morally relevant qualities of an action, which is a prerequisite for comparing it to another action, is highly dependent on background assumptions. Whether, for example, altering the promoter region of a certain plant gene amounts to an ‘unnatural modification’ of the plant or to a ‘simulation of a possible natural evolutionary process’ is largely dependent on what one considers to be ‘natural’ and ‘unnatural’.9 Secondly, whether two actions are similar depends on what one takes to be the relevant parameter of comparison. Cultivating a traditional food crop such as potatoes and cultivating a genetically modified version of it may be similar, or even identical, in terms of the agricultural techniques used but quite dissimilar in terms of the risk these plants pose to the environment. 7

8

9

This plea for consistency in evaluating moral problems has been made, for example, by Hare 1981: section 6.1, and has been specifically elaborated for risk evaluations by Gethmann in Gethmann and Kloepfer 1993:42–45. It might be worthwhile debating whether the psychological finding that most individuals live with inconsistent belief systems refutes the philosophical claim that they shouldn’t, but this is beyond our focus here. See below section 6.3.2 for a discussion of the concept of naturalness.

184

6 The ethical evaluation of pharming

In the following sections, some of the more important moral perspectives responsible for the moral debate on pharming are introduced. Furthermore, using the method proposed here some conclusion concerning pharming will be drawn. Beforehand, however, it needs to be acknowledged that, to some readers, the approach to bioethical problems outlined here will fall short of what they think the “real” or “proper” task of ethics is. Indeed, many prominent ethical approaches aim at solving bioethical conflicts by reference to such concepts as moral truth, the highest good, or ultimate justifications for moral judgements in contrast to aiming at the minimally sufficient commitments for mastering conflicts. This chapter is not the appropriate place for a detailed discussion of the strengths or weaknesses of these approaches to bioethics. It should be noted, however, that the pragmatic approach favoured here should not be considered as simply a truncated or defective form of debating – this would be to confound what might be a ‘practical’ or convenient way of acting in a certain conflict situation with what is a ‘pragmatic’, but nonetheless systematic, way of dealing with moral conflicts in general. The authors are well aware that the interest in moral reasoning is not limited to philosophically reconstructing moral conflict situations and moral argumentations, as is done in this chapter. Elucidating the way in which psychic, societal, and cultural mechanisms influence the generation and transformation of moral convictions over time is worthy of scientific effort.10 Knowing more about the moral convictions that members of the public actually hold, and knowing more about the psychic and societal mechanisms influencing public opinion, may contribute to the successful mastering of moral conflicts in biotechnology. Emphasising the worth of arguments for solving moral conflicts in this chapter should, therefore, not be misunderstood: we are not denying the importance (or even existence) of emotional and social influences on moral reasoning.11 Rather, the focus on moral argumentation is thought to enable us to find solutions to moral conflicts; while respecting the emotional and social influences, they can be transcended when this is helpful for conflict solving.

6.3 Common moral concerns regarding pharming In order to build on the results of chapter 5, we arranged the content of the following sections 6.3.1 to 6.3.4 roughly according to what one might call “clusters” of moral concerns that can be found through the analysis of public attitudes on pharming. We speak of a cluster, because public attitudes do 10 11

Such considerations take place partly in chapter 5. To put it even more bluntly: many public debates on moral issues involve rhetoric, persuasion, emotions and even ideology rather than moral reasoning as it is advocated here. It is obvious that the ethical approach suggested here will fail when one or more of the stakeholders is not willing to master their conflicts in an argumentative manner.

6.3 Common moral concerns regarding pharming

185

not consist of a systematic set of ethical principles or concepts that we usually find within a professional ethical discourse. Concerning public attitudes, we rather have to deal with a blurred mixture of partly overlapping intuitions, supplemented with bits and pieces of supporting moral arguments. Nonetheless, it seems preferable to start from an imprecise moral map drawn from reality, rather than to design a precise map that might only depict an ideal landscape. The reason for this is – as was already demonstrated in the last section – that the aim of this book is to come up with recommendations for a morally acceptable, but also socially accepted, development of the field of pharming. In view of this, it seems more promising to test the intellectual patience of professional ethicists rather than to lose the attention of the public worried about pharming.

6.3.1 The moral status of plants and animals The analysis in chapter 5 has shown that, in general terms, public attitudes towards the genetic modification of animals for the purpose of pharming are rather negative. Though many people assume that animal pharming is, to a certain extent, a useful and promising technique, the majority’s rejection of animal pharming is striking and is, it seems, partly due to strong views on the inherent value of animals. The view on plant pharming is less negative, but is also related to attitudes concerning the moral status of plants. Since the available data do not allow us to draw substantial conclusions on what exactly are the public’s underlying arguments on the moral value of animals and plants, one has to turn to ecological ethics for a clarification of the various perspectives on the moral status of animals and plants. In environmental ethics the traditional assumption that humans are the centre of moral reasoning is increasingly challenged. Instead, positions are favoured that shift the emphasis away from humans and claim that also animals, plants, and in some cases rocks, whole landscapes, and ecosystems may have entitlements or value and should, therefore, play a more important role in moral reasoning on interventions in nature. In the following we will give a brief overview of prominent positions in ecological ethics. We distinguish four positions that are fundamentally different with respect to the criterion that is decisive for ascribing moral entitlements or value to living and non-living entities.12 1) Proponents of anthropocentrism claim that only humans, having certain specific traits (for example rationality), qualify as bearers of moral entitlements or as having moral value in themselves.13 For an anthropocentrist the value of non-human animals, plants, nature, biotopes etc. is instrumental, i.e. these entities have entitlements or value only indirectly, insofar as 12 13

For an introduction see Krebs 1999. For a recent volume devoted especially to animal ethics see Sunstein and Nussbaum 2004. For an overview on the variety of anthropological positions see Krebs 1999, part II.

186

6 The ethical evaluation of pharming

their protection serves human aims – for example, technical, aesthetic, religious or pedagogic aims. From an anthropocentristic perspective neither pharming animals nor pharming plants are morally valuable in their own right, but only indirectly by way of their value for a human being. 2) The position called pathocentrism differs from anthropocentrism in that it takes the capacity to feel – and not the fact of being human or having specifically human traits – as the decisive criterion for having entitlements or having value. Pathocentristic positions claim that most but not necessarily all humans have moral entitlements or value, and that some but not all animals qualify likewise.14 Plants and entities such as species, nature and ecosystems are not usually seen as being endowed with a capacity to suffer and can, therefore, only have instrumental value. From the pathocentristic perspective pharming animals have moral entitlements in their own right, whereas pharming plants have instrumental value only insofar as they are valuable for those humans and animals that qualify as moral agents. 3) Advocates of biocentrism adopt the view that not only sentient beings but all forms of life have a certain value. In this case only non-living entities such as rocks and landscapes do not have moral value on their own, but are valued only because of their instrumental value for somebody. 4) Finally, an approach called holism, ecocentrism or sometimes physiocentrism, assumes that, in addition to animals and plants, also anorganic structures like mountains or rivers – have value not merely as a means for human aims, but have value in themselves. Sometimes it is assumed that even such ephemeral things as life, beauty, order, or diversity have value as such.15 Clearly, for biocentrists as well as holists both pharming animals and plants have moral value in their own right. This list of groups of positions on the moral status of animals and plants can be lengthened.16 Moreover, even the mentioned positions can be easily split in sub-groups. For example, some biocentrists and holists claim that all entities qualifying as value-bearers have the same value, whereas others think that there is a hierarchy of values.17 Further issues reflected in ethi14 15

16

17

Contemporary advocates of pathocentrisms are, for example, Peter Singer, e.g. 1975, and Bernard Rollin, e.g. 2006. See also Krebs 1999, part IV. An issue of debate amongst holists is whether nature as a whole has value (e.g. Leopold 1949) or whether everything in nature has value (e.g. Siep 2004:270 et seq.). For example, one could add an approach focusing on the ‘dignity’ of animals (and plants). Since this approach is drawing from heterogeneous philosophical and theological sources discussed elsewhere in this chapter, we do not list “dignity ethics” as a separate position (see Richter 2007 for an overview). An example for an “egalitarian” variant of biocentrism, where all living entities have the same value, likely is the position of Albert Schweitzer. A hierarchical holism using the concept of scala naturae is proposed by Siep (2004:270 et seq.).

6.3 Common moral concerns regarding pharming

187

cal approaches are, amongst other things, the justifications of the supposed values and entitlements, the exact scope and quality of values and entitlements, and the way of weighing colliding values and entitlements. Over the past decades an elaborate debate on genetic engineering of plants and animals has developed, about which the above can only give a cursory and abridged outline. Moreover, it is not to be expected that the disagreement on the correct ethical approach to environmental issues will be resolved soon. Despite this variety of ethical approaches and the disagreement about them, the authors think that pathocentrism is arguably a fruitful criterion for the evaluation of animal pharming – an approach that should be acceptable to proponents of most of the ethical standpoints sketched above for the following reasons: pathocentrism was introduced as the view that the capacity to suffer is the decisive criterion for having moral entitlements. Though not all animals have this capacity to suffer, it is agreed that the species typically used for pharming purposes – for example cattle, goat, sheep, pigs, rabbits, and poultry – can suffer18 and, therefore, qualify as moral agents from the pathocentric perspective. In contrast, the anthropocentrist claims that animals only have instrumental value, for example economic or pedagogic value for humans. Many anthropocentrists would agree that educating people to treat animals kindly is a fruitful pedagogic principle.19 Since the cruelty of a certain treatment is directly related to the suffering it inflicts, it is likely that many anthropocentrists will accept pathocentrism at least for evaluating the treatment of non-human animals – including animal pharming. Still, an anthropocentrist could claim that a certain, let us suppose, quite harmful treatment in the course of producing a pharming drug does not fall under the pedagogic proviso since, let us suppose again, it is the only way to develop an otherwise unavailable, lifesaving therapy. Clearly, this and other examples show that the anthropocentrist is not forced to accept a pathocentric approach for the treatment of non-human animals. Even though, in many cases, the anthropocentrist can adopt pathocentrism without giving up the core of his/her own position. At first sight biocentrists and holists will not have much use for pathocentrism, since they claim that many entities lacking the capacity to suffer nonetheless do have moral value. However, some biocentrists and holists assume that moral entities can be ordered in a hierarchical order according to certain criteria.20 One of these criteria could and probably will be the 18 19

20

EFSA 2005. The most cited reason is that those educated in this way will consequently treat other humans less cruelly too. The classic formulation is to be found in Kant’s “Metaphysik der Sitten”, II. 1., §17. A similar position is worked out in Tugendhat 1993, lecture 9:177 et seq. In contrast, economic interests in animals will serve to reject wanton cruelty (resulting in suffering) only if wanton cruelty endangers these interests. For example Attfield 1998:12 et seq., Siep 2004:270 et seq.

188

6 The ethical evaluation of pharming

capacity to suffer.21 Of course, not all biocentrists and holists would accept the capacity to suffer as an adequate criterion for ranking moral entities, and obviously non-hierarchic versions of biocentrism and holisms have no use at all for pathocentrism. But in view of developing recommendations for a socially acceptable development of pharming, it would be advantageous if it can be shown that pathocentrism is an attractive approach even for some of the adherents of non-pathocentristic ethical approaches. Moreover, pathocentrism not only has plausibility as a theoretical construct, but also has a well-established application in animal welfare science. As has been shown in chapter 4, animal pharming causes considerable concerns about possible suffering in the animals used, especially in the experimental phase – due to some invasive laboratory procedures, possible developmental problems, and the unavoidable trial and error with regard to the effects of transgenesis. But the production phase, too, even if to a lesser extent, can cause animal suffering due to special hygienic housing conditions that may be needed for manufacturing a high quality product. A first recommendation drawn from the argumentation in this chapter is, therefore, that in developing biotechnological techniques such as pharming more weight should be given to the potential suffering inflicted on animals.22 It should be noted, however, that scientists and others generating and using pharming animals in general have a strong interest in healthy animals, simply because this is most likely to secure product quality. The recommendation formulated above is therefore addressed predominately to the law and to animal welfare research.23 In contrast to pharming animals, both anthropocentrists and pathocentrists will treat pharming plants mainly as potential hazards against which they have to protect humans and animals capable of suffering, but not as moral entities in themselves. Biocentrists and holists see pharming-plants as morally valuable in themselves. If, however, they accept a hierarchy of values the constraints on plant pharming – as opposed to human and nonhuman animal needs – will be rather weak. Again, it should be quite clear that not all ethical approaches will subscribe to this analysis. Even from the perspective of the aforementioned hierarchical version of holisms, the genetic modification of plants might be judged an immoral act if it amounts to an undue infringement of the plants’ natural capacity to flourish24. How21 22

23 24

Such a hierarchy would rank animals higher than plants, and animals with a higher capacity to suffer higher than those with a lower capacity to suffer. As a side note, one should keep in mind that our highly industrialized way of agriculture and animal husbandry in many cases causes abundant suffering in animals. Criticizing pharming on pathocentristic grounds should then raise some doubts about conventional agriculture and husbandry, too. See also the chapter on law and animal welfare. For example Siep 2004:199 et seq.

6.3 Common moral concerns regarding pharming

189

ever, in view of the ethical analysis in this chapter and with regard to the empirical moral landscape described in chapter 5, the authors do not think it necessary to recommend treating plants as having moral value in their own right. (This, of course, does not mean that plant pharming is wholly unproblematic as will be shown in the following sections.)

6.3.2 Naturalness Arguments against the use of gene technology frequently take the form of ‘Doing x is morally problematic, because it is unnatural’. Arguments drawing in one way or another on naturalness are not only ubiquitous in academic literature but also deeply entrenched in everyday morality, even in everyday language – with ‘nature’ regularly being a positive reference mark to which human action should be aligned.25 It should not be forgotten, however, that ‘nature’ is also used to denote negative influences from which humankind should be protected.26 In a recent study27, showed that the term ‘naturalness’ is used in a variety of meanings. At least, one should distinguish between genetic and qualitative naturalness. (In this context ‘genetic’ refers to the Greek word for ancestry and not to the biological term for the theory of heredity (genetics).) Genetic naturalness refers to the formation of the entity that we qualify as natural or unnatural, whereas qualitative naturalness refers to the current composition or appearance of that entity. A new fish breed, for example, that has been altered for more efficient husbandry in aquaculture would be unnatural in terms of genetic naturalness, but possibly quite natural in terms of qualitative naturalness. A further qualification of the degree to which an entity is genetically natural or unnatural could be developed with reference to the depth of intervention in the natural origin of the analysed entity: creating new plant varieties by classical breeding would normally be less thoroughgoing than introducing new genes not normally present in its genome. In this sense, creating transgenic plants or animals for pharming would be quite unnatural, since the organisms are not only brought into existence through pathways invented by humans but in addition cross otherwise nearly impermeable species-borders as, for example, in the case of goats expressing a human antithrombin gene.28 25 26

27 28

See chapter 5. Mill, for example, argues that nature “impales men, breaks them as if on the wheel, casts them to be devoured by wild beasts, burns them to death, crushes them with stones … All this, Nature does with the most supercilious disregard both of mercy and of justice …” (Mill 1998:29). Also telling is the very last sentence of Voltaire’s “Candide” where the protagonist ironically responds to Pangloss’ unshakable belief that ours is the best of all possible worlds: “il faut cultiver notre jardin.” Birnbacher 2006:chapter 1. See the appendix for case-descriptions.

190

6 The ethical evaluation of pharming

An example of a profound change in qualitative naturalness is the creation of blind hens. Conventional conditions in egg production cause massive welfare concerns: hens, when kept in flocks with insufficient space, frequently injure each other by pecking. To prevent this, hens’ beaks are frequently trimmed so that severe injuries decrease. Another approach is to breed blind hens that, according to a study, do not display feather-pecking nor other abnormal behaviour29. Though blind hens routinely used in egg production would be unproblematic from the pathocentristic perspective, they are quite clearly profoundly changed in terms of qualitative naturalness. Another aspect of qualitative naturalness is the speed with which certain entities change their appearance. Whereas a cultivated landscape that has developed over hundreds of years may appear qualitatively natural (though genetically highly artificial), sudden changes – for example the appearance of dominating monocultures due to modern agricultural techniques – may be experienced as a rapid loss in qualitative naturalness. In the same vein, the sudden appearance of “animal factories” producing pharmaceuticals may be seen as highly unnatural to some observers. Unnaturalness is not only used as a criterion for the description of certain changes, but frequently also as a criterion for the moral evaluation of these changes. In chapter 5, it has been shown that negative attitudes towards both animal and plant pharming are regularly based on the strong belief that these techniques go against nature. In the following, two arguments, linking descriptive unnaturalness to a moral rejection will be introduced.30 A first argument claims that pharming is unnatural because it is an objectionable way of “playing God”. Some authors use this argument to criticize scientific hubris based on ever growing possibilities to intervene in nature.31 That, indeed, the scientific potential to intervene in nature causes widespread aversion and rejection explains, but does not by itself justify, a moral rejection or even legal prohibition of these interventions. A possible justification for this claim would be the theological point that only God is the legitimate creator of entities in nature. However, in secular societies such as ours theological arguments are of only very limited power. In a second argument, the allegation of unnaturalness points toward an assumed stability or well-ordered arrangement of natural processes where to intervene might be more risky than is obvious at first sight. Indeed, when humans intervene in ecosystems that have evolved over millions of years – as for example in the case of plant and animal species invading and sometimes disturbing habitats – there can be far-reaching consequences. The 29 30 31

Ali and Cheng 1985. For further analysis see for example Comstock 2000, 2002; Birnbacher 2006, 2007:67–70. Midgley 2000:14. A related position can be found in Leon Kass’ writing (e.g. Kass 2000).

6.3 Common moral concerns regarding pharming

191

esteem for naturalness as an attempt not to trigger unwanted consequences by incautiously intervening in natural processes has, therefore, some plausibility.32 It seems rather questionable, however, whether from the general appeal to act cautiously in nature one can draw any specific guidelines for creating pharming animals and plants. The latter would demand the specification of potential damages and the probability of their occurrence, in other words the performance of a risk-assessment.33 In conclusion, the reproach that certain actions are unnatural is undeniably a frequent ingredient of public moral debates. It is doubtful, however, whether one would be well advised to make a “naturalness-reasoning” part of the attempt to master moral conflicts on genetic engineering and biotechnology.

6.3.3 Integrity When cattle are dehorned or dogs’ tails are docked this sometimes causes suffering. But even in those many cases where it does not cause suffering, many people find these procedures morally objectionable. The concept of integrity was developed as an ethical underpinning for such feelings.34 Integrity draws on notions of ‘wholeness’ and ‘completeness’ of animals, their ‘species-specific balance’, and their ‘capacity to maintain themselves independently’. A great number of conditions violating one or more of these criteria can be listed – including the already mentioned blind hens with infringed wholeness, the spontaneous death in modern fast-growing broiler breeds infringing their species-specific balance, or birth difficulties in meat-rich cattle breeds decreasing their capacity to maintain themselves independently. As in the discussion on the perceived unnaturalness of many modern procedures in agriculture and animal husbandry, it is undeniable that the perceived infringement of an animal’s or plant’s integrity is a real concern amongst the general public. As with naturalness, however, the ethical question is how to justify those concerns beyond the pure emotional aversion.35 The explicit aim of proponents of integrity is to develop an ethical principle going beyond the (pathocentristic) evaluation of the animal’s well-being36. In pursuing this aim, proponents of integrity draw on the fact that in many legal frameworks the right not to be physically violated without consent is granted even to those human individuals who cannot give consent themselves. From this an analogy is drawn to animals that are – like children – unable to give consent and should, therefore, likewise be protected from 32 33 34 35 36

Myskja 2006 and Räikkä and Rossi 2004 argue similarly. See section 7.5 below. For an introduction see Bovenkerk et al. 2002; Heaf and Wirz 2002; Heeger and Brom 2001; and Vorstenbosch 1993. This point is clearly seen by the proponents of integrity (e.g. Bovenkerk et al. 2002, pass.) Heeger and Brom 2001:270.

192

6 The ethical evaluation of pharming

physical violation.37 One reason for protecting a human being’s physical intactness is probably because by this rule one can best protect their potential for autonomy. In its usual conceptualizations, however, autonomy is bound to higher mental capacities – mental capacities that animals (at least pharming animals) arguably lack. A moral justification of integrity, by drawing analogies between the moral status of human beings in our legal culture and the moral status of animals can, therefore, be contested. Another justification for the right not be physically violated would be that thereby one can best prevent suffering. However, proponents of integrity claim that this concept is not grounded in a pathocentristic approach, so that this justification is debatable, too. Another attempt to ethically justify integrity is to claim that any animal or plant has a ‘natural good’ – or in a (neo-)Aristotelian terminology a ‘telos’ – which should be protected.38 This position allows us to classify certain alterations of an animal’s or plant’s makeup as violations of their natural good, or impediments to their flourishing.39 It proves difficult, however, to precisely define this telos and the extent to which it may be infringed or changed from a moral point of view. In view of these difficulties, even proponents of integrity consider their position to be in need of further development40. And though it can be argued that the concept of integrity may help to structure discussions, it seems advisable to use the better-established criterion of pathocentrism for the mastering of conflicts on pharming.

6.3.4 Aims and means of using and manipulating animals and plants for pharming It is a widely accepted rule that ‘the end does not justify the means’. Nonetheless, in many cases the evaluation of a morally problematic situation takes – to a certain degree – into account the purpose for which a given means is to be used. The analysis in chapter 5 shows that this holds for pharming, too. The purpose of the drugs produced clearly is a discriminatory factor for opinions on genetic modification of plants and animals 37

38

39

40

Bovenkerk et al. 2002:18 et seq. Though plants, too, cannot consent to physical violation, the authors do not include them in the class of entities whose integrity should be protected. Proponents of such an approach are Heeger and Brom 2001. For detailed elaboration of the neo-Aristotelian basis of such an approach to environmental ethics – without using the notion of integrity – see Korsgaard 2004; Kallhoff 2002; and Siep 2004. Concerning transgenic animals and plants for pharming purposes, it seems plausible that transgenesis can infringe the criteria for integrity – e.g., when a transgenic animal would produce a certain toxic pharmaceutical in its milk, so that this animal could not suckle its offspring. It is not plausible, however, that creating transgenic animals as such is necessarily violating integrity and, therefore, morally objectionable. Bovenkerk et al. 2002:20.

6.4 Risk assessment and risk-benefit analysis

193

– certain purposes such as remedies for life-threatening or childhood diseases, as well as cheaper pharmaceutical drugs for the population of less developed countries, tend to make them more acceptable. One differentiation of ends to which pharming techniques can be used is the therapeutic purpose. Without going much into detail, it seems plausible to assume that using pharming to establish a treatment for an otherwise untreatable and life-threatening condition will be judged morally less problematic than producing certain ingredients for, say, cosmetics in transgenic animals.41 Further points on a spectrum could be treatments for non lifethreatening diseases, and alternative, for example safer or more effective, treatments that might replace existing therapies. From a slightly different perspective one might ask why pharming is used as a production pathway instead of conventional methods. In some cases it may turn out that pharming is simply the cheapest method compared to other production pathways.42 Other things being equal, it might seem morally more problematic to harm animals in order to produce cheaper drugs than to use them when they are the only way to produce that drug.43 Analysing the situations in which and the purposes to which pharming is used certainly is time-consuming work. Nonetheless, this kind of casuistic reasoning is indispensable for a balanced evaluation of bioethical issues.44 A second recommendation drawn from the argumentation in this chapter is, therefore, that in evaluating pharming projects one should take into account the specific aims for which the animals and plants are used.

6.4 Risk assessment and risk-benefit analysis To secure their very existence and to improve their living conditions, since their very early history humans have intervened in the genome of many animal and plant species by using traditional breeding techniques. Even though some are of the opinion that any intervention of this kind is morally objectionable, it seems unreasonable to claim that the “direct” intervention in the genome of an organism, using modern gene technology, is to be judged as morally different from the traditional way by breeding through reproduction. Admittedly, one could imagine an ethical stance that judges any intervention in the genome of animals and plants using the methods of modern gene technology as principally immoral, because of the unnatu41 42

43 44

This is supported by the analysis in chapter 5. The economic superiority of pharming is due to the promising, but until now unproven, advantages of this technique compared to other production methods (see the introduction to this volume). One should be careful, however, not to draw the possibly wrong conclusion that cost savings are beneficial only for the drug producer. This is also consistent with the survey results presented in chapter 5 – at least in most countries.

194

6 The ethical evaluation of pharming

ralness of the technical procedures applied or the undesirable infringement of genetic integrity resulting from it.45 We assume, however, that the vast majority, even of those who would like to prohibit methods like pharming, would do so because they judge the possible consequences of such methods (and not the methods themselves) as the crucial moral problem. (In addition, proponents of naturalness and integrity approaches frequently object only to some and not necessarily to all genetic interventions in the genome of animals and plants.) At first sight it may seem that with modern gene technology only the methods used for intervening in the genome have been improved, without changing the purposes of the intervention. Together with the assumption that the relevant human aims – such as improving human living conditions – are frequently deemed to be morally acceptable, perhaps even morally obligatory, one might conclude that methods of breeding using gene technology are also morally acceptable or obligatory. However, gene technology allows some procedures that would not have been possible with classical breeding technologies as can be easily illustrated by pharming: one of its most convincing applications for pharming is that it makes possible the production of specifically human proteins in animals and plants by introducing the respective human gene into the animal’s genome. It is obvious that the introduction of a human gene into an animal’s or plant’s genome could not have been achieved by classical breeding. Even if we assume that generating transgenic animals or plants is morally acceptable in principle, we are confronted with the new situation that breaking down hitherto impermeable species-borders might create certain ecological risks not connected to classical breeding. What furthermore distinguishes pharming is the way pharming animals are produced.46 Though invasive reproductive procedures used in transgenic technology (for example hormonal treatments, embryo transfer) are increasingly common in food production too, and though pharming involves much lower numbers of animals and involves a particularly strong motivation to keep the animals healthy, it is still the case that the experimental and production phases of pharming animals create special risks to animal welfare (for example through aberrant transgene integration).47 Earlier in this chapter it was assumed that as long as our societies allow, in principle, the use of animals and plants for purposes such as food production and pleasure, they should also allow, in principle, the use of animals and plants for the improvement of human healthcare.

45 46 47

See sections 7.3.2 and 7.3.3. Pharming plants are, from a technical point of view, also generated differently from classical agricultural plants, but this is rarely seen as morally problematic. See chapter 5.

6.4 Risk assessment and risk-benefit analysis

195

This claim was based on the, at first sight, plausible further assumption that using animals for food production is indeed morally similar to using them for pharming purposes. Now, it has turned out that this second assumption is wrong. At least with respect to certain ecological risks and certain welfare concerns, animal and plant pharming are in a morally relevant way dissimilar to conventional means of using animals and plants for human purposes. This result does not imply that pharming is morally objectionable as such, but what it does imply is that specific applications of pharming should be subjected to a careful risk assessment. Over the past decades, our understanding of risk and the ways risk can be measured has considerably improved due to contributions from disciplines such as sociology, cognitive psychology, philosophy and law. It should be noted that – due to the different aims underlying risk research – there is not just one concept of ‘risk’ but many48. Firstly, framing ‘risk’ as a concept that shows many similarities with, for example, the concepts of ‘taboo’ and ‘sin’ makes plausible that ‘risk’ is not an objective but a social concept deeply entrenched in the network of rules shaping a society49. Secondly, a way into risk research is to analyse the factors that influence how individuals perceive and evaluate risks and how their own actions are influenced by risks50. Apart from helping to understand on what grounds individuals make risk decisions, such data may also be helpful in proposing recommendations for mastering moral conflicts. Thirdly, in a highly developed technological culture there is a need to regulate the development and application of risky procedures, for example new technological developments such as pharming. In the following we will focus on this regulatory approach to risk while acknowledging that this approach is not fixed nor, as some authors seem to believe, objective. A central part of this approach is the employment of scientific procedures to identify the possible damages or dangers that are connected to pharming and to establish what their likely occurrence is – a task commonly called risk assessment. The methodology of risk assessment has been developed considerably over the past decades and it is neither possible nor necessary to display more than the basics of this field of research here51. Risk is normally defined as the product of the extent of a specific damage and the likelihood that this damage will actually occur (risk = damage x probability). Tasks that have to be fulfilled in this context are: 1) hazard identification: what can go wrong and why? 2) Probability analysis: how often do the events go wrong? 3) Consequence analysis: how much damage is caused by the event? 4) Risk calculation 5) Uncertainty and signifi48 49 50 51

Yearl 2001. Douglas 1992. Slovic 2001. Gethmann in Gethmann and Klopefer 1993; Shrader-Frechette 1991.

196

6 The ethical evaluation of pharming

cance analysis: how sure are we of the risk estimate and how important is this type of risk?52 The first task in assessing the risk of pharming is, therefore, to identify possible hazards for humans, animals and their environment. Hazards for humans and the environment can be caused in a multitude of ways: for example by the pharming organism itself, its parts (for example pollen or excrement), or their vestiges (for example straw, animal corpses). The unintended effects can be generated by the pharming product itself (for example because of toxic effects on humans, fauna, flora, soil and water) or through dispersal of the transgene to other plants and animals (vertical or horizontal gene flow) with unclear consequences for individual organisms and ecosystems. The striking problem in identifying these hazards is that pharming plants and animals create new combinations of trait (transgene), organism, and environment, so that it is hardly possible to draw on existing experience. Even though, for example, the basic mechanisms of transgene dispersal will not differ between pharming plants and other plants, the fact that pharming products are intentionally produced at very high concentrations in pharming plants makes the assessment of the probability of possible changes much more complicated.53 With regard to animals, although animal pharming relies on healthy animals whose welfare is good, there are a number of risks to the animals used for pharming54. It is especially the experimental phase of making pharming animals that currently arouses concerns: harm generated through the reproduction techniques used, as well as consequences of aberrant transgene integration in the animal’s genome, can have severe side-effects for the founder animals as well as their offspring. Studies of welfare issues arising from making transgenic animals are still in their infancy. Even if it should be possible to come up with convincing measures for specific risks of pharming by careful case by case evaluation, how to evaluate this risk must be discussed. The vague phrase “How much risk is too much?” already shows that calculating a risk does not tell us whether we should take this risk or not: from calculating risks – and likewise calculating benefits – we do not learn how to balance risks versus benefits. Some authors have developed guidelines for balancing risks and benefits that seem quite plausible.55 However, convincing as such guidelines may seem at first sight, they are burdened with hardly solvable problems when one tries to make them work in practice: it has already been said that getting reliable data on risks may be problematic in some cases. Also, it might be a dif52 53 54 55

For a more detailed analysis see chapter on environmental risks. For a general discussion of the difficulties of performing risk assessment in genetically modified organisms see Myhr and Traavik 2002. See chapter 4. Gethmann (in Gethmann and Kloepfer 1993) proposes a ‘principle of pragmatic consistency’, Lassen et al. (2006:1003) advance a ‘principle of proportionality’.

6.4 Risk assessment and risk-benefit analysis

197

ficult task to determine (let alone measure) the expected benefits with high precision. The need to bring into relation qualitatively different risks such as ecological risks and animal suffering with equally varied benefits such as health care improvements, innovation, and a cut in production costs adds to the difficulties of employing risk-benefit analyses for practical decision making purposes. Finally, risks and benefits of a development like pharming are unequally distributed: those individuals or collectives at risk are not necessarily the same as those that benefit from that development. A conclusive risk-benefit analysis would, therefore, also have to achieve a sound balancing across collectives. The issue becomes ultimately muddled when one is confronted with risks and benefits of a technique that affects future generations.56 What consequences arise from these restrictions of risk assessment and risk-benefit analysis? It is likely that we will not be able to give detailed descriptions and calculations of all the relevant risks and benefits. For example, though we have identified the basic mechanisms of gene drift from pharming plants, we will not be able in the foreseeable future to make very precise calculations on how likely gene drift from all known pharming plants is and how severe the resulting damage will be. If, however, we lack the database for a scientific risk assessment, we cannot say much about risk. We can still believe that there is actually no risk, or we may believe that there is a risk about which we simply do not know yet. Strictly speaking, however, both beliefs stand outside the risk discourse. What is discussed is not risk but beliefs. In this situation it is not justified to speak either of belittlement of risks nor is it justified to speak of fear mongering – at least not with scientific rigour. Certainly, we will be able to assess some risks of pharming, especially in the area of animal welfare, with sufficient precision to base recommendations on them. In addition, we will be able to further develop our methods for doing risk-analyses. However, a conclusion of our considerations is that the alleged competition between scientists calculating so-called “objective” risks and laymen arguing with their beliefs – measured as “subjective” risks – cannot be easily decided in favour of scientists. This persistent uncertainty in performing a risk-benefit assessment of pharming and other biotechnological techniques has led to the development of the precautionary principle which is well embedded in the regulatory framework (see chapter 8). A third recommendation drawn from the argumentation in this chapter is, therefore, that in view of the difficulties in performing a systematic risk-benefit assessment of both animal and plant pharming there should be a careful case by case analysis of pharming projects using the precautionary principle. 56

The case examples in the appendix further illustrate the complexity a risk-benefit analysis would have to master.

198

6 The ethical evaluation of pharming

6.5 References Ali A, Cheng KM (1985) Early egg production in genetically blind (Rc+/rc) chickens in comparison with sighted (Rc+/rc) controls. Poultry Science 64:789–794 Attfield R (1998) The comprehensive ecology movement. In: Morscher E, Neumaier O, Simons P (eds) Applied ethics in a troubled world. Kluwer, Dordrecht, pp 9–27 Birnbacher D (2006) Natürlichkeit. De Gruyter, Berlin Birnbacher D (2007) Pharming – Ethical Aspects. In: Engelhard M, Hagen K, Thiele F (eds) Pharming. A new branch of biotechnology. Graue Reihe 43 (November 2007). Europäische Akademie, Bad Neuenahr-Ahrweiler, pp 65–72 Bovenkerk B, Brom FWA, van den Bergh BJ (2002) Brave new birds. The use of ‘animal integrity’ in animal ethics. Hastings Center Report 32(1):16–22 Brom FW (2000) The good life of creatures with dignity. Some comments on the Swiss expert opinion. Journal of Agricultural and Environmental Ethics 13:53–63 Charlesworth M (1993) Bioethics in a liberal society. Cambridge University Press, Cambridge Comstock G (2000) Vexing Nature. Springer, Heidelberg Comstock G (2002) Ethics and Genetically Modified Foods. In: Ruse M, Castle D (eds) Genetically Modified Foods. Prometheus Books, New York, pp 88–107 Douglas M (1992) Risk and Blame: Essays in Cultural Theory. Routledge, London Eaton M (2007) Commercial Pharming. Managing the challenges of drug manufacturing in plants and animals. In: Engelhard M, Hagen K, Thiele F (eds) Pharming. A new branch of biotechnology. Graue Reihe 43 (November 2007). Europäische Akademie, Bad Neuenahr-Ahrweiler, pp 31–64 EFSA AHAW (Animal Health and Animal Welfare) Panel (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 Frankena WK (21973) Ethics. Prentice Hall, Englewood Cliffs Gethmann CF (1992) Universelle praktische Geltungsansprüche. Zur philosophischen Bedeutung der kulturellen Genese moralischer Überzeugungen. In: Janich P (ed) Entwicklungen der methodischen Philosophie. Suhrkamp, Frankfurt, pp 148–175 Gethmann CF, Kloepfer M (1993) Handeln unter Risiko im Umweltstaat. Springer, Heidelberg Hare RM (1981) Moral Thinking. Its Levels, Method and Point. Clarendon, Oxford Hare RM (1997) Sorting Out Ethics. Clarendon Press, Oxford Heaf D, Wirz J (2002) Genetic Engineering and the Intrinsic Value and Integrity of Animals and Plants. Ifgene, Dornach Heeger FR, Brom FWA (2001) Beyond feeling well: our direct duties towards animals. In: Food safety, food quality and food ethics. Preprints 3rd Congress of Eursafe 2001, A&Q, Milan, pp 270–273 Hegselmann R (1998) What is Moral Philosophy and what is its Function? In: Morscher E, Neumaier O, Simons P (eds) Applied Ethics in a Troubled World. Kluwer, Dordrecht, pp 251–272 Kallhoff A (2002) Prinzipien der Pflanzenethik. Die Bewertung pflanzlichen Lebens in Biologie und Philosophie. Campus, Frankfurt Kass L (2000) The moral meaning of genetic technology. Human Life Review 26(1):76–87 Korsgaard CM (2004) Fellow Creatures: Kantian Ethics and Our Duties to Animals. In: Peterson GB (ed) The Tanner Lectures on Human Values Vol 25. University of Utah Press, pp 79–110

6.5 References

199

Krebs A (1999) Ethics of Nature. A Map. DeGruyter, Berlin Lassen J, Gjerris M, Sandøe P (2006) After Dolly – Ethical limits to the use of biotechnology on farm animals. Theriogenology 65:992–1004 Leopold A (1949) The Land Ethic. From: A Sand County Almanac. Reprinted in: Light A, Rolston H (2003) (eds) Environmental Ethics. An Anthology. Blackwell, Malden, pp 38–46 Light A, Rolston H (2003) (eds) Environmental Ethics. An Anthology. Blackwell, Malden Midgley M (2000) Biotechnology and Monstrosity. Why we should pay attention to the “yuk factor”. Hastings Center Report 30 no. 5:7–15 Mill JS (1998) Nature. In: Mill JS, Three essays on religion. Prometheus, Amherst, pp 3–65 Müller-Terpitz R (2007) The Use of Genetically Modified Animals for the Production of Pharmaceuticals – Legal Considerations from an Animal Welfare Perspective. In: Engelhard M, Hagen Km Thiele F (eds) Pharming. A new branch of biotechnology. Graue Reihe 43 (November 2007). Europäische Akademie, Bad Neuenahr-Ahrweiler, pp 73–86 Myhr AI, Traavik T (2002) The precautionary principle: scientific uncertainty and omitted research in the context of GMO use and release. Journal of agricultural and environmental ethics 15:73–86 Myskja BK (2006) The moral difference between intragenic and transgenic modification of plants. Journal of agricultural and environmental ethics 19:225–238 Nussbaum MC (2004) Beyond “Compassion and Humanity”. Justice for Nonhuman Animals. In: Sunstein CR, Nussbaum MC (eds) Animal Rights. Current Debates and New Directions. Oxford University Press, New York, pp 299–320 (Reprinted in Nussbaum MC (2006) Frontiers of Justice. Harvard University Press, Cambridge, pp 325–407) Räikkä J, Rossi K (2004) Bioethics and the moral significance of “gut feelings”. T Klin J Med Ethics, Law and History 12:79–82 Richter D (2007) Die Würde der Kreatur. Rechtvergleichende Betrachtungen. ZaöRV 67:321–349 Rollin B (2006) Animal Rights & Human Morality. Prometheus Books, New York Sandøe P, Forsman B, Hansen AK (1998) Transgenic Animals: The Need for Ethical Dialogue. In: van Zutphen LFM, van der Meer M (eds) Welfare Aspects of Transgenic Animals. Springer, Heidelberg, pp 90–101 Shrader-Frechette KS (1991) Risk and Rationality. Philosophical Foundations for Populist Reforms. University of California Press, Berkeley Siep L (2004) Konkrete Ethik. Grundlagen der Natur- und Kulturethik. Suhrkamp, Frankfurt Singer P (1975) Animal Liberation. Random House, New York Slovic P (2001) The Perception of Risk. Earthscan, London Sunstein CR, Nussbaum MC (2004) (eds) Animal Rights. Current Debates and New Directions. Oxford University Press, New York Thiele F (2004) Bioethics. Its foundation and its application in political decisionmaking. In: MacHamer P, Wolters G (eds) Science, Values and Objectivity. Pittsburgh University Press, Pittsburgh, pp 256–274 Tugendhat E (1993) Vorlesungen über Ethik. Suhrkamp, Frankfurt Vorstenbosch J (1993) The concept of integrity. Its significance for the ethical discussion on biotechnology and animals. Live Stock Production Science 36:109–112 de Vries R (2006) Genetic Engineering and the integrity of animals. Journal of Agricultural and Environmental Ethics 19:469–493 Yearley S (2001) Risk, Sociology and Politics of. In: Smelser NJ, Baltes PB (eds) International Encyclopaedia of the Social & Behavioural Sciences. Elsevier, Amsterdam, pp 13360–13364

7 The role of patents in the development of pharming

Many activities in the area of pharming are funded by private money. The greater part of this money is not invested for philanthropic reasons but because of a strong interest in an economic reward for providing venture capital. Since patents promise the patent-holder economic compensation, it is not surprising that many genes or coding regions (like promoter-regions), and many biotechnological procedures relevant for developing marketable biotechnological products, including biopharmaceuticals, are indeed protected by patents. The question to be addressed here is whether – and if so why – such patents are morally, legally, or economically questionable with respect to pharming.

7.1 The general justification of patents Patents are instruments of economic policy and are intended to promote innovations. This includes the safeguarding of material investments and the protection of intellectual property (for example inventions). In the latter case, the inventor shall receive commensurate reward for his services in providing useful benefits to the community in general – for example for the development of a novel pharmaceutical compound improving the provision of health care.1 The (ultimately moral) justification of patents is based upon the hope that by issuing patents it is possible to raise the general standard of prosperity in a society.2 It is for this reason that – at least in Europe – patents are issued only if both the novelty and the economic potential of the invention have been demonstrated in principle. An invention alone, without demonstrating its potential benefit for society, is not deemed to warrant special protection. For this purpose the patent system seems to have proven useful over time. This does not exclude, however, the possibility that it may be justified to raise objections against the issuing of a particular patent or even against awarding patents in certain areas like the biotechnology sector. The European patent law, which will be our main focus and which is discussed in detail below, is subordinate to national and European legal 1 2

For a general introduction see Poland 2000; Wreen 1998. See also Thiele 2003.

202

7 The role of patents in the development of pharming

provisions which regulate research and the marketing of research results. 3 Thus a patent alone does not authorize the owner of the patent to actually use the patented invention. This authorization is bound to other regulations, for example the gene technology law or the animal protection law. This means, therefore, that a patent is merely an excluding right; i.e. the owner of the patent may exclude a third party from using the content of the patent and is entitled to issue licences (for which fees may be charged).

7.2 The existing regulatory framework There are a variety of different sources of patent law that are or may be relevant in the field of pharming. At the international level: the Paris Convention on Industrial Property and the World Trade Organization agreement on Trade-Related Aspects of Industrial Property Rights (TRIPS). The Paris Convention does not establish substantive standards for granting patents, but assumes a broad scope of patent protection for the purpose of recognition of foreign patents, and in this context requires non-discrimination (national treatment) of foreigners. TRIPS obliges member states to protect inventions with respect to processes and products, prohibits discrimination and requires foreigners to be treated like nationals of the most-favoured nation. At the Pan-European level: the European Patent Convention of 1973, in force in Germany since 1976. (Its revision of 2000 entered into force in December 2007.) Under the convention, a European patent can be granted by the European Patent Office. This patent has the effects of a bundle of national patents. However, the convention also contains provisions on patentability that are relevant to pharming. At EU level: the Directive 98/44 on the legal protection of biotechnological inventions. This directive has been implemented under German law by a recent amendment of the German Patent Act. The project of establishing a single Common Patent for EU and EEA countries (Convention and later on EC Regulation on Common Patents) has, as yet, not been concluded because of controversies about the language of the patent claim as well as the introduction of software patents. At national level: the national patent acts, for example the German Patent Act.

3

The relevant legal text is: Directive 98/44/EC of the European Parliament and of the European Council of 6 July 1998 on the legal protection of biotechnological inventions.

7.3 Basic rules on patentability of biological products

203

7.3 Basic rules on patentability of biological products, biological material and biological and microbiological processes The relevant rules relating to the patentability of biological products, biological material, and biological and microbiological processes in the European Patent Convention (article 53 lit. b and section C-IV 2.a Guidelines of 2001), Directive 98/44 (articles 2 I, II, 3 I, II, 4, 5, and 8–11) and the German Patent Act (§§ 1 II, 1a, 2 II, 2a, 9a–9c) are by and large identical, although their interpretation has been the subject of quite some controversy and given rise to extensive litigation. However, it should be noted that Directive 98/44 is limited in scope and only covers biotechnological inventions. Modern law goes beyond the protection of technological inventions. Patents can also be granted when the object of an invention is a natural product or a product that consists of, or contains, biological material. The same is true of a process with which a natural product or biological material is produced or processed or in which it is used. Biological material is defined as to mean material that contains genetic information and reproduces itself, or can be reproduced in a biological system such as genes, gene sequences or fragments of gene sequences. Patentability does not necessarily depend on the “artificial” nature of a product. Substances, including biological material which previously occurred in nature, can be patented when they are isolated or produced from their natural environment by means of a technical process. The technical nature of isolation, or other production which can be repeated, marks the step from a mere discovery to an invention. Examples include the synthesis of a natural substance or the isolation of a gene sequence from a plant or an animal. The patentability extends, to a certain degree, also to human biological material or biological material that is identical to it. This is important because the use of human biological material in pharming is common. For moral reasons the legislator has ruled that the human body at all stages of formation and development, as well as the mere discovery of one of its elements, especially a gene sequence or fragment of a gene sequence, cannot be subject to patent protection. However, an element of the human body such as a gene sequence or fragment of a gene sequence isolated from it can be a patentable invention. The same is true of biological material otherwise produced by a technical process, for instance from an animal or a plant, but which is identical to human biological material. A further requirement of patentability of GMOs under Directive 98/44 and in conformity with it national law derived from the general prerequisites of patentability – identification, innovation and capability of industrial application – is that the claimed function and the industrial use of the gene sequence, or fragment of a gene sequence, must be described and thereby revealed in a concrete manner. As regards human biological material, Ger-

204

7 The role of patents in the development of pharming

man law goes a step further in limiting the patent claim on the product to the described industrial application (§ 1a IV Patent Act). Patents will not be granted for plant and animal varieties as well as essentially biological processes for the breeding of plants and animals. In the case of plant varieties, plant breeders’ rights are provided for. However, patents relating to plants or animals can be granted under two conditions. One is where the carrying out of the invention is not technically limited to a particular plant variety. This is true of the European Patent Convention4 as well as Directive 98/44 and national law implementing it. An example would be a method for producing herbicide resistance in plants belonging to the same species but constituting several plant varieties. A second condition where patent protection is provided is for inventions relating to plants or animals whose object is a microbiological or other technical process or a product obtained by means of such a process unless it relates to a plant or animal variety. While under the European Patent Convention the delimitation of biological and microbiological processes is largely open, Directive 98/44 and national law implementing the directive now provide for the relevant definitions. A biological process is defined as a process which consists entirely of natural phenomena such as cross fertilization or selective breeding. A process is of a microbiological nature when microbiological material is used, an intervention in such material is carried out or such material is produced. The legal situation in the United States is less clear. American law does not make a distinction between discovery and invention, but limits patent protection by the requirement that the discovery or invention must relate to a new useful process, machine, manufacture, or composition of matter or improvement of them (section 101 Patent Act). In a leading case decided in 19805 the US Supreme Court, departing from the proposition that not every discovery in the field of nature is patentable, considered the modification (“engineering”) of bacteria so that they could break down and metabolize crude oil to be a patentable invention. Patent protection is also provided for transgenic plants and seeds. Although there seems to be no case law in point, it may be assumed that patent protection also extends to genes, gene sequences and fragments of gene sequences, at least where substantial new utility can be claimed. Taken in conjunction, this means that under modern European and US patent law there is a far-reaching industrial property protection for biotechnological processes as well as products and material obtained through such processes, including products and material that have previously occurred naturally or are derived from the human body. This protection covers proc4 5

Enlarged Board of Appeal of the European Patent Office, decision of 20.12.1999, Transgenic plant/Novartis II, Official Journal 2000:111. US Supreme Court, decision of 16 June 1980, Diamond v. Chakrbaty [1980] 65 Lawyers’ Edition 2nd 150.

7.4 Extent of protection

205

esses for the genetic modification of plants and animals used for the production of pharmaceuticals, the genetically modified plants and animals themselves as well as genes, gene sequences or fragments of gene sequences isolated from them or inserted into them.

7.4 Extent of protection The patent owner has the exclusive right to produce, market and use the patented product or use the patented process, especially for production. This is also true of GMO patents. However, there are some special rules that amount to an extension of protection. In case of biological material or a process for obtaining such material, this protection extends to propagation and multiplication of the patented biological material and the material derived by means of the patented process including its propagation and multiplication. However, the usual farmer’s privilege is maintained and at least under German law (§ 9c III Patent Act) the farmer is also protected in case of adventitious presence and use of GMOs in the cultivation of nontransgenic crops. Where, due to an invention, a patented product consists of, or contains, genetic information, the protection extends horizontally to another material into which the product has been inserted and which contains the information and fulfils its function. Nevertheless, modern patent law also contains some limitations of protection of transgenic inventions that are designed to enable future innovations in the field. One is the requirement for patentability of gene sequences and fragments of gene sequences whereby their claimed function and industrial use must be described. The consequences of this requirement for the extent of patent protection are not quite clear. The official German interpretation of directive 98/446 is that an “absolute” product patent can be granted whose scope, however, is limited to the fragment of a gene sequence that is relevant for the claimed function. Within this frame, the patent owner enjoys product protection. The recognition of a – limited – product patent would manifest itself where a second inventor developed an alternative use for the gene fragment. This development would constitute a dependent invention which can only be used with the consent of the holder of the product patent. In legal literature, it is frequently sustained that the grant of an absolute product patent for gene sequences, however limited its scope may be, is not consistent with Directive 98/44 and constitutes an unjustified degree of protection that hampers innovation.7 However, the German legislature was of the opinion that the inventive step normally consists in the proof of a new 6

7

Federal Government, Exposition of reasons for the draft, Bundestags-Drucksache 15/1709:11; Report of the Parliamentary Committee on Economic Affairs, Bundestags-Drucksache 15/4417:8. See Kraßer 2004:227 et seq.

206

7 The role of patents in the development of pharming

industrial application, rather than in the general determination of a particular function, and hence the patent claim would be voluntarily limited by the inventor. In any case, those parts of the gene that are not relevant for the claimed function are not covered by patent protection. By this technique the law tries to reduce overlaps8 and ensure that, apart from the research exception (experimentation defence) discussed below, research on, and industrial use of, the whole gene sequence is not severely hampered. As regards human biological material, patent protection is clearly narrower. It is not only the function of the fragment of the gene sequence but also the disclosed industrial application that limits the scope of protection. This is also true where the gene sequence has not been isolated from the human body but from an animal or a plant. The legal consequence of this limitation of protection is that a new application does not constitute an invention that is dependent on the product patent.

7.5 Mandatory licenses When the exercise of an invention depends on a prior patent owned by a third party, the inventor may find himself in a situation where a commercial use of the invention is impossible. Patent law has long recognized a need for mandatory licences, although under very restrictive conditions. The applicable rules are by and large based on national law. Neither the European Patent Convention nor Directive 98/44 contain pertinent legal provisions. Rather, they leave the field to national law. However, the TRIPS Agreement imposes some restraints on mandatory licensing. Traditionally, a mandatory licence could only be granted where the exercise of an invention was dependent on a previous patent, the inventor was unable to secure a licence under appropriate conditions and the grant of a mandatory licence was in the public interest. This approach is however deeply problematic, since (i) the definition of “public interest” is wide open to interpretation, and (ii) trust in the reliability of governmental economical policy may be seriously weakened.9 In the more recent legal development, the prerequisite of a paramount private interest has replaced that of a public interest. Under § 24 II German Patent Act, the grant of a mandatory licence requires that the invention, in comparison with the patented invention whose use is claimed, constitute an important technical progress of considerable economic importance. Although the shift from the public to a paramount private interest lowers the hurdles of mandatory licens8 9

Overlaps of protection due to parallel claims regarding the same gene sequence. Compulsory licences may however work in some cases: e.g. compulsory licences for AIDS drugs in Brazil, where the failure of the patent system (from the perspective of the Brazilian government primarily looking after the welfare of the Brazilian population) was demonstrable in a single, clearly definable case.

7.6 Patents as obstacles to innovation in pharming?

207

ing, it is safe to say that the new prerequisites are still very demanding and limit mandatory licences to exceptional cases. Particularly in the initial stage of innovation it will be difficult to prove the economic importance of the invention, since it depends on a variety of different factors including acceptance of the new product on the market.

7.6 Patents as obstacles to innovation in pharming? We will take it for granted that the promotion of innovation in the area of pharming is morally recommendable in principle, since we can reasonably expect that innovation in this sector can be used by and contribute to the benefit of humanity. In recent years significant investments in biotechnology, to a large extent from private sources, have led to a rapid increase in research activities in this field. Since these investments were made in anticipation of future profits, investors are highly interested in safeguarding their investments. An instrument to this end is, as has been mentioned previously, the application for a patent in order to safeguard the future utilization of, and benefits to be gained from, research results. It is therefore reasonable to assume that a large proportion of biotechnological research, progress, and innovation – including the area of pharming – would not exist if investors had not been able to protect their investments by patents. It should be remembered, however, that the primary aim in issuing patents is not to help patent-holders earn money, but to increase a society’s welfare by promoting innovation. Issuing patents usually, but not necessarily, is the best way to achieve this goal. Indeed, there are some indications that the existing patent system might become a veritable obstacle to the innovation process in the biopharmaceutical and biotechnological industry.10 Central to such considerations is the worry that so called strategic or key patents are concentrated in the hands of a few patent-holders that thereby may exert far-reaching control over the marketing of a broad range of pharmaceutical and biotechnology products. In this situation it may become unattractive to sponsor the development of a new biotechnology – product or pharmaceutical – even though research itself is not affected by patent rights at all. Unfortunately, those statutory precautions implemented in patent regulation for ensuring the possibility of future innovation are not very relevant in the field of pharming. Firstly, it is safe to say that process patents such as patents that cover particular methods of insertion or nuclear fusion must be respected under any circumstances, although in practice licences can normally be obtained. Secondly, the limitations relating to transgenic product patents are not really geared to pharming practices. It is exactly the function and industrial application of a gene sequence or fragment of a gene sequence isolated by the patent holder which is essential for further 10

See also Cockburn 2004; Heller and Eisenberg 1998.

208

7 The role of patents in the development of pharming

development in pharming. The idea of the legislature that future development would occur “around” the protected function and industrial application is not realistic in practice. Another safeguard of future innovation is provided by the “research exception” (“experimentation defence”; article 31 lit. b European Patent Agreement, § 11 No. 2 German Patent Act). Under this exception, the protection of the patent does not extend to activities for test purposes relating to the object of the patented invention. The extent of this exception is controversial. This concerns in particular the question relevant to pharming as to whether the exception goes beyond purely scientific research and may also cover research that is undertaken for commercial purposes. Under the traditional interpretation, the exception only covers tests for scientific purposes which aim at determining the capability and manner of functioning of an invention. This relates in particular to tests for proving the lack of an inventive activity or the pre-publication of the invention, as well as tests which are necessary for the development of science and technology. Tests outside the area of research that already mark the beginning of commercial use, for example tests used in authorization procedures, are considered to violate the patent.11 In the United States, too, the experimentation defence is in principle limited to non-commercial purposes. However, in Germany a more liberal interpretation of the research exception is gaining ground. The Federal Supreme Court, in a judgement of 11 July 199712 concerning national and European patents for a transgenic polypeptide with interferon-gamma properties, declared clinical tests carried out by third parties to be covered by the research exception when they served to find out whether, and in which form, the active ingredient was capable of curing or mitigating particular human illnesses. Relying on the purpose of the research exception under § 11 No. 2 German Patent Act of 1981 – patent protection as a reward for, and promotion of, scientific and technical innovation, but in view of freedom of research and the social obligation of property no obstruction of the innovative process – as well as its legislative history, the Federal Supreme Court discarded the traditional distinction between scientific and commercial purposes of tests. In its view, the only requirement was whether or not the tests relate to the object of the patented invention, and this was held to cover, beyond the mere substance, the claimed technical precept including the industrial use of the substance. The court held all tests to be encompassed by the research exception that 11

12

Dutch Supreme Court, decision of 18 December 1992, Gewerblicher Rechtsschutz und Urheberrecht, Internationaler Teil (GRUR Int.) 1993:887; Dutch Court Den Haag, decision of 3 February 1994, GRUR Int. 1995:53; British Court of Appeal, decision of 11 June 1985, GRUR Int. 1987:108; formerly also German Federal Supreme Court, judgement of 21 February 1989, Bundesgerichtshof in Zivilsachen (BGHZ) 107:46. BGHZ 130:259.

7.6 Patents as obstacles to innovation in pharming?

209

served to generate knowledge about the substance, including its utilization and potential for further development. Therefore, tests carried out in order to determine the effects of the substance or find new, previously unknown applications were considered to be legal, irrespective of whether their ultimate goal was purely scientific or commercial. By contrast, according to the opinion of the court a protected invention may not be used in the framework of a test relating to another object, i.e. where the invention is only the means of a testing activity. What that means exactly is not clear. In legal literature, instrumental uses are described as tests relating to the market success of the product.13 The court considered the interests of the patent holder as sufficiently protected because, when a second party secured a patent for a new application of a patented substance, this use patent is dependent on the substance patent and can only be exercised with the consent of the holder of the substance patent. The safeguard of dependency may not fully operate in the field of biological material when the original patent claim is (and needs to be) limited to biological material in conjunction with its claimed function and industrial application, in other words the first patent holder does not have a patent claim on the biological material as such.14 Despite the evident theoretical advantages the recent German Supreme Court’s construction of the experimentation clause offers for research and development in general, its consequences in the field of pharming appear to be slight. The reason for this is similar to that sustained with respect to the limitation of patent protection for biological material. There is less commercial interest in finding and developing new, previously unknown applications of a gene sequence or fragment of a gene sequence but, rather, in using it in its revealed application for the production of a pharmaceutical. Unfortunately, it is an exceedingly difficult task to establish exactly to what extent specific patents can or do restrict innovation processes in pharming. Nonetheless, some general remarks can be made: – Very early in establishing a pharming platform transgenic plants and animals carrying the relevant gene have to be generated. The methods of creating transgenic organisms, for example transfection and cloning techniques can and have been subject to patent protection. Insofar as there are not many procedures known for the successful and reliable generation of transgenic organisms, patents on those procedures may be the limiting factor in developing new pharming products (especially in the light of the unclear status of the research exemption). 13 14

Kraßer 2004:813. See e.g., Kirin Amgen Inc. v. Transkaryotic Therapies Inc, House of Lords, decision of 21 October 2004, [2004] House of Lords 46, where a simple process patent was in litigation.

210

7 The role of patents in the development of pharming

– The most likely production path in animal pharming will be the expression of pharming products in the milk of cows, goats, etc. To achieve expression of those products in the milk, the coding gene sequence has to be attached to a gene-promoter determining the expression site. Similarly, in plant pharming one will like to control the expression site (leaves, crop, etc.) of the transgene by using a specific promoter. Clearly the number of suitable promoter-regions for this task is limited so that patents on them may severely restrict possibilities to establish pharming animals. Though some key patents for pharming have already expired or are due to expire,, we recommend that this issue is observed closely – not, as has been stated previously, because of fundamental objections against the patent in the biotechnology sector but because intellectual property restrictions are likely to be very important in the short to mid term. Practices such as the granting of excessive and protectionist claims by national governments are a significant negative influence that stifle innovation.

7.7 References

211

7.7 References Cockburn IM (2004) The Changing Structure of the Pharmaceutical Industry. Health Affairs, 23(1):10–22 Heller MA, Eisenberg RS (1998) Can Patents Deter Innovation? The Anticommons in Biomedical Research, Science 280:698–701 Kraßer R (2004) Patentrecht. 5th edition. CH Beck, München McGee G (1998) Gene Patents Can Be Ethical. Cambridge Quarterly of Healthcare Ethics 7:417–421 Poland SC (2000) Genes, Patents and Bioethics – Will History Repeat Itself? Kennedy Institute of Ethics Journal 10(3):265–281 Thiele F (2003) Zur moralischen Bewertung der Patentierung von Genen. In: Steigleder K, Düwell M (eds) Bioethik – Eine Einführung. Suhrkamp, Frankfurt, pp 388–396 Wreen M (1998) “Patents”. In: Encyclopedia of Applied Ethics. Academic Press, San Diego 3:435–447

8 Legal problems of pharming

8.1 Introduction The regulatory framework of pharming is highly complex. As regards the sources of regulation, one must distinguish between European and national levels. The various phases of pharming, from the initial development to the authorization procedure to final manufacturing, are partly subject to different legal rules. In addition, the various regulatory issues at the relevant stages are addressed by different pieces of regulation. The following text is organized so as to follow the “life cycle” of pharming products. When appropriate, the phases are further divided into segments according to the regulatory issues arising and the type of regulation concerned. The following overview provides an initial orientation.

8.2 Development phase I: Protection from risks to the environment caused by the use and release of GMOs As regards risks to the environment caused by the use and release of GMOs in the development phase, the fundamental question concerns the necessary safety requirements in order to prevent harm to the environment from the relevant transgenic plants or animals. Alternative strict safety requirements exist under the contained use regime established by the Directives 90/219 and 98/8 and the more liberal precautions under the release regime laid down by Directive 2001/18.

8.2.1 Sources of legal regulation and their scope of application In this respect, two major EC directives1 may be relevant: – Directive 90/219 on the contained use of genetically modified microorganisms (contained use directive) as amended by Directive 98/81 (which has replaced almost entirely the former directive)2, – Directive 2001/18 on the deliberate release into the environment of genetically modified organisms (release directive). 1 2

All the more recent EC directives and regulations can be accessed free of charge through www.eur-lex.europa.eu. Except for articles 1 and 17. The original directive is published in O.J 1990 No. L 117:1.

214

8 Legal problems of pharming

Table 8.1: Overview of the legal regulations and institutions involved in

pharming

1. Development phase Risks to the environment Directives 90/219 + 98/8 on contained use or Directive 2001/18 on release of GMOs, Part B and national law on biotechnology Waste: in addition Regulation 1774/2002 on disposal of animal parts and national law Commission guidelines (releases) P/S: National authorities P: Commission (release in simplified procedure) Occupational safety and health Directives 90/219 + 98/8 on contained use or Directive 2001/18 on release of GMOs (conditions of permit) and national law on biotechnology Directive 2000/56 on biohazards and national law S: National authorities

Animal welfare Directive 86/609 and national law on anima1 protection

P/S: National authorities Quality and safety of developmental product; risks from the environment; ethical justification of clinical testing Regulation 726/2004 + Directive 2001/83 on pharmaceuticals, Directive 2001/20 (ethical review) and national law on pharmaceuticals (production and clinical testing) P: National authorities (production and clinical testing); S: National authorities

2. Marketing authorization phase Quality, safety and effectiveness of medicinal product (including assurance of safety of production) Regulation 726/2004 + Directive 2001/83 on pharmaceuticals and national law (production) EMEA guidelines P: Commission/EMEA, National authorities (production) 3. Production phase Risks to the environment Animal welfare Directives 901219 + 98/8 on contained use National law or Directive 2001/18 on release of GMOs, Part C and national law on biotechnology P: National authorities (contained use) or S: National authorities Commission/EFSA (cultivation in the open environment); S: National authorities Occupational safety and health Coexistence (impairment of economic interests through production) Directives 90/219 + 98/8 on contained use EC recommendation on coexistence or Directive 2001/18 on release of GMOs National law (regulation, labelling and (conditions of pernit) and nationallaw Di- liability) rective 2000/56 on biohazards and national law S: National authorities S: National authorities Quality and safety of medicinal product; risks from the environment Regulation 726/2004 + Directive 2001/83 on pharmaceuticals and national law (production) EMEA guidelines P: National authorities (production); S: National authorities

P: permit/authorization; S: surveillance/inspections; for other abbreviations see list of abbreviations.

8.2 Development phase I: Protection from risks to the environment

215

These directives need to be, and have been, implemented and partly supplemented by national law so that the relevant national laws3 are applicable “on the ground”. These are in: – Germany: the Act on Biotechnology of 1990, consolidated in 1993, as last amended in 2008, and various regulations promulgated under the act, in particular the Regulation on Biotechnological Safety of 1990, consolidated in 1996, as amended (contained use) and the Regulation on Biotechnology Procedure of 1990, consolidated in 1996, as amended (contained use and releases); – United Kingdom: the Environmental Protection Act 1990, part VI, the Genetically Modified Organisms (Contained Use) Regulations 2000 and the Genetically Modified Organisms (Deliberate Release) Regulations 2002; – France: Code de l’environnement, Book V Title 3 (legislative and regulatory parts), in particular articles L532-1 to L532-6, R515-32 to R515-36, R532-1 to R532-l7 (contained use), articles L533-1 to L533-7, R533-1 to R533-48 (releases). Directive 90/219 is an environmental directive and therefore only contains a minimum standard (article 176 EC Treaty, formerly article 130t EC Treaty). In contrast, Directive 2001/18 is based on the legislative competence of the Community to harmonize national law. A national deviation from such a directive is subject to severe restraints (article 95 [4], [5], formerly article 100a [4] EC Treaty). Since both directives are very detailed, the degree of divergence between the relevant national laws is not very large. Therefore, for reasons of convenience, the following analysis focuses on the two directives, and discusses national law implementing them only where suggested by particularities. The scope of application of the two directives is essentially equal as regards the type of activities covered. They comprise the cultivation, keeping, propagation, storage, destruction, disposal, discharge, transport and other use or handling of GMOs. However, there are two differences. One is that Directive 90/219 only applies to genetically modified microorganisms (GMMs), while Directive 2001/18 comprises all genetically modified organisms. Microorganisms are defined as microbiological entities capable of replication or of transferring genetic material, including viruses, viroids, 3

Unless the exact source is indicated, access to the national laws and regulations cited here is possible free of charge at the following internet addresses: Germany: www.bmu.de (July 2008) or www.gesetze-im-internet.de (July 2008); United Kingdom: www.opsi.gov.uk (July 2008); France: www.legifrance.gouv. fr. (July 2008); United States: www.law.cornell.edu/uscode (July 2008) or www. gpoaccess.gov/uscode/index.html (July 2008) and (as for regulations) www. gpoaccess.gov./ecfr (July 2008). Many member state texts can also be found in an English translation in: European Commission, Technical Regulations Information System (TRIS) at www.ec.europa.eu/enterprise/tris (July 2008).

216

8 Legal problems of pharming

and animal and plant cells in culture.4 “Naked” DNA, r-plasmids and cell nuclei are not encompassed by the definition of both microorganism and organism. In the context of pharming, the distinction between microorganism and organism is not very important at the initial development stage, due to the frequent use of cell cultures as well as virus-based DNA sequences as promoters. However, the subsequent stages of development, especially the handling of GM plants and animals after genetic modification, are no longer covered by directive 90/219. Since the scope of application of the national laws and regulations on contained use extends to all GMOs, especially transgenic animals and plants, this restriction does not mean that there are no rules on contained use in place. The second and fundamental difference between the two directives concerns the kind of handling. While Directive 90/219 and implementing national laws apply to the contained use of GMMs, releases of GMOs into the environment, including placing them on the market, are covered by Directive 2001/18. From an environmental perspective, the operator can in principle choose between the two regimes. Development with containment under Directive 90/219 and national laws on contained use can be carried out in order to secure an authorization for the finished pharmaceutical product under Regulation 726/2004. This is especially attractive in animal pharming because the number of animals needed for development is limited. Even in plant pharming, development with containment, for example in glass-houses, would make sense. Alternatively, the operator can start development operations without containment under the regime of the release directive, and then initiate the authorization procedure for the medicinal product derived from the transgenic plants or animals. Finally, the operator can divide the development process into two phases, first using the contained use directive and later on switching over to the release directive. The operator might consider it useful to gain experience with a release, especially if he/she wants to perform the later production in an open system under part C of Directive 2001/18, or wants to introduce the transgenic plant as seed (see section 8.7). Directive 2001/18 is based on the “step by step” principle of continuous generation of new knowledge, whereby containment of GMOs is gradually reduced and the scale of release increased if the evaluation of the earlier steps, in terms of human health and the environment, indicates that the next step can be taken (Directive 2001/18, recital 24).5 Although the step-sequence concept is not strictly mandatory,6 it may serve as a guideline for the interpretation of the directive. Therefore, it may appear reasonable to follow this line in devising the stages of the development process for pharming drugs. 4 5 6

The definition intentionally goes beyond the common scientific understanding of a microorganism. See Winter et al. 1993:49 et seq. Administrative Court Berlin, in: Ebersbach/Lange/Ronellenfitsch 1995 vol. 4, § 16 GenTG No. 7.

8.2 Development phase I: Protection from risks to the environment

217

However, Regulation 726/2004 in conjunction with Directive 2001/83 and the applicable requirements relating to the quality of development pharmaceuticals and the production process for such pharmaceuticals may limit these options. As will be discussed in more detail later (see section 8.5), in contrast to medicinal products that contain, or consist of, GMOs, medicinal products that are merely derived by using recombinant DNA techniques such as the normal pharming drugs need only to undergo the normal authorization procedure under Regulation 726/2004. An environmental risk assessment relating to a release is not required. From this perspective, a release would not be necessary but, of course, possible. However, under pharmaceuticals regulation the production process for pharmaceuticals under development (“developmental pharmaceuticals”) requires a considerable degree of protection from risks from the environment in order to ensure a sufficient quality and safety of the product to be submitted for testing, especially clinical testing. This may limit or even exclude the use of the release regime, at least with respect to animal pharming (see sections 8.4, 8.5.3). In the United States, from an environmental perspective the development of transgenic medicinal products can be performed either as a contained use or as a release of GMOs. Under the sectoral approach that the US follows in this field, contained uses and releases of GMOs are regulated according to products rather than processes. The Toxic Substances Control Act (TSCA; 15 U.S.C. §§ 2601–2692) only applies to genetically modified microorganisms, which are considered by the Environmental Protection Agency as “new substances” in the meaning of section 3 [2] [a] [i] TSCA. It is not applicable to transgenic plants and animals. The contained use and release of transgenic plants are regulated by the US Department of Agriculture under the Plant Protection Act (7 U.S.C. §§ 7701–7758) and implementing regulations (7 C.F.R part 340), provided the GMO constitutes a plant pest. A plant pest includes bacteria, fungi, viruses or similar organisms, as well as infectious agents or substances that may cause damage to plants (7 C.F.R § 340.1). Although the regulations assume that introduced genetic material that encodes products intended for pharmaceutical use may be covered (7 C.F.R § 340.3 [b] [4] [iii]), pharmaceutically active agents produced in pharming activities, for example vaccines or antibodies, as such are not encompassed by this definition. Rather, it is necessary that the donor or recipient organism, or the vector or vector organism constitutes a plant pest, which may be the case where gene fragments from pathogens, viruses or bacteria are used as promoters.7 The regulatory activities of the US Department of Agriculture (Animal and Plant Health Inspection Service – APHIS) relating to GMOs focus on the release of transgenic plants used for the production of food and 7

See USDA, 65 Fed. Reg. 53976 (2000); Anderson et al. 2001:3,4,15–24; Steines 2002:147.

218

8 Legal problems of pharming

feed. However, transgenic plants for pharmaceutical and industrial use are also regulated. There is as yet no parallel regulation of transgenic animals. However, powers to regulate exist under the Animal Health Protection Act (7 U.S.C. §§ 8301–8317), and the US Department of Agriculture is considering regulating transgenic animals in the future.8 As in Europe, the regulation of medicinal products under the Federal Food, Drug and Cosmetic Act (21 U.S.C. §§ 301–399), administered by the Federal Food and Drug Administration (Center for Biologics Evaluation and Research – CBER – and Center for Veterinary Medicine – CVM), also addresses safety and quality issues of products under development derived from transgenic plants or animals. The agency is also competent to regulate risks to the environment that may be associated with the development process.

8.2.2 Development of recombinant medicinal products with containment Directive 90/219, which may be applicable to the initial development of recombinant medicinal products with containment, aims for the protection of human health and the environment from risks arising from activities that engender the use of GMMs and hence also initial pharming operations. Although the formulation of the relevant article 1 of the directive, even after the fundamental reform of 1998, does not contain any language to the extent that it could be an expression of precautionary elements, the directive has to be interpreted in conformity with the precautionary principle laid down in article 174 [2] EC Treaty. This is not only suggested by the jurisprudence of the European Court of Justice,9 but also by the recitals of the directive (recitals 1 and 3) and its article 5, which requires the member states to avoid adverse effects on human health and the environment which might arise from the contained use of GMMs. Moreover, as can be derived from the definition of accident, the directive aims to minimize the risk of escape of GMMs into the environment as well as exposure of workers. The definition of accident that has to be prevented refers to mere hazards to human health and the environment (article 2 [d]). In view of the uncertain nature of risks presented by GMMs and the resulting impossibility of a definitive risk assessment, this application of the precautionary principle is indispensable. Directive 90/219 employs a system of four risk classifications (containment categories) ranging from no or negligible to high risks, which trigger containment measures of increasing stringency.10 The attribution of a par8 9

10

Based on 7 U.S.C. § 8303; see 72 Fed. Reg. 69762-63 (2007). Animal cloning is already regulated; see 73 Fed. Reg. 2923 (2008). 1998 ECR I-2265 No. 90 – BSE; 2006 ECR I 53 No. 40 – Commission v. Germany; see also European Court of First Instance, 2002 ECR II-3305 Nos. 135 et seq. – Pfizer Animal Health. Based on American National Institute of Health guidelines; see Marx 1997:295 et seq.; OECD 1986:34 et seq.

8.2 Development phase I: Protection from risks to the environment

219

ticular activity to one of the risk categories must be based on a risk assessment to be performed by the operator, using the criteria contained in annex III (article 5 [2], [3]), and notified to the competent authority. The factors to be considered concern the risk potential presented by the recipient organism, the genetically modified insertion, the vector, the donor organism and the resulting GMM, the characterization of the activity and the severity and likelihood of harm. Classification is the responsibility of the operator. As a safeguard against under-classification by the operator, the directive (article 4 [4]) provides that, in case of doubt, the more stringent category shall be applied unless sufficient evidence, in agreement with the competent authority, justifies the application of the less stringent measures. The competent authority has a margin of discretion in attributing a particular use to a risk class.11 The directive prescribes a consent procedure when premises are used for contained use of GMMs (maximum delay: 90 days; article 10). This requirement is limited to the risk classes 3 and 4. In other cases, a notification is sufficient. Articles 8 and 9, in conjunction with article 11 [2], [3] of the directive, mandate the member states to grant their authorities sufficient powers to review all notifications, the correctness of the risk assessment including the classification of the activity in question, and the suitability of the containment and protection measures proposed by the operator. The operator may also be obliged to set forth an emergency plan for the case of an escape of GMMs falling under the higher risk categories. As stated, national law relating to contained use extends to all GMOs. It more clearly spells out the precautionary principle as a basic standard of protection, while following relatively closely the risk classification and duties of care established by the directive. In setting permit and notification requirements, Germany and the United Kingdom distinguish between the facility in which a contained use is to be carried out (facility accreditation) and the process as such (contained use). Moreover, the degree to which national law also adheres to the procedural design of the directive varies. The most liberal regime exists in Germany. Here, the construction and operation of a facility for use of classes 1 and 2 GMOs and the initial use of such GMOs are only subject to a notification procedure (coupled with alleviations regarding repeated uses). A facility and use permit is required for the higher risk categories. In the United Kingdom, facilities for the use of GMMs of all risk categories need to be notified. With respect to the use of class 1 GMMS, a further notification is not necessary. The initial use of all other risk classes requires a consent. A further differentiation between the relevant risk classes is made with respect to repeated uses. With respect to GMOs other than GMMs, the regulations provide that a notification and 11

Administrative Court of Appeal Mannheim, Neue Zeitschrift für Verwaltungsrecht 2002:224.

220

8 Legal problems of pharming

consent are not required where the potential risk is equivalent to that of non-transgenic organisms. In France, initial and repeated uses of GMOs classified into all risk categories are subject to the requirement of a consent, while some procedural alleviations are provided for repeated uses. Normally, in all countries there are relatively short delays for reacting to a notification or granting a consent. In practice, pharming activities are classified under the risk category 1. This is based on the assumption that the risk associated with the development operations is negligible. However, as recognized by annex III point A and B 4, risk is a function of hazard, probability of exposure and harm, and kind and extent of assumed harm. The requisite probability is relative to the kind and extent of harm. Therefore, one may put the generality of this classification practice into question, especially in view of possible accidents. One of the criteria, among others, that may be considered when assessing risk is the biological activity of the GMOs (see section 3.7). Directive 2001/18 seeks to phase out the use of antibiotic resistant marker genes (article 4 [2]). This is predicated on a normative judgement, whereby substances that have pharmaceutical properties associated with potential deleterious effects on the possibility of treating a disease should not enter the environment through release of GMOs, even if the probability of a (horizontal) gene transfer associated with the release is considered to be very low. Without an express political decision laid down in the contained use directive or national law, this phasing-out policy cannot be transferred to other areas and substances. However, it calls for a higher degree of caution. With respect to activities of all categories, the directive (article 6 [1], annex IV Point 1) and implementing and supplementary national regulations12 in the first place require a minimum of containment and protection measures of a general nature (so called principles, in Germany: basic obligations). In this connection it must be noted that, since the amendment of the directive in 1998, the notion of containment has become more fluid and its delimitation from mere confinement used under the release regime more difficult. Under the new article 1 [1] [c] of the directive, containment measures are not only physical barriers, they can also be a combination of physical barriers with chemical and/or biological barriers. The notion of containment requires that the relevant measures are capable of limiting the contact with, and providing a high level of safety for, the general population and the environment. Among others, the “principles” require that the environmental exposure must be kept to the lowest practicable level, control measures and equipment must be checked and maintained adequately, which includes 12

Germany: Section 6 [2] Act on Biotechnology, section 8 [2], [5] Regulation on Biotechnological Safety; United Kingdom: Section 106 [4] Environmental Protection Act 1990, section 17 [1] Genetically Modified Organisms (Contained Use) Regulations 2002; France: Articles R532-1 to 532-17, R515-32 to 515-36 Code de l’environnement.

8.2 Development phase I: Protection from risks to the environment

221

the prevention of accidents, and, when necessary, testing for the presence of viable process organisms outside the primary containment must be undertaken. It may be assumed that the interpretation of the broad legal terms “practicable” and “adequate” varies from country to country, in particular regarding the requirement of proportionality to risk and the degree to which costs incurred are considered.13 In Germany, reduction of exposure even to the lowest possible level is required, which, however, does not rule out the application of the principle of proportionality.14 In the United Kingdom, the concept of best available technology not entailing excessive costs is applied. In addition, the directive provides for special measures whose stringency depends on the risk class. Regarding category 1 activities, very few mandatory special containment and protection measures are prescribed. Rather, annex IV lists some specific measures of an “optional” character, especially for activities in animal units and for activities outside glass-houses and animal units. These measures concern viable microorganisms, bulk culture fluids, possible spillage, air pollution and final discharges of waste. “Optional” means that the operator may apply these measures on a case by case basis, subject to the risk assessment. The practical legal consequence of this somewhat cryptic formulation is that, depending on the risk assessment, such measures may be mandatory by virtue of the general obligation of care set forth by article 6 [1] of the directive and specified by the general principles under annex IV point 1. National law normally follows very closely the annex IV but extends it, as stated, to all GMOs. Sometimes, the safety standard has also been specified. This is, for instance, true in Germany with respect to the keeping of transgenic animals. It is permissible to keep them outside with double fencing where there is no risk of horizontal gene transfer, measures for preventing theft are taken and a warning system in case of escape is in place.15 In contrast, it would seem that open air cultivation of transgenic plants is not possible under the contained use regime because the contact of GMOs with the environment cannot be excluded. The basic competences for applying national law implementing the directive are vested in the member states. Institutional arrangements vary. In Germany, the regional authorities are competent, but have to consult a central expert body (Central Commission for Biosafety, Committee on Biotechnological Works). In the United Kingdom the Health and Safety Executive, and in France the Minister for the Environment is competent. In both countries, too, expert bodies have to be consulted (Health and Safety Commission, Commission de génie génétique). The expert bodies exert a con13 14

15

See Hughes et al. 2002:356. Federal Administrative Court, Entscheidungen des Bundesverwaltungsgerichts (BVerwGE) 119:329,333/34; Neue Zeitschrift für Verwaltungsrecht 1991:1187; 1997:497. Regulation on Biotechnological Safety, Annex V I.

222

8 Legal problems of pharming

siderable influence on the contents of the consent, which has been criticized as a disguised form of governance by expert bodies.16 In Germany, the Act on Biotechnology (section 10 [7]) expressly requires that the opinion of the expert body must be considered and the grounds for deviating from it stated in writing. Nevertheless, from a legal point of view the ultimate responsibility is vested in the competent administrative authority.17

8.2.3 Development of recombinant medicinal products without containment 8.2.3.1 Scope of application and regulatory principles of Directive 2001/18

The development of transgenic medicinal products without containment is governed by Directive 2001/18. The directive in principle applies to the same type of activities as the contained use directive. It distinguishes between a release in the strict sense and the placing on the market (article 2 [3] and part B, article 2[4] and part C). A deliberate release of GMOs is an intentional introduction into the environment of GMOs for which no specific containment measures are used to limit their contact with people and the environment. A release which consists in making the GMOs, as such or in products, available to third persons is defined as placing on the market. However, a supply to contractors for the exclusive purpose of performing, or cooperating in the performance of, a simple release does not constitute a placing on the market (article 2 [4], 3rd indent). Directive 2001/18 is designed to protect human health and the environment from risks associated with the release of GMOs. The standard of evaluating and controlling these risks is the precautionary principle.18 In contrast to Directive 90/219, the applicability of the precautionary principle is expressly spelt out in the goals provision of article 1 and the fundamental obligation set forth in article 4 [1] of the directive. Recital 8 of the directive adds that the precautionary principle must be taken into account when implementing the directive. Likewise, the national laws and regulations that implement the directive adhere to the precautionary principle.19 Despite this clear pronunciation of legislative intent, it remains to be seen whether, and to what extent, the precautionary principle really guides the application of the directive and national law implementing it. 16 17 18

19

Reinhardt 2003:1446. Administrative Court of Appeal Mannheim, Neue Zeitschrift für Verwaltungsrecht 2002:224. See Christoforou 2004: 645; Boy 2002:9–13; Calliess and Korte 2006:11/12; with respect to the predecessor Directive 90/220: Hill 1994:180; Schenek 1995:182/83,199 et seq. Germany: Sections 1 No. l, 16 [1] Act on Biotechnology; United Kingdom: Section 106 [4] Environmental Protection Act 1990; France: Articles L110 [2] No. 1, L531-3, L531-4 and L533-3 Code de l’environnement.

8.2 Development phase I: Protection from risks to the environment

223

Transgenic animals can be kept with a minimum of containment, such as double fencing. An escape of these animals or an intrusion of other animals that then could disperse GMOs into the environment can normally be effectively prevented,20 although an escape of small mammals or theft or sabotage, for example by members of animal rights groups, cannot entirely be ruled out. The more important potential adverse impacts of animal pharming relate to the transport of the transgenic crude bulk material to the manufacturing premises, if any, the disposal of urine and manure from the herd and the disposal of excess animals. Therefore, the focus of possible risks to the environment clearly is on transgenic plants, especially from the perspective of spread of genes that have pharmaceutical properties. The crucial question is what distinguishes plant pharming from the “normal” cultivation of transgenic crops, and whether the legal requirements relating to information and the risk assessment to be supplied and performed by the applicant and the prerequisites for granting the authorization adequately respond to these concerns. 8.2.3.2 Information requirements and risk assessment

The directive (article 6) and implementing national law21 introduce an authorization procedure for releases of GMOs. Before starting a release, the operator must notify his intention to perform the release to the competent authority. The notification must be accompanied by documentation prescribed in article 6 and specified in annex III of the directive. Article 6 [2] of the directive obliges the operator to supply information relating to the GMOs, the conditions of release and the receiving environment, the interactions between the GMOs and the environment, and control and remedial methods as well as monitoring, waste treatment and emergency response plans. Annex III distinguishes between genetically modified organisms other than higher plants, to which category also belong transgenic animals, and genetically modified higher plants. A number of human health and environmental problems raised by plant pharming activities, such as direct adverse effects on human and animal health and the consequences of possible gene spread, are addressed by items listed in annex III B as part of the requisite information. Adverse effects on human health caused by the genetic modification are covered. Moreover, annex III B lists a variety of ecological impacts or impacts on agriculture that are relevant to plant pharming. This is true of sexual compatibility of the transgenic plant with other cultivated or wild plant species, as well as the presence of such plants at or near the site of release, the ways and extent of dissemination, the 20 21

Critical Schmitt 2004:30, from the perspective of risk from the environment. Germany: Section 14 [1], [5] Act on Biotechnology; United Kingdom: Sec. 111 EPA 1990 in conjunction with section 8 Genetically Modified Organisms (Deliberate Release) Regulations 2002; France: Article L533-3 Code de l’environnement.

224

8 Legal problems of pharming

proximity of officially recognized biotopes or protected areas which may be affected, methods for managing the release site, including cultivation practices and harvesting methods, and post release treatment methods. Since Directive 2001/18, due to its specificity, considerably limits the discretion of the member states with respect to the information requirements, national law closely follows annex III.22 The applicant must also include in the notification the environmental risk assessment, which every operator has to carry out in accordance with annex II before commencing the release (article 4 [2]). The risk assessment identifies, describes and evaluates the possible effects on human health and the environment associated with the release and, if necessary, indicates measures for risk management. Annex II sets forth the general principles that govern the risk assessment. Commission Guidelines on environmental risk assessment published in 200223 specify these requirements. Although adopted in the legal form of a Commission decision which is in principle binding on the member states, the guidelines only claim to supplement Annex II of the directive and therefore cannot have more extensive legal effects than the directive itself.24 There also is an EFSA Guidance document on the risk assessment of genetically modified plants and derived food and feed of 2004.25 Although this document is limited to EFSA involvement in the placing on the market of GMOs, it will be considered already in the development process where a later authorization under part C of Directive 2001/18 is required. Annex II prescribes, as part of the risk assessment, (1) the identification of inherent properties of the GMOs that may cause adverse effects, such as gene spread or transfer, genetic instability and interaction with other organisms, (2) an evaluation of the potential consequences of these adverse effects if they occur, to be classified, according to the guidelines, into 4 categories, (3) an evaluation of the likelihood of occurrence of such potential effects, (4) an estimation and classification of the risk posed by the each identified characteristic of the GMOs, (5) the definition of risk management measures and (6) a determination of overall risk taking into account these measures (annex II C. 2.). The guidelines specify these requirements and add to the estimation of the risk an element which is also important for plant pharming, namely that the operator must, in relation to the severity of the adverse effects to be expected, also identify, describe and evaluate the 22

23 24 25

Germany: Section 15 [1] Act on Biotechnology, sections 5, 6 and Annex III Biotechnology Procedure Regulation; United Kingdom: Section 111 [4] EPA 1990, section 11 and Schedules 1 and 2 Genetically Modified Organisms (Deliberate Release) Regulations 2002; France: Article R533-3 Code de l’environnement. Commission decision 2002/623, O.J. 2002 No. L 200:22. Brand and Winter 2004:273/74. Final edited version EFSA 2006a. EFSA is preparing a new guidance document on GMOs used for non-food or non-feed purposes; see EFSA 2008.

8.2 Development phase I: Protection from risks to the environment

225

degree of uncertainty. National law once again closely follows annex II of the directive.26 The risk assessment prescribed under the directive is based on the example of chemical substances. Its transfer to GMOs is problematic in view of a lower quantification potential and a higher degree of uncertainty. However, this is a general reform problem of GMO regulation and not a pharmingspecific one. Within these possible limitations, the risk assessment model of the directive is also capable of addressing potential adverse effects that are associated with plant pharming releases. Annex II D.2 lists a number of risk configurations that are also relevant to plant pharming. The most significant problem appears to be posed by the spread of genes having pharmaceutical properties from the transgenic, cultivated plants to sexually compatible, non-transgenic species of cultivated or wild plants. Thereby the GMOs might establish themselves in the natural environment and even enter into the feed/food chain. This type of potential risk is only partially addressed in annex II D.2. point 2 and 3. However, since annex II is not definitive, risk assessment is not limited to identifying, describing and evaluating gene spread from the perspective of selective advantage and disadvantage of the transgenic plants. Rather, the whole potential direct and indirect risks to human health and the environment presented by GMOs that have pharmacological properties must be identified, described and evaluated. Although not specific to plant pharming as far as the possible causation of gene spread as such is concerned, the evaluation of the risk associated with it may engender different implications due to the pharmaceutical characteristics of the GMOs. Moreover, adverse interactions with non-target organisms (annex II D.2. point 5) cannot be ruled out. Another, probably less relevant aspect of plant pharming releases is possible indirect or delayed effects on health of people who come into contact with, or are in the vicinity of, the release (annex II D.2. point 6). The same is true of effects on animal health and the consequences for the feed/food chain resulting from consumption of the GMOs (annex D.2. point 7), since the transgenic plants used for pharming are not intended to be used as animal feed. However, in this situation there may be misuse and adventitious admixtures with conventional feed. In the United States, a permit is required for releases of transgenic plants that may cause harm to plants (7 C.F.R. § 340.0, § 340.2). With respect to transgenic plants that are deemed to be associated with no or minor risks (“non-regulated status”), the applicable regulations only require a notification. This applies to about 90 percent of all releases. However, the reg26

Germany: Section 15 [1] No. 4 in conjunction with section 6 (1) Act on Biotechnology, section 5 Biotechnology Procedure Regulation; United Kingdom: Section 108 [1] [a] Environmental Protection Act 1990, sections 6, 11 [1] [c] Genetically Modified Organisms (Deliberate Release) Regulations 2002; France: Article R533-3 No. 4 Code de l’environnement.

226

8 Legal problems of pharming

ulations exclude introduced genetic material which encodes products for pharmaceutical use from this procedural alleviation (7 C.F.R. § 340.3 [b] [4] [iii]). The operator is obliged to submit information which, by and large, resembles that needed in Europe, but is somewhat less specific regarding environmental impacts such as escape, dissemination and persistence in the environment (7 C.F.R. § 340.4). An environmental assessment may be prescribed where this appears necessary to evaluate the potential environmental effects of the release, for instance where a new genetic modification raises problems that cannot be addressed through normal safety measures (7 C.F.R. § 372 [d] [4]). Moreover, the assessment serves to determine whether the proposed action may significantly affect the quality of the environment and thereby be subject to an environmental impact assessment under the National Environmental Policy Act (7 C.F.R. Parts 1b and 272). The latter requires that the decision on the release is a major federal action, which is not normally considered to be the case with pharming development operations. The Department of Agriculture has recently published a guidance document relating to the relevant permit process which in particular applies to plant pharming development.27 As stated, the development and manufacture of pharmaceuticals from transgenic plants and animals can also be regulated within the framework of pharmaceuticals regulation under the Federal Food, Drug and Cosmetic Act, especially as part of the pre-marketing permit procedure. Under the applicable regulations, an environmental assessment may also be needed (21 C.F.R § 25.21 and § 25.22). 8.2.3.3 Authorization prerequisites

Basic standard. Article 4 [1] of the directive establishes the basic standard for granting the authorization for a release. Direct or indirect adverse effects on human health and the environment which might arise from the release, in the case of plant pharming especially through the spread of genes having pharmaceutical properties, must be avoided. The decision on the application for an authorization is taken on a case by case basis considering the information and risk assessment provided by the operator, but also using other information that is available. This decision-making technique opens the process to political influence. This is most visible at EC level and perhaps less so at national levels. However, its advantage is a higher degree of flexibility and capability to adjust to novel configurations and achieve regulatory innovations. Moreover, it can be justified on grounds of political accountability and democratic legitimacy.28 Article 4 [1] of the directive has to be interpreted in the light of the precautionary principle, which also means that controversies relating to the 27 28

USDA/APHIS 2008. To this extent European Court of First Instance 2002 ECR II 3305 No. 201 – Pfizer Animal Health; Christoforou 2004:679–682,695,705; see also Breyer and Heyvaert 2000:330–337.

8.2 Development phase I: Protection from risks to the environment

227

interpretation of this principle may gain some importance. A clear Community standard of decision-making that would have to be uniformly implemented under national law and applied on the ground is not provided, although the annexes to the directive have been literally transposed into national law. This is in particular true of the questions as to which effects are insignificant and whether there are spatial and temporal limits to the scrutiny of a release.29 The consequence is a certain degree of divergence in the statutory formulation of the authorization prerequisites in the member states as well as in the practice of the authorities that are responsible for applying the respective national laws. For example, section 16 [1] of the German Act on Biotechnology establishes two major prerequisites for granting the authorization, namely that the operator must have taken all safety measures that are necessary, according to the state of science and technology, and that the release may not cause unacceptable adverse effects on human health and the environment. The relationship between these two prerequisites is controversial. In accordance with the German tradition of applying the precautionary principle and in view of the systematic structure of the provision, one should assume that the requisite safety measures have to be taken independent of concrete risks insofar as the measures are technically available and scientifically necessary as a precaution against, and proportionate to, potential risks presented by the release. By contrast, the second requirement of avoiding unacceptable adverse effects addresses risks presented in spite of such safety measures being taken.30 Moreover, section 16[1] of the Act enlarges the authorization prerequisites by a risk-benefit analysis, whereby the acceptability of the risks presented by a release has to be determined in the light of the benefits conferred by it. Risk-benefit analysis is said to operate as an additional filter for eliminating low risks that are not justified by the benefits associated with the use of biotechnology. It is not meant to make higher risks acceptable in view of the benefits derived.31 One can justify this as an extension of proportionality which governs the application of the precautionary principle.32 The benefits of the release can be regarded as economic chances of placing transgenic products on the market; if these chances are foregone they are costs that have always to be considered in applying the precautionary principle. Nevertheless, it is 29 30

31 32

Lewidow et al. 1996:145,146; Ostertag 2006:339,340. In this sense Jörgensen and Winter 1996:296; Winter 1998:106 et seq.; Brand and Winter 2004:233; in the reverse sense Hirsch and Schmidt-Didczuhn 1991, § 16 No. 12; Dederer in: Ebersbach/Lange/Ronellenfitsch 2007 § 16 No. 70; Ostertag 2006:372. A matrix for structuring the decision using the criteria of likelihood, extent of possible harm, quality of the effects and degree of certainty is proposed by Winter 2006:459. Winter 2006:456,457/58; Brand and Winter 2004:251; in favour of the risk-benefit assessment also Ostertag 2006:373. COM 2001, 1 final:20.

228

8 Legal problems of pharming

clear that Directive 2001/18 does not provide for a risk-benefit analysis. Where Community law has opted for such an analysis, such as in the regulation of pharmaceuticals, plant protection products, biocides, and particularly hazardous chemicals for general use, it has expressly provided for it. Therefore, the insertion of a risk-benefit evaluation into the authorization prerequisites by German law raises some legal doubts relating to possible pre-emption,33 although it is clearly desirable as a matter of policy. The risk-benefit evaluation does not play any role in the practice of the German permit authorities. In the United Kingdom, the authorization prerequisites can be implied from the basic safety obligations of the operator under section 109 [4] Environmental Protection Act 1990. The operator shall not release the GMOs if it appears that, despite the precautions that can be taken, there is a risk of damage to the environment being caused as a result of the release; moreover, the operator has to apply the best available technology while not entailing excessive costs. It seems plausible that the relationship between precautionary measures and acceptability of remaining potential risk is the same as suggested for German law.34 The possible benefits derived from the release are not considered. Under French law, the decision on the application for an authorization to release GMOs is discretionary.35 Article 533-3 Code de l’environnement does not contain any express authorization prerequisites. These can, at best, be derived from the statement in the law that the decision on the application for an authorization to release GMOs is taken after an investigation of the risks to human health and the environment. In any case, they are much less specific than in Germany or the United Kingdom. In the United States, as can be concluded from the definition of plant pests and the information requirements of the Department of Agriculture regulations, the primary concern of the permit procedure for the release of GMOs is the prevention of risks to agricultural plants and also of risks presented to the environment, especially the environment beneficial to agricultural plants. The environmental perspective comes into play insofar as the competent agency performs an environmental assessment of the relevant releases of GMOs, which normally is the case with pharming development operations.36 33 34 35

36

See generally Christoforou 2004:671/72,677,682/83. See Macrory and Purdy 1998:69/70. See, with respect to marketing, Conseil d’État, Revue juridique de l’environnement 1999:561,563; confirmed by European Court of Justice 2000 ECR I 1676 No. 39 – Greenpeace; Brand 2004:148 et seq. See USDA 1986 Reg. 23302, 23313-19; USDA 2007 Reg. 14649; example of an environmental assessment of a trial release: USDA/APHIS, Environmental Assessment of 22 June 2007 (06-363-1035) concerning the use of sunflower as a pharming development platform.

8.2 Development phase I: Protection from risks to the environment

229

Likelihood of effects. In keeping with the precautionary principle, under the directive also such possible adverse effects must be identified, evaluated and eventually controlled which cannot be excluded because, according to the present state of scientific knowledge, the possibility of future harm being caused can neither be positively determined nor ruled out, but there is reason to believe this may occur. The question as to the necessary degree of scientific substantiation of this “potential for concern” is controversial. The German Act on Biotechnology (section 16 [1] No. 3) requires that adverse effects that are expected must be excluded. The German administrative courts37 go relatively far in assuming that, in the case of GMO releases, there are justified grounds for concern. In the United Kingdom, for denial of the authorization it is sufficient that it “appears” that a risk of damage is being caused (section 109 [4] Environmental Protection Act 1990). This suggests that reasonable grounds for concern are sufficient but also necessary. The European Court of First Instance, in two pharmaceutical cases that involved the withdrawal of a permit and in a more recent case regarding the listing of an active ingredient for a plant protection product,38 has decided that a mere hypothetical risk does not permit precautionary action. Rather, there must be some scientific foundation for believing that adverse effects may occur (scientifically plausible grounds for concern), although the court does not necessarily require empirical findings and emphasizes the normative nature of the decision on tolerability of risk. These decisions are of general importance beyond the narrow field of pharmaceuticals law. They have been followed by the Commission of the European Union in its Communication on the precautionary principle.39 Their reasoning is also shared by a number of commentators. Therefore, one may assume that the decisions set the future standard of precautionary analysis also in the context of national biotechnology law, insofar as it implements Directive 2001/18. As a matter of policy they make sense, even if one considers that the first two cases do not concern an initial authorization but the withdrawal of an authorization already granted where vested interests are at stake. In granting an initial permit the standard of scrutiny may be stricter. It is clear here that remaining uncertainties must be taken into account and evaluated as to the question of whether they are tolerable or not. However, the rule of law and the protection of economic fundamental rights militate against the prohibition of business activities by a denial of a permit, in case of uncertainty, that is 37

38

39

Administrative Court Gießen, Neue Zeitschrift für Verwaltungsrecht – Rechtsprechungs-Report 1993:534,537/38; Administrative Court Berlin, Neue Zeitschrift für Verwaltungsrecht – Rechtsprechungs-Report 1994:150,152; Zeitschrift für Umweltrecht 1996:146,147. 2002 ECR II-3305 Nos. 143–146,152 – Pfizer Animal Health; 2002 ECR II-4945 Nos. 181 et seq. – Artegodan; case T-229/04, judgement of 11 July 2007, Nos. 161,170 – Sweden/Commission (not yet published). COM 2001, 1 final:15–18.

230

8 Legal problems of pharming

merely based on speculative concerns. Post-event monitoring provides an additional safety net provided it is extensive and long-term. The interpretation of the precautionary principle is particularly relevant with respect to horizontal gene transfer from transgenic plants to sexually non-compatible organisms, which quite a number of scientists would denote as purely speculative. However, the more immediate concern in plant pharming is gene spread from transgenic cultures to conventional ones, and the entrance of genes with pharmacological properties into the feed and food chain. The likelihood, frequency and extent of such gene spread or adventitious commingling, as well as the safety that can be provided for receiving cultivated plants by confinement and other management measures, are in principle amenable to scientific research although there are limits to the validity of the findings stemming from the long-term and systemic effects of the releases of GMOs.40 Scientific uncertainty, of the kind which the precautionary principle is designed to address, comes in any case into play where one attempts to transfer empirical research results, gained from a particular crop and physical environment, to other crops and environments. Although this kind of transfer of findings is in principle scientifically accepted, and uncertainties are accommodated for by inserting “prudential” elements in the risk assessment process, drawing conclusions from studies relating to other crops and physical environments is scientifically problematic in the field of agricultural and ecological effects. In the United States the precautionary principle is not accepted in the regulation of GMOs. Rather, the competent agency determines whether there is an unacceptable risk to agricultural plants and the environment. In case of uncertainty or ignorance the release will be permitted.41 The decision on the application is taken on the basis of a classical risk assessment which is divided into four steps, namely hazard identification, risk assessment, risk evaluation including determination of risk management measures, and riskrisk comparison. A risk-benefit evaluation does not take place.42 Definition of adversity. Apart from the question of possible causation of effects on human health and the environment, the definition of harm or adversity of an effect is far from clear. Problems of interpretation of the directive, in this respect, are addressed in the case by case evaluation but deserve more fundamental discussion. The first starting point is the notion of environment, which encompasses all media and elements that constitute the environment as well as their interrelationship. The agricultural environment is included.43 Furthermore, it is clear from the text of the direc40 41 42 43

See Breckling 2004:52–64,68,69–77; Sukopp 2004:100–113. Fisahn 2004:186. Anderson et al. 2001:18; Dederer 1998:281/282; Steines 2002:172. Annex III B section E 3., G 1. [a] Directive 2001/18; No. 3, 1st and 3rd indent Directive 2002/623 (Guidance on risk assessment of GMOs); Brand and Winter 2004:231; Ostertag 2006:275–379; Herdegen 2004:19.

8.2 Development phase I: Protection from risks to the environment

231

tive (article 1 No. 8, article 4 and annex II D.2.5) that direct and indirect, immediate and delayed effects are covered. Finally, as adversity is a qualified effect and more than a simple alteration of the environment, a certain threshold of significance is inherent in the term. It is a fact, corroborated by spectacular incidents such as the Starlink case, that gene spread through pollen dispersal and seed dispersal and commingling cannot be entirely prevented. Consequently, the admission of biotechnology in agriculture by Directive 2001/18 (and its predecessor Directive 90/220) implies that simple alteration of the environment as such cannot be deemed to be unacceptable in principle. In German practice, gene spread and transfer of genes, for example by out-crossing to sexually compatible agricultural plants or wild plant species, by volunteers of the transgenic plant or by commingling, are not yet considered as adverse effects.44 This position is shared by section 107 [6] of the British Environmental Protection Act 1990 which requires an interference with, rather than a simple alteration of, the ecological system. The European Food Safety Authority (EFSA), which is involved in the risk assessment of commercial cultivation of transgenic crops, also seems to take the same position.45 In German practice, an evaluation according to the standards of naturalness (equivalence to nature) and selective advantage is carried out. Effects of the same kind as occur in nature, or which can be caused by conventional breeding, are not considered as adverse or at least are considered as acceptable. In the absence of a new selective advantage, adversity is normally denied. While new properties can spread in the environment, one assumes that nature can adjust to them. However, a special evaluation is necessary when a release is associated with a novel process. New selective advantages, due to strong propagation potential and higher vitality, are relevant where they can lead to the establishment of a transgenic plant in the environment. These advantages must be evaluated regarding their kind and consequences for nature.46 In British practice, too, only an additional risk is considered as relevant; factors such as the scale of the release, the genetic predictability of the organism, and the relationship of the size of the genetic change of the gene construct and its likely consequences for the affected environment are considered.47 In France, one deems risks originating from a release to be acceptable if there is sufficient knowledge. 44

45 46

47

See Administrative Court Berlin, Zeitschrift für Umweltrecht 1996:146,147; Brand and Winter 2004:246; Fisahn 1998:38 et seq.; Fisahn himself is critical of this position and pleads for growth intervals (as they partly already exist in Denmark); see Fisahn 2004b:145. EFSA 2006b. The Commission has not yet granted the authorization but, rather, remanded the case to EFSA for further review. Fisahn 1998:186; Department of the Environment/ACRE, Guidance Note 1, The Regulation and Control of the Deliberate Release of GMOs, 1993, as amended 1995, Nos. 2.10,2.11. Under the 2002 regulations, there is no new guidance. Guidance Note, supra note 46, No. 2.9; Lewidow et al. 1996:145.

232

8 Legal problems of pharming

It is doubtful whether the concept of equivalence to nature sufficiently addresses, in view of the limited knowledge about the systemic, spatial and temporal interrelationships in nature, the dynamics and complexities of ecological systems. Therefore, it has been proposed to supplement it by recourse to natural variation of affected ecosystems and their development trends, as well as those of affected species.48 However, this is not an issue that is specific to pharming. The genetic modification of plants used for pharming does not normally entail a new selective advantage, since the genetic modification concerns other properties of the plant. Such an advantage could only be an unintended side effect of the genetic modification that aims to equip the plant with pharmaceutical properties. The more problematic aspect of pharming arises from these properties of GMOs. Notwithstanding the fact that plants may contain pharmaceutical properties, the kind of genetic modification that is generated by pharming – for example the introduction of pharmaceutical properties based on human or animal antibodies or vaccines into plants – is alien because it could not be brought about by conventional breeding. Therefore it constitutes a potentially adverse effect per se. This requires a case by case evaluation as to the effects on nature. There is another reason why the appropriateness of the criteria of naturalness and selective advantage for determining adversity is limited. They are geared to structural effects on the environment, and do not necessarily address the protection of human and animal health. The same is true of eco-toxic effects, as the case of transgenic Bt-maize shows. In this field, certain properties of a transgenic plant as such may constitute a potential adverse effect. Pathogenic, toxic and allergenic effects presented by transgenic plants must, in principle, be classified as adverse, independent of whether they are directly associated with the inserted gene sequence or caused indirectly by a change in the metabolism of the plant. The same is true of eco-toxic effects. A qualification of this classification may ensue from considering significance (threshold concentrations) and exposure.49 In the latter respect, the dispersal behaviour, the propagation patterns, the existence of sexually compatible food or feed plant or wild plant species as well as the selective advantage of the transgenic plant species, independent of the genetic modification, are most important. The analysis of naturalness and selective advantage of a transgenic plant is not relevant when exposure cannot be ruled out. Selective advantage may come into play, as an additional factor, in order to determine the kind and extent of exposure but is not essential. 48 49

Sachverständigenrat für Umweltfragen 1998:Nos. 813 et seq.; Breckling 2004:79–83. Administrative Court Berlin, Zeitschrift für Umweltrecht 1996:146,147; Hirsch and Schmidt-Didczuhn 1991:§ 16 No. 15; Dederer, in: Ebersbach/Lange/Ronellenfitsch 2007:§ 16 No. 100; Guidance Note, supra note 46, Nos. 2.9,2.10.

8.2 Development phase I: Protection from risks to the environment

233

An adverse effect on human and animal health may also be constituted by compromise of prophylactic or therapeutic medical or veterinary treatments. The involuntary intake of GMOs in food or feed may cause resistance to pharmaceuticals. Such effects may occur either by gene spread and the resulting propagation of transgenic cultivated plants on neighbouring fields, or by adventitious commingling. Directive 2001/18 contains a phase-out programme for antibiotic resistant marker genes (article 4 [2] subparagraph 2). This implies that causation of resistance to treatment with antibiotics through gene transfer, or otherwise, may constitute an adverse effect. Moreover, annex II (C.2. point 2, 5th indent) generally includes the issue of causing resistance to pharmaceuticals in the risk assessment. Hence, under the directive adversity of such indirect effects on health cannot be questioned as a matter of principle. Whether concrete pharmaceutical properties of transgenic plants used in pharming present such risks is a matter of individual assessment of the kind and extent of effect and likelihood of causation. From an initial perspective, apart from the properties of the host plant, including its transgene content and the state of the recipient environment, one might have to distinguish here between different kinds of pharmaceutical effects, in particular the degree of bioactivity (see section 3.7). In any case, the relevant risks are more difficult to trace and a determination that the risks presented are negligible appears problematic. In contrast to antibiotic resistant marker genes, where the probability of horizontal gene transfers to microorganisms that could cause such effects, according to the present state of knowledge, is very low, plant pharming arguably presents a higher potential risk because some gene spread, due to pollen or seed dispersal, is to be expected and commingling cannot be ruled out either. In the United States, a variant of the concept of equivalence to nature and selective advantage is applied as well. In principle, one assumes that the risks presented by the release of GMOs do not fundamentally differ from those associated with traditional plant breeding, although their extent may be different. Apart from adverse effects on human health, persistence in the environment including the agricultural environment, development of weediness, and risks of commingling with non-transgenic seeds are the relevant factors that are considered.50 Risk management. Depending on the outcome of the risk assessment, a significant gene spread and commingling may have to be reasonably excluded. If this is not possible, the authorization must be denied. The measures to be imposed are governed by the principle of proportionality.51 They largely 50

51

USDA/APHIS, Environmental assessment of 22 June 2007, supra note 36; USDA/ APHIS, Final Environmental Assessment of 28 June 2007, 05-354-015 concerning the use of tobacco as a production platform for a pharmaceutical; Anderson et al. 2001:18; Fisahn 2004:186. Christoforou 2004:706/07; Commission, supra note 32:20/21.

234

8 Legal problems of pharming

depend on a weighing of conflicting values and will also be influenced by public attitudes towards GMOs. Radical solutions would require a contained use, limiting the plant species that can be used for pharming to nonfood/feed plant species, or prohibiting an expression of pharmaceutical properties in the pollen of the plant. This is, at least, plausible in the case of highly bioactive genes. In other cases using some form of confinement, establishing safety distances and employing good farming practices may be sufficient. The duties of care and rules of good professional practice, established under national law for avoiding adverse effects on human health and the environment and ensuring coexistence, can serve as a model, although not directly applicable because they are limited to cultivation and handling of transgenic products whose placing on the market has been authorized.52. For instance, section 16b German Act on Biotechnology contains a list of measures to prevent gene spread and commingling such as minimum distances, selection of suitable plant varieties, combating volunteers, use of pollen barriers, segregation in storage and transport and cleaning of containers. The necessary safety requirements can be imposed on the operator by a condition attached to the authorization. Nature conservation areas. A special regime under nature conservation law is provided for releases of GMOs, especially transgenic plants, that may affect a protected habitat established under the Habitats Directive (Directive 92/43, as amended), more exactly under national law that implements the directive.53 Although the risk assessment to be carried out by the operator under the release directive covers certain ecological effects, such as impacts on species and habitats affected by the release, Article 6 [3] of the directive and national law54 require the performance of a habitat impact assessment whenever a release of GMOs, within or outside the protected area,55 may considerably impair such an area. The habitat impact assessment is designed to determine whether the release is compatible with the conservation objectives established for the area. If it is not, an authorization for the release can only be granted where cogent reasons of paramount public interest justify the project and no other feasible alternative is available (article 6 [4] Habitats Directive). GMO regulation and habitats regula52

53 54

55

Section 14 [2] German Act on Biotechnology which declares sections 16a and 16b to be applicable to products governed by a special regime is limited to authorisations for the marketing of the product and does not cover mere releases. See also EFSA 2008:24–26. See Winter 2006:456; Palme and Schumacher 2007:16. Germany: Sections 32, 34 Protection of Nature Act; United Kingdom: Sections 18–21 Conservation (Natural Habitats) Regulations 1994, as amended; France: Articles L414-4, R214-19 to 23 Code de l’environnement. Germany had previously excluded releases outside the protected area but, following a judgement of the European Court of Justice (2006 ECR I 53 Nos. 39–45) that held this to be in violation of the directive, has extended its law to this type of release.

8.2 Development phase I: Protection from risks to the environment

235

tion apply cumulatively, although in case of releases that concern a specific, defined area – in contrast to placing on the market – it is difficult to distinguish between the two kinds of risk assessment. 8.2.3.4 Special issues relating to animal pharming

The authorization procedure, the authorization prerequisites and the associated obligation of the operator to supply prescribed documentation and perform a risk assessment under Directive 2001/18 are also applicable to the creation of transgenic animals. Annex II D (risk assessment) and annex III A (information to be supplied) contain special requirements applicable to GMOs other than higher plants. This includes transgenic animals. The hazards presented by elements of the development procedure, such as viral vectors, and by the pharmaceutical properties expressed in the founder animal and the production herd must be identified and evaluated. Moreover, a possible gene spread through an escape of the transgenic animals, as well as an uptake of transgenic material by intruding animals and the ensuing possibility of survival and propagation, constitute a relevant potential risk that must be identified and evaluated. This is particularly true of small mammals. Other pathways of potential gene spread, such as discharge of urine and gaining and processing the crude bulk material such as milk derived from transgenic animals, are also relevant. However, it is safe to say that these risks can be reasonably excluded through confinement and other protection and control measures, in such a way that they do not create obstacles to the granting of the authorization. 8.2.3.5 Institutional arrangements

In contrast to placing GMOs on the market, the evaluation and authorization of releases is entrusted to the authorities of the member states (article 6 [1] Directive 2001/18). Community involvement in the procedure is, in principle, limited to information exchange, including an opportunity for the Commission and the other member states to present to the permit authority observations on the planned release and try to influence its decision (article 11). However, the permit authority is sovereign in its ultimate decision. In this information exchange, the Commission is supported by the European Food Safety Authority (EFSA) (article 28 [2] Directive 2001/18). An exception is possible where a member state authority intends to apply a simplified procedure; here the Commission can decide on the conditions of the release (article 7 Directive 2001/18). Moreover, it should be noted that “forum shopping” is easy. If a national authority denies a permit, the applicant can try again in another country. National law attributes the competence for authorising releases to central authorities. In the decision-making process, expert bodies play an important role while public participation appears to be less influential. In Germany, the Federal Office for Consumer Protection and Food Safety

236

8 Legal problems of pharming

(Bundesanstalt für Verbraucherschutz und Lebensmittelsicherheit – BVL) is responsible. It decides after coordination with other federal agencies and consultation of the Central Commission for Biosafety (Committee on Releases and Marketing). In the United Kingdom, the Department of Environment, Food and Rural Affairs (DEFRA) is responsible. It normally has to consult the Administrative Commission on Releases to the Environment (ACRE) and, with respect to effects on human health, decides with the agreement of the Health and Safety Executive (HSE). In France, administrative competences are divided among the ministers or agencies responsible for the environment, health, agriculture and research depending on the kind of release; the agreement of the Minister for the Environment is always required. In the case of pharming, the Minister for Agriculture is competent. Before making the decision, an expert body (Commission de génie biomoléculaire – CGB) must be consulted. As regards public participation and access to information, article 9 Directive 2001/18 requires that the public, including environmental groups, is consulted on the proposed release and mandates the member states to make available to the public information on all part B releases. The public must be given an opportunity to express an opinion. The directive also limits confidentiality of information. In particular, the general description of the GMO, the purpose of the release, the location of the release, the intended uses, monitoring and emergency plans and the risk assessment are not deemed to be confidential (article 25 [4]). These provisions have been implemented by the member states in a narrow and somewhat different way.56 Germany and the United Kingdom require that the full dossier accompanying the application (except for confidential information) is made available for inspection. By contrast, France limits general access to information on the risk assessment to a mere summary, so that an interested person would have to take recourse to the provisions of the Code de l’environnement on environmental information (articles L124-1 to L124-8) in order to get full access. In all three countries public participation is in the form of individual comments which can be made during a limited period of time. There is no public hearing. The attribution of administrative competences for the control of releases is based on a deliberate decision that is motivated by the principle of subsidiarity (article 5 [2] EC Treaty). It includes cases where the authorization for the end product is centralized. A certain variation in the national authorization practice is taken into account. It is mitigated by the consultation procedure set forth in article 11 Directive 2001/18. Therefore, the mere fact that 56

Germany: Section 18 Act on Biotechnology, section 5 Biotechnology Procedure Regulation, Biotechnology Hearing Regulation of 1990, as amended in 1996; United Kingdom: Sections 11, 12, 33 and 34 Genetically Modified Organisms (Deliberate Release) Regulations 2002; France: Articles L 535-5, R533-5 and R533-10 Code de l’environnement.

8.2 Development phase I: Protection from risks to the environment

237

recombinant medicinal products are subject to a centralized authorization procedure is not sufficient to require that the competences for the release of transgenic plants and animals used in the pharming process should also be transferred to the European level. While one could argue that the novel kind of risk involved justifies such a transfer, as long as decentralization does not prove to be clearly inadequate in the field of pharming the extension of Community powers should be renounced. Decentralized decision making offers better chances for social learning, and this is important in an area of high uncertainty. This does not exclude the establishment of EU guidelines that collect already-existing national experience and provide for minimum harmonization.

8.2.4 Coexistence between experimental cultivation of GMOs and organic and conventional agriculture Adverse impacts of gene spread and commingling on the ability of farmers to maintain production and market their products according to their quality preferences are not considered as harmful and, hence, are neither a subject of the risk assessment nor of the authorization and conditions attached to it. Article 26a Directive 2001/18 and the Commission Recommendation on coexistence57 conceive the problem of coexistence between pharming, including experimental releases of GMOs, and conventional and organic agriculture as a purely economic problem which may be addressed by the member states. This is based on the assumption that adverse effects on health and the environment, including the agricultural environment, that may be caused by gene spread and commingling during the development process are already practically avoided by confinement and other protection measures taken under the directive. The problem of coexistence in agriculture mainly concerns large-scale cultivation of transgenic crops which occurs on the basis of a marketing authorization. If one stresses the very concept of coexistence between different methods of agriculture, it seems natural not to include trial releases of transgenic plants, the more so since such releases are normally performed subject to some form of confinement. However, seen under the perspective of the victim who suffers economic losses due to a release, the concept of coexistence could be understood in a broader sense. Article 26a [1] Directive 2001/18 generally empowers the member states to take action to prevent the adventitious presence of GMOs in other products. It is only from article 26a [2] of the directive that one could, at best, derive a limitation of the concept to coexistence between genetically modified, conventional and organic agriculture. The Commission Recommendation on coexistence58 seems to interpret article 26a of the directive in this sense, although this conclusion 57 58

Commission 2003:36. See note 57.

238

8 Legal problems of pharming

is not cogent. In national practice there is a clear focus on conflicts between different methods of agriculture. However, certain rules, such as the provisions relating to the location registry, cover experimental releases as well. Moreover, some states such as Germany have special liability rules with respect to authorized releases (section 23 Act on Biotechnology).

8.2.5 Waste disposal In pharming, development processes generate different kinds of waste. In plant pharming, straw and other plant materials are generated from processing. Animal pharming is a source of urine, manure, excess animals and unusable animal parts that must be disposed of. This waste contains, or is at least likely to contain, the inserted genes or gene products. Therefore, special controls are necessary. 8.2.5.1 GMO-specific regulation

The legal regulation of GMO waste is primarily provided by Directive 90/219 on the contained use of genetically modified microorganisms (as amended by Directive 98/81) and Directive 2001/18 on the release of genetically modified organisms into the environment, as implemented and – in case of contained use – supplemented by national law. The applicability of the two regimes depends on whether the pharming operations are carried out with or without containment. The risks to human health and the environment, presented by GMM and GMO waste originating from pharming activities, constitute part of the risk assessment the operator has to perform (article 4 [2] Directive 90/219; article 4 [2], annex II Directive 2001/18) and the competent authority can subject the contained use or the release to conditions relating to waste disposal (article 11 [3] Directive 90/219; article 6 [7] Directive 2001/18). However, the two directives and national law implementing and supplementing them do not establish an exhaustive regime. Therefore, unless there are specific legal provisions or conditions attached to the authorization, reprocessing, incineration and disposal under general law are in principle possible. Whether pre-treatment is necessary depends on national law. Where GM plant material is processed as animal feed, a special authorization must be secured under Regulation 1829/2003. Directive 90/219 sets forth some provisions on the treatment of waste produced by genetic activities with containment that have been implemented and supplemented under national law.59 The applicable requirements are different according to the relevant safety level (containment level). As regards the safety level, 1 the directive sets forth relatively leni59

Germany: Regulation on Biotechnological Safety; United Kingdom: Genetically Modified (Contained Use) Regulations 2000; France: Articles R532-1 to R532-17 Code de l’environnement.

8.2 Development phase I: Protection from risks to the environment

239

ent and not very specific measures for waste disposal. The operator must include information about waste disposal (annex V part A) in the initial notification to be made under article 7 of the directive. According to the “General Principles” contained in annex IV, in laboratory activities inactivation of GMMs in contaminated material and waste is optional and only required as of safety level 2. As regards development activities using glasshouses and rooms for cell cultures, minimization of dissemination, including waste-related dissemination, of GMMs is prescribed. There are no specific waste-related requirements applicable to animals kept in laboratory units in level 1 facilities. With respect to activities other than laboratory activities, regarding safety level 1 activities Annex IV Table II requires the inactivation of GMMs in contaminated material and waste, including those in process effluent before final discharge, only where the results of the risk assessment suggest this in an individual case, and declares it to be generally mandatory only as of safety level 2. Assuming that developmental pharming will be classified as level 1 (no or negligible risk; see above 8.2.2), the requirements set forth in annex IV of the directive do not appear to be particularly stringent. However, national law supplements EC regulation through more extensive (all GMOs) and more stringent requirements. In Germany, section 13 of the Regulation on Biotechnological Safety sets forth the general requirement of disposal of GM waste according to the state of science and technology; general waste law remains applicable. With respect to safety level 1, the regulation dispenses the operator from special pre-treatment of waste when animals or plants are used and where adverse effects to human health or the environment are not expected, or the waste is so slightly contaminated that the risk is negligible. Otherwise, inactivation of GMOs is prescribed. In the UK, there are no special duties of care relating to waste generated in the process; rather, under section 17 [1] of the Genetically Modified Organisms (Contained Use) Regulations, the general duty of minimising exposure and risk applies. Moreover, British law goes beyond the directive in generally prescribing inactivation of GMMs in waste also for class l activities (schedule 8 No. 16). France has literally transposed the relevant provisions of the directive. In contrast, with respect to releases without containment Directive 2001/18 exclusively addresses the waste problem at the level of the environmental risk assessment. As regards GM animals, the applicant must include information about the type and amount of waste generated by the planned release (article 6 [2] [a], annex IIIA point III.V.C) in his/her notification. In plant pharming the operator must describe the post-release treatment methods (article 6 [2] [a], annex IIIB point G 3). Although not specifically mentioned, the waste problem must also be dealt with in the risk assessment to be performed by the applicant. Based on the risk assessment, the competent national authority can make the authorization subject to condi-

240

8 Legal problems of pharming

tions relating to waste disposal (article 6 [8]). In doing so, it may consider the requirements applicable to contained use under Directive 219/90. The relevant provisions of Directive 2001/18 are implemented under national law. While the waste regime applicable to contained use is satisfactory, the concept of sole reliance on case by case assessment under the release directive appears problematic from a legal point of view, the more so since general law is not very demanding either. However, as in the case of contained use in practice one normally requires incineration of the waste which solves the relevant problems. 8.2.5.2 Regulation under general waste law

As regards general waste management law, the relevant EC directives do not contain special regulation on GMO waste. Directive 91/689 on hazardous waste, as amended by Directive 94/31 and Regulation 166/2006, theoretically applies to several categories of GMO waste as the European List of Hazardous Waste (Council Decision 94/904) includes animal tissues, animal urine and waste from milk processing, production of pharmaceuticals and research and development. However, the directive concentrates on hazards to human health and therefore only covers those categories of pharming waste that are toxic, or otherwise potentially harmful, to human health (for example cancerogenic or teratogenic), and provided the GMO substances contained exceed certain concentrations levels (Council Decision 94/904). Hazards to the environment are not included (see Directive 91/689, annex III). Therefore, even recourse to article 1 [3] 2nd indent of the directive, which empowers the member states to also classify unlisted waste as hazardous, is not available where environmental hazards are at issue. Nevertheless, the member states could enact more stringent national regulation that applies to such waste under article 176 EC Treaty. As far as can be seen, no member state has used these powers with respect to GMO waste as yet. 8.2.5.3 Disposal of excess animals and animal parts

As regards excess animals and animal parts, there is special, although not GM-specific regulation relating to the disposal of animal side products under EC Regulation 1774/2002 and national regulations specifying it. Regulation 1774/2002 applies to all animal side products such as whole animals, animal parts, skins, wool und urine. As a reaction to animal pests and human diseases caused in the past by feeding reprocessed protein feeding stuff, the regulation is designed for the protection of human health, the health of animals and the environment against significant risks by controlling the relevant waste. The regulation classifies animal side products into different risk categories and attributes to them particular waste management options (articles 4–6). Category 1 comprises special risk material, including among others parts from experimental ani-

8.3 Development phase II: Animal protection

241

mals; here, incineration or pre-treatment in special facilities, before normal incineration or closed deposit on land, is required (article 4). Category 2 applies, among others, to animals slaughtered for purposes other than trading purposes. In this category, a variety of waste management methods is permitted such as incineration, pre-treatment for incineration or deposit on land, processing for the production of biogas, fertilizers, cosmetics, pharmaceuticals and medical equipment (article 5). This also applies to urine from animals which, unless there is a special risk of illness, can in addition be sprayed on land as fertilizer. Category 3 comprises low risk materials such as milk and wool. Milk can be processed as pet feed (article 6). The latter two provisions raise the question as to whether urine from pharming animals must be regarded as presenting a special risk of illness, and milk from transgenic herds can be treated just as any other milk. In view of the limited purpose of the regulation, one should arguably answer the question in the negative, since the kind of risk presented by GMO traces in the waste materials does not correspond to the kind of illness the regulation wants to counteract. However, it is safe to say that a new exposure of the environment to GMO traces is subject to case by case scrutiny under Directive 2001/18. The regulation also contains detailed obligations for collecting and transporting category 1 and 2 animal waste (articles 4 [2] and 5 [2]), as well as provisions on the organization of collection, deposit and processing (articles 10–15, 17–19) which are supplemented by national law. It also sets forth rules relating to the placing on the market of certain recycling products such as animal pet food, dog chews, certain technical products and organic fertilizers. As the regulation in general, these provisions aim to prevent animal pests and hygienic problems associated with recycling of animal side products, and are not geared to the specific problem presented by GMOs contained in the recycling products. However, it should be noted that products manufactured from transgenic animal side products are also subject to GMO-specific controls. With respect to products such as pet food, soap or wool, an authorization for placing on the market would be required under article 6 [9] of Directive 2001/18, even if the end product no longer contains GMOs (see below 8.7). Moreover, national product regulation may apply. Therefore, in this respect the regulation is, by and large, sufficiently protective of human health and the environment.

8.3 Development phase II: Animal protection Development activities for the generation of transgenic animals to be used as production platforms for pharmaceuticals invariably involve animal trials, at least in the initial stages of development. Therefore, the question arises as to what extent the existing regulation on animal protection might present obstacles and limitations to pharming development.

242

8 Legal problems of pharming

8.3.1 Sources of regulation There are a variety of legal sources of regulation in the field of animal protection and welfare. Three levels must be distinguished. Pan-European conventions, EC directives and national laws and regulations may be relevant for the protection of animals in the course of development activities in the field of pharming: – European Convention for the protection of vertebrate animals used for experiments and other scientific purposes of 1986, as amended by the Protocol of 1998 (in force since 2 December 2005), – European Convention for the protection of animals kept for farming purposes of 1976, as amended by the Protocol of 1992 (not yet in force), – Protocol on protection and welfare of animals, annexed to the Treaty of Amsterdam amending the EU Treaty and the EC Treaty of 1997, – EC Directive 86/609 on animal trials as amended by Directive 2003/65, – EC Directive 98/58 concerning the protection of animals kept for farming purposes, – national laws on animal protection and welfare in general and protection of animals used for experiments and other scientific purposes, such as the German Animal Protection Act of 2006 and the Regulation on the Keeping of Useful Animals of 2006, the British Animal Welfare Act 2006 and the Animals (Scientific Procedures) Act 1986, and articles R214-1 et seq. of the French Code rural on the protection of animals (codified in 2000). The scope of application of these legislative texts is quite different. One category comprises the regulation of animal trials. These texts may be relevant in the development phase of pharming. However, it will be seen that the delimitation between an animal trial and production is not easy to draw. A second category deals with the keeping of animals, either specifically or as part of more general rules. Theoretically, these texts could be applicable to the protection of animals during part of the pharming development and, which will be discussed later, the whole production process. However, subject to some qualifications such as the German Animal Protection Act, as a rule there is no comprehensive regulation that covers all pharming production operations during the development and later manufacturing stages.60

8.3.2 Animal trials: Scope of application of the relevant laws The most extensive regulation exists in the field of animal trials, especially relating to vertebrates, and yet their relevance to pharming activities is open to some doubt. The reason for this is that the protection of experimen60

The reason is that pharming animals are not considered as being used for farming purposes.

8.3 Development phase II: Animal protection

243

tal animals is limited to experiments using animals and does not extend to production activities. This raises the fundamental question as to whether and to what extent the various steps to be taken in developing medicinal products from transgenic animals can be considered as experiments or as normal development, breeding and manufacturing activities. The conventional distinction between animal trials, on the one hand, and normal breeding and production activities, on the other, is that a trial is designed to generate new knowledge, while activities that focus on breeding or production belong to the production stage. However, in the case of pharming this distinction is difficult to make, the more so since all definitions provided in the legal texts are more or less circular. The European Convention on animal trials defines trials (called animal procedures) as “any experimental or other scientific use of an animal which may cause it pain, suffering, distress or lasting harm”, starting with the preparation of the animal and the time when no further observations are made for that procedure (article 1 [2] [c]). This definition is resumed in article 2 of Directive 86/609. Conversely, Directive 98/58, relating to the protection of animals kept for farming purposes, excludes experiments on laboratory animals from its scope of application (article 1 [1] [c]). Similar descriptions of the scope of application of animal welfare laws can be found in national law. In the specific context of pharming, one could sustain that the activities aimed at the mating of donor animals and insertion of the gene construction into the fertilized eggs of these animals, as well as their transfer to recipient females (foster animals), are based on known procedures and, as such, do not constitute experiments but rather are essentially breeding and production activities. The following stages of selecting the individual or individuals from the offspring that are a copy of the donor animal’s transgenic eggs and express the inserted gene sequence (founder animals), as well as the propagation of the founder animals, could be regarded as resembling so very closely normal breeding procedures that they cannot be classified as a process whose focus is the generation of new knowledge. Furthermore, the initial generation of crude bulk material, and its processing for manufacturing a developmental medicinal product, could be denoted as a normal production process. It is clear that the final production of vaccines and antibodies from transgenic animals following the authorization of the medicinal product is a production process, even if associated with interventions in the animals.61 However, one must make a clear distinction between the development and production stages of pharming. As expressly spelt out in French law (article R214-87 Code rural), applied research constitutes a part of research. There are a number of uncertain61

Bundesministerium für Landwirtschaft, Ernährung und Forsten 2001:87; Caspar 1999:436; Goetschel, in Kluge 2002:§ 10a Nos. 2–4; Lorz and Metzger 1999:§ 10a No. 6.

244

8 Legal problems of pharming

ties associated with creating transgenic animals and developing their offspring to the stage of production. Even if the methods of transfer of the relevant gene sequences into fertilized eggs of the donor animals and their transfer to recipient females (foster animals) are known, there are no standardized methods. There is limited knowledge on the expression of the relevant gene constructs in founder animals. The outcome of the process is uncertain (see section 2.3). Therefore, the development process entails a fair degree of applied research. This is true of the gene transfer to the eggs of the donor animal as well as their reproduction in a suitable founder animal and in the latter’s direct offspring. Therefore, good arguments militate for the proposition that the development stage can still be considered to constitute a series of animal trials. This seems to be the position of German and British practice, which treats the creation of transgenic animals up to the second generation of offspring (founder animal and two generations of progeny) as animal trials.62

8.3.3 The European Convention and Directive 86/609 The existing regulatory texts on the protection of animal trials can be classified into two different categories, namely the first and second generation of animal protection laws. The European Convention of 1986, as amended in 1998 (in force since 2005), already sets forth the principle of justification when deciding whether an animal trial has to be performed. Moreover, at least in essence it establishes the so called 3-R concept (replacement, reduction, refinement)63, to be used when deciding on how a trial will be carried out. Justification means that only certain purposes are admissible for vertebrate animal trials, and that the trial must be indispensable in order to achieve this purpose. Directive 86/609 (as slightly modified by Directive 2003/65), among others, was adopted for implementing the European Convention. More demanding than the European Convention, the directive does not only require that animal trials entailing serious pain and suffering be specially justified. It further limits animal trials under these circumstances by prescribing a rudimentary risk-benefit evaluation, whereby the trial must be sufficiently important for the fundamental needs of man or animals (article 21 [2]). Recently, the Commission has published detailed guidelines for the accommodation and care of animals used for experiments and other scientific purposes.64 Although the directive is based on the harmonization competence under former article 100 EC Treaty (now article 95 EC Treaty), it 62

63 64

Bundesministerium für Landwirtschaft, Ernährung und Forsten 1997:110; Goetschel, in Kluge 2002:§ 7 No. 8,22,28; Müller-Terpitz 2007:84; AEBC 2002:56 (minimum of two generations of offpsring); see also House of Lords Select Committee 2002:16/17. Russel and Burch 1959. Recommendation of 18 June 2007 (2007/526).

8.3 Development phase II: Animal protection

245

only aims for minimum harmonization. The maintenance or introduction of more stringent member state law is permissible. This explains why, under the umbrella of the directive, a second generation of national laws for the protection of trial animals could develop. A negative – at least partly negative – effect of minimum harmonization is that there have not been strong pressures for the modernization of the directive.

8.3.4 National law National law, insofar as it is more stringent, is neither pre-empted by the European Convention nor, as stated, by Directive 86/609. It embodies, in many respects, more modern concepts of animal welfare. However, it should be noted that the focus of animal welfare law is on animal trials for the testing of chemical substances. Pharming activities are not of central importance, and there has been practically no regulatory experience in this field. Germany. In Germany, animal protection and welfare has a general constitutional status. Article 20a Federal Constitution (“Grundgesetz”) declares the protection of animals to be a task of the state. This ethical imperative is addressed to all state powers, in particular the legislature, although it does not establish subjective rights. A major legal effect of the state obligation to protect the animals consists of an enrichment of broad statutory terms, the performance of prescribed weighing of conflicting concerns and other exercise of discretion granted to the authorities. However, it is important to note that article 20a Federal Constitution does not afford absolute protection but, rather, only requires responsible treatment of animals.65 The German Animal Protection Act in its new version of 2006 describes the purpose of the law, in ethical terms, to protect life and welfare of animals in “responsibility of man for the animals in their capacity as co-creatures”, but, in an anthropocentric stance, goes on to prohibit the infliction of pain, suffering or harm to animals without a reasonable cause (section 1).66 Sections 7–9 of the law contain more specific provisions on animal trials, so that the general justification requirement under section 1 is not relevant here. The notion of animal trial, and in particular trial on vertebrates, under the act is not limited to live animals but also comprises genetic material. It is defined so as to include interventions in, and treatment of, animals or genetic material for trial purposes which are associated with pain, suffering and harm to these animals or, in the case of genetic material, pain, suf65

66

Federal Administrative Court, BVerwGE 101:l,37; Administrative Court of Appeal Mannheim, Natur und Recht 2006:111; as to the concept of dignity of animals see Richter 2007:321 et seq.; Teutsch 1995. Pain and suffering does not include mere discomfort; Federal Administrative Court, in: Buchholz 2000, Entscheidungssammlung des Bundesverwaltungsgerichts, 418.9 Tierschutzgesetz No. 11; Federal Supreme Court, Neue Juristische Wochenschrift 1987:1833,1834.

246

8 Legal problems of pharming

fering or harm to the genetically modified animals or their founder animals (section 7 [1]). In the case of genetically modified animals, breeding up to the second generation after the founder animal is considered an animal trial.67 Animal trials need to be justified by a variety of particular purposes described in the law, among others the prevention, determination or treatment of disease, suffering, physiological harm and distress, the testing of substances and fundamental research. The performance of an animal trial needs to be indispensable for achieving this purpose. In deciding whether this is the case, the state of scientific knowledge must be considered and it must be ascertained whether the purpose cannot be reached by using non-animal methods (section 7 [2]). Apart from this general justification requirement, the law prescribes an ethical justification whenever the trial is to be performed on vertebrates. This goes beyond Directive 86/609. Vertebrate animal trials may only be carried out when the expected pain, suffering or harm caused to the animals is ethically justifiable in relation to the purpose of the trial. Where such animal trials lead to longer lasting or repetitive considerable pain, suffering or harm, the review standard is strengthened. There must be reason to believe that the envisaged results will be of outstanding importance for the essential needs of humans or animals, including the solution of scientific problems (section 7 [3]). This is a severe restriction of animal trials, because it practically limits such animal trials to the prevention, determination and treatment of serious diseases.68 However, as a rule, it does not concern pharming operations as they are not normally associated with the degree of pain, suffering or harm that the law has in mind. Other restrictions on animal trials are formulated as requirements relating to the operative performance of animal trials, especially those on vertebrates. In keeping with the 3-R concept, such trials must be limited as much as possible and the state of scientific knowledge respected. Vertebrate animal trials are only permissible where trials on other animals are not sufficient for the envisaged purpose, the minimal number of animals is used and pain, suffering and harm are only inflicted on the animals insofar as this is indispensable for the purpose of the trial (section 9 [2]). The requirement of ethical justification engenders a balancing process between the “costs” incurred by the animal trial, in terms of pain, suffering and harm caused to the animals, and the expected benefit to be derived from the trial. Relevant cost factors are also the high mortality of trial animals and the generation of “excess” animals in the trial that have to be killed, especially when using micro-injection methods. The exact contents of the balancing process are controversial69 and the criteria for deciding on 67 68 69

See supra note 62. See Goetschel, in Kluge 2002, § 7 No. 45. Goetschel, in Kluge 2002:§ 7 Nos. 48 et seq., 43; Kloepfer 2004:§ 11 Nos. 313,343.

247

8.3 Development phase II: Animal protection

the right balance are uncertain. In this respect, some authors refer to the “social morality” of the population70 which provides a conduit for public attitudes about animal welfare, although, in contrast to section 5 [c] of the Norwegian Animal Welfare Act, ethical reactions of the public alone do not justify a prohibition of an animal trial. Besides, for making the necessary value judgement on admission or prohibition of an animal trial operative, one uses tables that classify the costs and benefits into three different categories ranging from low to medium to high, and attribute to them negative or positive decisions.71 Table 8.2: Illustration of classification of costs and benefits Benefit low medium high + = permissible

low + + - = prohibited

Cost (impairment of animals) medium +

high +/-

+/- = controversial72

However, the heuristic value of such classifications seems to be limited. It is clear from the text of the German Animal Protection Act, especially its inclusion of interventions into genetic material in the definition of animal trial, that the provisions relating to justification, ethical review and the 3-R concept also apply to animal trials in the course of generating and propagating transgenic animals for the production of pharmaceuticals. One also sustains that from a policy point of view, the existing law is sufficient.73 An important issue is whether, and to what extent, cost savings ultimately expected from the results of an animal trial can be listed on the positive side of the balance. Section 9 [2] No. 3 of the German Animal Protection Act expressly excludes any cost argument with respect to the performance of an animal trial as an element of the evaluation of indispensability for the intended purpose. Pain, suffering and harm may not be inflicted on an animal in order to achieve savings in work, time or costs. From the systematic position of this provision in the act, one might conclude that it only relates to the question as to how an animal trial is performed and not to the ethical review that scrutinizes the question as to whether an animal trial may be permissible at all, in other words that it only concerns trial costs 70 71 72 73

Kluge 1994:871. See Bundesamt für Veterinärwesen 1994; Goetschel, in: Kluge 2002, § 7 No. 60; Maisack 2007:197, 2002. In favour of a differentiation within this category Scharmann and Teutsch 1994:191; De Cock Buning & Theune 1994:107. Lorz and Metzger 1999:§ 7 Nos. 14 et seq.; Gruber and Kolar 1997:373 et seq.; Gruber 1995:239.

248

8 Legal problems of pharming

rather than cost savings achieved by exploiting the results of the trial. However, there is some authority to the contrary, although relating to the general justification clause set forth in sections 1 and 2 of the act.74 Even if one adhered to this broad interpretation, this would only exclude the consideration of individual cost savings of the operators, not economic cost savings of society as a whole. In the market for medicinal products, rapid availability of new products and a cheap supply to social health care systems are, in particular in societies with an aging population, important and legitimate factors on the positive side of the cost-benefit balance that should not come under the verdict of mere costs savings for the operator.75 Moreover, it does not appear permissible to look at the “purpose behind the purpose” when deciding on indispensability. Innovation and competition in the market, through the development of new pharmaceuticals, is considered as legitimate purposes. The question as to whether a new pharmaceutical, whose development requires animal testing, is needed on the market is not a relevant question in this context, even if this development “only” serves the ultimate purpose of saving costs (nor is it under pharmaceuticals regulation; see section 8.5.3). In order to establish an animal production platform, animal trials are indispensable. Therefore, there is no cogent reason to deal with this production method in a way that is fundamentally different from normal development operations. At best, the broader perspective – the purpose behind the purpose – can be considered in the ethical review process, and here the considerations relating to societal costs appear entirely legitimate. Of course, based on a need analysis, the legislature could ban animal trials in the development process entirely – as has been done with respect to cosmetics. However, such a fundamental decision should not be taken by the executive and the courts through the interpretation of existing law but rather politically, and it could not be confined just to pharming products. Another open question relates to the relevance of animal welfare risk analysis. The definition of animal trial (section 7 [1] German Animal Protection Act) refers to the mere possibility that a trial may be associated with pain, suffering or harm to the animal. However, in the framework of ethical review, Section 7 [3] of the act partly refers to “expected” pain, suffering and harm, and partly seems to require certainty about the future causation of pain, suffering and harm by an animal trial. If one considers the latter formulation which concerns trials on vertebrate animals which cause 74

75

District Court Hamburg, 313 O 565/00, not published (a case based on the law of unfair competition that, due to an amendment of the Act on Unfair Competition in 2004 [section 4 No. 11], could no longer be brought to such a court); in the same sense Goetschel, in Kluge 2002, § 10a No. 7; Maisack 2007:176–179,221/2,234; Caspar 1999:455; Müller-Terpitz 2007:78/79,83 (who, however, suggests an exception in case of prohibitively high costs for patients); contra Lübbe 1994:472. District Court Düsseldorf, Recht der Landwirtschaft 1980:189,191; contra Maisack 2007:222 with respect to food.

8.3 Development phase II: Animal protection

249

longer lasting or repetitive pain or suffering as inaccurate, as suggested by the context of the clause, the standard of knowledge is that of “danger” in the conventional German categorization of risk.76 This means that, while certainty is not necessary, there must be sufficient probability, based on past experience or other reliable evidence, that the trial will be associated with pain, suffering or harm. The precautionary principle does not apply. In view of wide-spread uncertainty about adverse effects caused by animal trials, including those that are inflicted in pharming development, for example uncertainties about harmful effects of insertion and transgene expression, this limitation of ethical review that excludes uncertain, but plausible future harm short of sufficient probability appears questionable. The Act establishes a general requirement of a project authorization for any trial involving vertebrate animals (section 8). The authorization may only be granted under certain prerequisites, among which the following are the most important ones: The applicant must have demonstrated in a scientific manner that the prerequisites of indispensability and ethical justification are fulfilled and that the envisaged results of the trial are not sufficiently known despite exhaustion of accessible means of information or need to be reviewed, and it can be expected that when performing the trial the 3-R concept is complied with (section 8 [3]). The burden of proof and proffering evidence is placed on the applicant in a differentiated, somewhat confusing way ranging from full proof to mere demonstration (section 8 [2]). Demonstrate in this context means that the applicant must give substantiated and plausible information to support the application.77 The differentiation made by the Act reflects both the knowledge available at the time of submitting the application, and the necessary knowledge base for proof of the authorization prerequisites (section 8 [2]). Apart from the role of the applicant, it is also doubtful what degree of scrutiny – plausibility control vs. in-depth review – has to be employed by the competent authority in processing the application. Even within the judicature, there is no uniform opinion. It has often been sustained that the authority may only carry out a qualified plausibility control, but may not substitute its judgement for that of the scientists who perform the animal trial.78 Others plead for a more extensive review of the application, in which plausibility review is limited to the purpose of the trial while all other authorization prerequisites are fully reviewed.79 Still others take the view that at least after the insertion of animal protection in article 20a Federal 76 77 78

79

In this sense Kloepfer 2004:§ 11 No. 332. Lorz and Metzger 1999:§ 8 No. 12. Federal Constitutional Court, Neue Juristische Wochenschrift 1994:869; Administrative Court Berlin 1995 Zeitschrift für Umweltrecht:201; Kloepfer 2004, § 11 No. 130. Federal Administrative Court, Entscheidungen des Bundesverwaltungsgerichts (BVerwGE) 105:73,82; Caspar 1999:460/61; Kluge 1994:870.

250

8 Legal problems of pharming

Constitution, full scrutiny of the authorization prerequisites, including the purpose of the trial, is warranted.80 A decision of this controversy is not easy. German administrative procedure law is based on the principle of full agency review of the prerequisites for an authorization, unless a statute prescribes another standard. However, this seems exactly to be the case under the act, insofar as section 8 [3] only requires the applicant to demonstrate the fulfilment of certain authorization prerequisites, especially those that involve scientific judgement or a prediction of future behaviour. Article 20a Federal Constitution protects animal welfare only subject to legislation. In a more recent law relating to stem cell research, the legislature has expressly provided for a mere plausibility control. There is no cogent reason to reinterpret the Animal Protection Act in the light of the Constitution. The law also mandates the nomination of one or more officers for animal protection in all facilities where animal trials on vertebrates are performed (section 8b [1], [2]). The officer for animal protection is obliged to pay attention to compliance, advise staff responsible for performing animal trials and keeping trial animals, comment on applications for the authorization of such trials, and promote the introduction of non-animal trial methods (section 8b [3]). The establishment of internal ethical commissions is not prescribed, except for public research facilities such as universities where state law or internal regulations provide that such commissions must be established. Ethical review is, in principle, considered a matter of administrative control rather than self-regulation. Section 15 [1] subparagraphs 2, 3 provide for the establishment of animal protection commissions inside the administration, whose task is to advise the state agencies on decisions on applications for authorizations. Apart from requirements as to qualification of the members of these commissions, the law mandates a fairly high representation of animal protection interests. In the development phase of pharming, the breeding prohibitions set forth by section 11b Animal Protection Act may also be relevant. The modification, by bio-technological or biological means, is prohibited where it must be expected that the animals or, for genetic causes, their offspring lack parts or organs for proper functioning or these parts or organs are unsuitable or modified in such a way that pain, suffering or harm occurs. The same is true where, in the offspring for genetic causes, behavioural disturbances occur that are associated with suffering. In these cases, the competent authority can order sterilization of the animals. In pharming, these prohibitions may become relevant, especially with respect to the propagation of the donor animal. United Kingdom. In the United Kingdom, the Animal (Scientific Procedures) Act 1986 is largely based on the European Convention, but contains 80

Administrative Court Gießen 2004, Natur und Recht:64; Goetschel, in Kluge 2002, § 7, No. 32a; § 8 Nos. 9 et seq.; Maisack 2007:168–170,357.

8.3 Development phase II: Animal protection

251

a number of innovative elements. The definition of animal trial (section 5 [3]) follows the definition given by the convention. At the core of the act, as part of the prescribed authorization procedure, is a cost-benefit analysis whereby the competent authorities, when deciding whether and on what terms to authorize the project, weigh the likely adverse effects on the animals concerned against the benefits likely to accrue as a result of the programme (section 5 [4]). Furthermore, the 3-R concept is laid down in the law. An animal trial is only permissible where its purpose cannot be achieved satisfactorily by other reasonably practicable methods not entailing animal trials (section 5 [5] [a]). In performing the trial a method must be employed that uses minimum numbers of animals and uses vertebrates with the lowest degree of neuro-physiological sensitivity, causes the least pain, distress or lasting harm and is most likely to produce satisfactory results (section 5 [5] [b]).81 The act requires a project authorization for every animal trial (section 3). The restrictions on animal trials as described are prerequisites for granting the authorization for a project, and must be dealt with in the documentation accompanying the application. The burden of proffering evidence is described by the Home Office Guidance on the Operation of the Animals (Scientific Procedures) Act of 200082 as the need for the applicant to “demonstrate” fulfilment of the permit requirements. As in Germany, this means that the applicant must provide substantiated and plausible information to support the application. In making the decision on the application, the competent authority, the British Home Office and its Inspectorate, has a wide margin of discretion. According to the British practice, not only the cost-benefit evaluation but also the application of the 3-R concept entails a weighing of conflicting concerns. The transparency of the weighing process and the predictability of its outcome have been criticized as being low.83 The guidance document84 interprets the cost-benefit analysis as an optimization process, whereby the benefits must be maximized and the (social or external) costs minimized. The likely benefit is derived from the utility of the data or products to result from the programme of work. It relates to the progress likely to result directly from the programme. The guidelines do not give any indication that efficiency of the end production could be a relevant beneficial factor. Rather, in relation to safety testing, they specifically rule out that the utility or benefit of the end product could be taken into consideration. However, it is difficult to see how the utility of the data generated can be assessed if one ignores the benefits to be derived from 81 82 83 84

Radford 2001:297; House of Lords Select Committee 2002:37/38; AEBC 2002:34 et seq. 23 March 2000 Appendix I:10/11. House of Lords Select Committee 2002:30. Sections 2.44, 2.45 and Appendix I:10/11; see also Animal Procedures Committee 1997 Appendix F, chapter 2.

252

8 Legal problems of pharming

their use. Moreover, it is not clear whether the objective of creating a transgenic animal would be questioned at all, or constitutes the frame within which the cost-benefit evaluation operates. Consequently, the application of the British law on animal trials is confronted with the same kind of problem addressed in the context of German law. This means that, in principle, savings of individual costs at the level of the firm do not justify an animal trial,85 and even cost savings within the social insurance system may be irrelevant. In the context of the cost-benefit evaluation, cost is equivalent to social (external) cost. It is considered as the immediate or delayed adverse welfare effects (pain, suffering, distress or lasting harm) likely to be experienced by the animals used, as the consequence of the trial or the result of the care and handling systems. A better expression would be harm-benefit analysis.86 Cost-benefit analysis is not to be performed by mathematical calculations in the sense of monetization; rather, it is conceived as a qualitative assessment. Apart from the project license, the Animals (Scientific Procedures) Act 1986 also requires a personal license, which relates to the qualification of staff that takes part in animal trials. The keeping of trial animals is subject to the requirement of having appropriate animal accommodation and veterinary facilities which is confirmed by a certificate of designation (section 6). Moreover, every operator must, by virtue of a standard condition, appoint one or more animal care and welfare officers whose task is to ensure proper husbandry, care and welfare of the animals (section 10 [6B], Guidelines 4.48, 4.55–4.58). Codes of practice regulate the details. Besides the statutory controls, especially the requisite cost-benefit analysis and the 3-R concept, since 1999 the regulation of animal trials in Britain also operates through controlled self-regulation based on a local ethical review process. This procedure is not set forth by law. Rather, it has been introduced as a standard condition under section 10 [1] of the act. All operators that carry out animal trials must establish an ethical review committee, which may be composed of staff members but normally contains outsiders. The objective of the ethical review process is to provide ethical advice to the operator, promote the use of ethical analysis, increase awareness of animal welfare issues and develop initiatives for the widest possible application of the 3-R concept.87 This includes the encouragement of nonanimal alternatives for a projected license and consideration of the care and accommodation standards for trial animals. In practice, the ethical review process has developed into the second core element of animal protection in the field of research and development. 85 86 87

Board on Agriculture and Natural Resources 2002:51 et seq.; Boyd Group 1999; see also ECVAM 1898:21 et seq. In this sense House of Lords Select Committee 2002:30. Guidance document, supra note 81, appendix J No. 3.

8.3 Development phase II: Animal protection

253

In contrast to Germany, the United Kingdom relies somewhat less on administrative regulation. While ethical review is a part of the authorization procedure in Germany, it is based on self-regulation in the United Kingdom. However, in assessing the differences between the two countries, one must also consider that the cost-benefit analysis prescribed in the United Kingdom is part of what German law understands by an ethical review process. In a way, cost-benefit analysis and ethical review overlap, which has been criticized.88By and large, the actual authorization practice seems to be strict and rather bureaucratic. In particular, the 3-R concept is pursued strictly.89 Whether the often made claim, that the United Kingdom has the most stringent system of animal protection in the world,90 is justified could only be verified by an empirical investigation. France. In France, articles R 214-1 et seq. of the Code rural regulates animal protection in the course of animal trials. This is the former Regulation 87-848, as amended by the Regulation 2001-464, which has been codified in the regulatory part of the Code rural.91 Following the European Convention, article R214-87 permits animal trials for a variety of purposes. The prerequisites for admissibility of animal trials are formulated without a clear distinction between obligations of the operator or authorization prerequisites and required information to be supplied by the operator. Every project that involves an animal trial requires an authorization (article R214-93). In performing an animal trial, the number of animals must be kept to a strict minimum and the proposed trial be justified in the application for an authorization. Where the trial entails intensive or prolonged pain or suffering, or the risk of such pain or suffering, this must be expressly declared and justified (article R214-91 [2]). The application must be accompanied by a dossier with prescribed items. These items reflect the 3-R concept. However, an ethical justification in the strict sense is not provided. The applicant must justify the choice of the animals used for trial. He/she must establish that there is no alternative method available which could be substituted for the animal trial, and that the animals are the most suitable for the type of envisaged research (article R214-99 [2] No. 1). The justification requirements relate to the kind and the manner of the trial. The choice must be guided by the objective to minimize the number of animals, selecting the least sensitive from a neurophysiological point of view as well as selecting animals that present the best chances of deriving satisfactory results from the trial (article R214-99 [2] No. 2). However, in practice, the applicant must state the reasons for the use of the animals only in very broad terms. 88 89 90 91

House Select Committee 2002:34. Radford 2001:298/99; House of Lords Select Committee 2002:34/35, 37/38. See House of Lords Select Committee 2002:12. See Ziani 2006:425–441.

254

8 Legal problems of pharming

Although the law does not state this expressly, the prescribed contents of the application also serve as prerequisites for granting the authorization. The decision on the application is discretionary and can be restricted or granted subject to conditions (article R214-100). Therefore, the competent authority, namely the prefect of the department and its veterinary inspector, has the power to decide on the application, relying on a cost-benefit analysis or an ethical review even if this is not formally provided, and the applicant is not required to specifically demonstrate costs and benefits of the trial. Besides the project authorization, the law also requires a personal authorization which, in practice, contains elements of a project authorization (article R214-93). It is sustained in legal literature that the administration is not very sensitive to the concerns of the public about animal trials and, by and large, decides in favour of the applicants from industry and research institutions.92 The degree of controls seems to be low. The facilities where the trial shall be performed need a certification (agrément) (article R214-100 to 103). Breeding facilities must be notified to the prefect and also require a recognition (article R214-107). The keeping of the animals is subject to a general duty of care (article R214-17). Finally, the law establishes two national commissions in the field. One is the National Commission on Animal Trials (article R214-116), and the other is the National Committee for Ethical Reflection on Animal Trials which forms a part of the former commission (article R214-122). Both are consultation bodies of the competent Ministry. United States. In the United States, animal trials are covered by the Animal Welfare Act of 1966, as amended (7 U.S.C. §§ 2131–2159) and Department of Agriculture regulations (9 C.F.R. parts 1 and 2).93 Moreover, most large companies receive accreditation from, and are inspected by, the private Association for Assessment and Accreditation of Laboratory Animal Care (AALAC),94 which, however, focuses on animals used for the testing of hazardous substances. Apart from the requirement of a facility license, American law is dominated by the concept of controlled selfregulation. The Animal Welfare Act (7 U.S.C. § 2143 [b)[1]) prescribes that all research facilities (including industry) establish an Institutional Animal Care and Use Committee, composed of people selected on the basis of their experience and expertise, including outsiders. In its capacity as an agent of the firm, the committee must regularly review the animal welfare practice of the enterprise and needs to approve any animal experiment 92 93

94

Ziani 2006, supra note 91. To the extent that research is government-funded, the Policy on Humane Care and use of Laboratory Animals of the Public Health Service under the Health Research Extension Act may apply in addition; see House of Lords Select Committee 2002:12/13. House of Lords Select Committee 2002:13.

8.4 Development phase III: Protection of occupational safety and health

255

(9 C.F.R. § 2.31 [8] [d]. Compared with European standards, the prerequisites for approval are couched in rather weak terms. The Animal Welfare Act only refers to scientific necessity (7 U.S.C. § 2143 [a] [3] [D]). The regulations require avoidance or minimization of discomfort, distress and pain, the mere consideration of non-animal alternatives and written demonstration that they are not available, and the avoidance of duplications of animal trials. An ethical review process or cost-benefit analysis is not provided, and the implementation of the requirements within the enterprises seems to be lenient.

8.4 Development phase III: Protection of occupational safety and health in the development of recombinant medicinal products Since the development of pharming products will normally be carried out within an enterprise or a research facility by the staff, it also raises problems of occupational safety and health. The relevant problems are in general addressed by the two major Community regulatory texts on GMOs, namely Directive 90/219, as amended by Directive 98/81 (containment) and Directive 2001/18 (release without containment) as well as national law implementing these directives.

8.4.1 Contained use Directive 90/219 is designed to protect human health and the environment (article 1). As must be concluded from the general principles of containment and other protective measures set forth in the annex IV, this notion includes the protection of occupational safety and health. The annex contains a variety of measures for good occupational safety and hygiene. As regards concrete measures in the field of occupational safety and health, one can distinguish between information and operational obligations. The obligation of the operator to perform a risk assessment in order to determine the kind and magnitude of risks associated with the activity, and to classify it into one of the four risk levels (article 5 [2]), is also designed to protect occupational safety and health. A summary of this risk assessment must be communicated to the competent national authorities as part of the notification required for starting operations under article 7 Directive 90/219. With respect to category l activities the directive does not require more specific information on occupational safety and health. In contrast, the envisaged containment and other protection measures must be described in detail for the following safety categories (annex V). Pursuant to national law, the competent authorities can impose upon the operator of category 1 activities concrete obligations for the protection of workers (cf. article 5 [1]).

256

8 Legal problems of pharming

Apart from that, Directive 90/219 contains general and specific requirements for the protection of human health that in particular concern occupational health. For all activities involving GMMs, according to annex IV the principles of good microbiological practice and specified principles of good occupational safety and hygiene apply. Among others, the operator must minimize exposure to GMMs to the lowest practicable level, perform engineering control measures at the source of risk, provide protective clothing and equipment, test and maintain control measures, provide appropriate training, instructions and warnings, and formulate and implement local codes of practice. This catalogue of general obligations is specified by particular measures, which vary according to the relevant risk category. As regards Category 1 activities, there are only very few additional optional requirements which must be evaluated in the risk assessment (annex IV, General principles, No. 2). It is only beginning with category 2 that the Directive sets forth more demanding requirements. National law by and large has transposed the provisions of the directive relating to occupational safety and health literally but extended them to all GMOs. Apart from this important extension and from providing specific agency powers, it does not add much to it in terms of protection.

8.4.2 Release without containment By contrast, Directive 2001/18 and national law implementing the directive are almost silent about risks to occupational safety and health presented by releases of GMOs without containment. This does not mean that occupational safety and health are not subjects of protection at all. Rather, as in the case of Directive 90/219, the notion of health (articles 1, 4 [1] and [3]) includes occupational safety and health. This interpretation is confirmed by the notification requirements set out in annex IIIA that applies to animals. The applicant must, among others, include in the risk assessment effects on human health resulting from direct or indirect contact of persons working with the GMOs (annex II D.1 point 6) and provide information about workers’ protection measures to be taken during the release (annex IIIA point III A 8). As regards higher plants, the risk assessment also comprises occupational health (annex II D.2 point 6), while the applicable annex IIIB does not require supplying information on occupational safety and health. This gap may be explained by the assumption of the drafters of the directive that the problem is less relevant for the release of higher plants. However, when the competent authority has reason to believe that there might be a problem of workers’ protection in the case of release of transgenic higher plants in the course of pharming activities, it can request additional information (article 6 [7]). As can be concluded from article 4 [2] of the directive which describes the content of annex III as information that may be necessary to carry out the requisite risk assessment, annex III is not definitive.

8.5 Development phase IV: Regulation of development medicinal products

257

Insofar as the competent national authority, on the basis of information supplied by the applicant on his/her own or upon its request, sees a particular workers’ protection problem being presented by a release, which in particular may arise in case of consecutive releases (article 6 [9]), it can grant the authorization subject to conditions (article 6 [8]). In doing so, it may use the requirements set forth by Directive 90/219 as guidance. However, at least as regards pharming, more explicit language would be appropriate.

8.4.3 General regulation of occupational safety and health In addition, the general regulation of occupational safety and health established by Directive 2000/54 on the protection of workers against hazards by biological substances at work and implementing national regulations may apply, both with respect to development under containment and release without containment. This directive is a daughter directive under the umbrella of Directive 89/391 for the improvement of the safety of workers and protection of the health of workers at work. It also covers transgenic microorganisms but not all GMOs (article 2 [1]). Directive 2000/54 is merely supplementary. Equivalent or more stringent regulation set forth by directives 90/219 and 2001/18 directives is paramount (article 1 [3]). It may be assumed that the primacy of specific GMO regulation includes conditions attached to an authorization granted under the two directives. Apart from this, Directive 2000/54 is not of major relevance for typical potential health risks presented by GMM releases in the course of initial pharming development activities. The reason for this is that the system of risk classification established by Directive 2000/54 (article 2) is confined to the risk of infection from biological substances at work. In practice, one could think of the potential exposure of workers to pathogens contained in the cell cultures or viral vectors used in the process. However, this appears to be a marginal problem.

8.5 Development phase IV: Regulation of development medicinal products During the development stage, the operator also has to comply with the substantive and procedural requirements applicable to medicinal development products. These are set forth – by way of reference – by Directive 2001/83, as amended in particular by Directives 2003/47 und 2004/63, and various guidelines in the field95 as well as national pharmaceuticals law. The reason for this overlap between GMO and pharmaceuticals regulation of “upstream processing” is, on the one hand, that the marketing authorization 95

For example with respect to the product: EMEA 2000; ICH 1998a and other ICH guidelines such as: ICH 1998a; ICH 1997; ICH 1995; with respect to production: guidelines on good manufacturing practice (GMP); see also Schmitt 2004:33/34 and 47/48.

258

8 Legal problems of pharming

for medicinal products requires the submission of the results of previous analytical, toxicological and clinical testing.96 The developmental product needs to have an appropriate quality and stability and the production process must be carried out in certified facilities and comply with good manufacturing or agricultural practice (GMP or GAP) in order to generate test results that are valid also for the final product. On the other hand, ethical considerations play a role in deciding whether the development product can be subjected to clinical testing. The product must have a sufficient degree of biosafety and the testing operations must comply with good clinical practice (GCP). The commencement of clinical testing requires an authorization (article 9 [4] and [7] Directive 2001/20). This is granted, after consultation of a regional or national ethics committee, under national law by the central authorities of the member states that are responsible for pharmaceutical safety. Similar requirements apply in the United States. With respect to pharming development, the regulatory problem is not the protection of the environment against risks associated with the release of GMOs, but rather the protection of the production process and the product against risks originating from the environment, such as infectious and viral agents that may spread to the production premises from the outside. This is especially true of animal pharming. However, the required safety measures are not necessarily such that options for development operations that are available under GMO regulation, especially an open release or even a contained use for class 1 GMOs with double fencing, cannot be used in practice. They affect more the safety and control measures to be taken regarding the gaining and processing of the crude bulk material and the manufacturing of the developmental preparation. Since the requirements to be observed are, by and large, similar or closely related to those relevant for securing the marketing authorization, they shall be addressed under 8.6.

8.6 Market authorization phase 8.6.1 Regulation 726/2004: Objectives and scope of application Placing recombinant pharmaceuticals on the market is regulated by EC Regulation 726/2004 laying down a Community procedure for the authorization and supervision of medicinal products for human and veterinary use and establishing a European Medicinal Agency. Regulation 126/2004 refers 96

Biomedical testing of the development medicinal product does not normally entail a release of GMOs because the finished product no longer contains, or consists of, GMOs. These have been removed from the bulk material during the manufacturing process. It is only in rare cases that the finished product generated in pharming would contain, or consist of, GMOs. This aspect of pharming can be neglected.

8.6 Market authorization phase

259

to Directive 2001/83, as amended in particular by Directives 2003/63 and 2004/27, with respect to quite a number of questions such as definitions, the documentation to be supplied with the application, labelling and the production process.97 National law plays a major role in the production process for medicinal products. The purpose of the authorization procedure is to protect human health and the health of animals as well as the environment (article 6 [2] subparagraph 2). As evidenced by the prerequisites for granting the authorization and recital 7 of the Directive 2001/83 which refers to potential risks, the regulation is guided by the precautionary principle. The quality, safety and effectiveness of the medicinal product must be cumulatively proven in a proper and sufficient manner. The burden of proof is placed on the applicant (article 12).98 Regulation 726/2004 applies to a limited number of categories of pharmaceuticals contained in the annex to the regulation (article 3 [1]). All medicinal products developed by recombinant DNA techniques, controlled expression of genes coding for biologically active proteins in prokaryotes and eukaryotes including mammalian cells, and hybridoma and monoclonal antibody methods are listed in the annex and hence are subject to the centralized authorization procedure. These products are defined by the manufacturing process. The recombinant DNA is not required to still be present in the finished preparation. All techniques presently used in pharming operations are encompassed.

8.6.2 Special regime for recombinant pharmaceuticals? The inclusion of pharmaceuticals developed by means of biotechnological processes in the centralized authorization procedure suggests that such products should be subject to a special procedure designed to ensure the control of typical risks associated with this process. Indeed, the regulation establishes a special regime for medicinal products that contain or consist of GMOs. In an application for authorising such a medicinal product, the consent to the deliberate release for research and development purposes under part B of Directive 2001/18, the complete technical dossier under annexes III and IV, and the environmental risk assessment performed in accordance with the principles set out in annex II, must be provided (article 6 [2]). The environmental safety requirements laid down by Directive 2001/18 must be respected in the evaluation of the application (article 6 [3] subparagraph 4). However, in contrast to the regulation of food and feed (Regulation 1829/2003), the special regime does not apply when the pharmaceutical is merely produced by a biotechnological procedure (or from GMOs) but 97 98

Thereby the directive has the legal status of a regulation. European Court of First Instance 2002 ECR II 3305 Nos. 114,135 et seq. – Pfizer Animal Health; 2002 ECR II 4945 Nos. 181 et seq. – Artegodan; Blattner 2002:281/282; Lorenz 2006:167.

260

8 Legal problems of pharming

does not contain, or consist of, GMOs. Rather, these products are subject to the normal central authorization procedure in the framework of which special problems regarding safety and quality that may be associated with the manufacturing method have to be addressed.99 In the case of pharming, the crude bulk material gained from the transgenic herd is purified so that the finished active substance, and hence the preparation, no longer contains any organisms that are able to replicate or transfer genetic material (see the definition of organism in article 2 [1] Directive 2001/18). An extensive interpretation of article 6 [2] of the regulation is not warranted since, as will be shown, the possible effects of the biotechnological procedure of deriving the active substance such as impurities, contamination with host cell contaminants or risks of viral infection can be addressed in the normal authorization procedure. There are guidelines on recombinant pharmaceuticals issued by the European Medicines Agency (EMEA) in 1995100. As they are outdated in view of new scientific knowledge about possible quality problems and adverse effects, EMEA is preparing new guidelines.101 The lack of up-to-date guidelines means that the producer needs to use informal contacts to EMEA in order to receive sufficient and reliable information about the pertinent requirements, especially as regards the requisite product safety and quality of the medicinal product and the clinical trial programme. Pre-submission meetings with EMEA are common practice.102 While problematic from the point of view of legal certainty, this procedure has the advantage that hand-tailored solutions can be devised for the development of a particular pharming product and for preparing the submission of an application for that product. In the United States, medicinal products derived by recombinant DNA technology are in principle treated like conventional biological pharmaceuticals. The Federal Food and Drug Administration (FDA) is of the opinion that specific differences as to safety and quality can be addressed in the framework of existing regulation.103 Transgenic pharmaceuticals have to be authorized as biologics under part V of the Federal Food, Drug and Cosmetic Act and implementing regulations (21 C.F.R. parts 58, 210, 211, 600 and 680). There are FDA guidelines relating to the manufacture and testing of pharmaceuticals derived from transgenic animals104 which are outdated and presently under consideration. Where the medicinal product contains, or consists of, GMOs, an environmental impact assessment under 99 100

101 102 103 104

See EMEA 2007. Production and Quality Control of Medicinal Products Derived by Recombinant DNA Techniques, Guidance document 3AB1a, 1995; Use of Transgenic Animals in the Manufacture of Biological Medicinal Products for Human Use, Guidance document 3AB7a, 1995; see Schmitt 2004:51. As a first step not relevant here see EMEA, supra note 99. Schmitt 2004:35,39,48,51; Schneider 2003:96 et seq. FDA 1986. FDA 1995; FDA 1996.

8.6 Market authorization phase

261

the National Environmental Policy Act must be performed (21 C.F.R. part 25).105 However, this is not relevant with respect to the normal medicinal products derived from transgenic plants or animals.

8.6.3 Authorization prerequisites and procedure The basic prerequisite for securing an authorization for pharmaceuticals derived from DNA technology is that the applicant must prove in a proper and sufficient manner the quality (consistency and purity), safety and effectiveness of the product (article 12 [1] Regulation 726/2004). According to general opinion, this entails a risk-benefit evaluation and a corresponding weighing, whereby certain risks to human health or uncertainties as to such risks presented by the new product can be compensated for by the therapeutic value of the product and the expected benefits from using it, taking into account the kind and extent of undesirable side effects, the severity of the illness to be combated, and the therapeutic alternatives.106 The wording of the provision does not support this interpretation. However, recital 14 of the regulation refers to a risk-benefit evaluation. Moreover, arguments based on consistency militate for such an interpretation. The parallel Directive 2001/83 which regulates the decentralized procedure for medicinal products for human use107 expressly provides for a risk-benefit evaluation (article 1 Nos. 1, 28a, article 26 [1] [a]). It would be odd that pharmaceuticals should be treated differently depending on the attribution of competences either to the member state authorities or the Commission. Finally, even more cogently, the requirement of a risk-benefit evaluation must be derived from article 16 [2] subparagraphs 2 and 3 of the regulation. This provision sets forth obligations of the holder of the authorization to provide information with respect to new knowledge that might influence the evaluation of the risks and benefits of the product. Furthermore, it empowers EMEA to ask the holder of the authorization to forward any new data demonstrating that the risk-benefit balance of the new drug remains favourable. This provision would not make sense if the decision on the original authorization were not based on a risk-benefit evaluation. The Commission and the expert bodies preparing the decision have a wide margin of discretion when determining whether the authorization prerequisites are fulfilled, in particular when performing the risk-benefit evaluation, in view of the scientific complexities that have to be addressed.108 105 106 107

108

Example: the authorization of a transgenic variant of the hormone BST, 58 Fed. Reg. 59946 (1993). Lorenz 2006:167/168. Directive 2003/82 relating to medicinal products for veterinary use will not be considered as the regulation essentially equals that on pharmaceuticals for human use. European Court of Justice 1999 ECR I-223 No. 34 – Upjohn; European Court of First Instance, 2002 ECR II-4945 No. 201 – Artegodan; Lorenz 2006:169.

262

8 Legal problems of pharming

However, an evaluation of the need for a particular new medicinal product is not required and not even permissible.109 Although perhaps desirable as a matter of policy, article 12 Regulation 726/2004 contains no language from which one could seriously conclude that something like a need analysis could be added to the considerations that the competent agency must perform in the authorization process. This is also true of the consideration of therapeutic value, which does not allow a comparison with products that are already on the market. The only legal basis for such an analysis would be the discretion afforded the Commission by article 12 of the regulation. However, this discretion must be exercised in coping with the objectives of article 12, which does not indicate that beyond quality, safety and effectiveness other factors, such as the need for the medicinal product, could be relevant. The rejection of a need analysis under existing law also concerns the production method. The Commission cannot deny an application for a marketing authorization for a pharming product because it feels that, in view of existing conventional production alternatives, there is no need for such products. The member states could try and refuse to reimburse prescriptions of the pharmaceutical by their social security systems, but this is not very probable if one accepts the basic premise of pharming that it generates cheaper products. Under regulation 726/2004, the application must be accompanied by information and documentation as required under Directive 2001/83, as amended in particular by Directives 2003/63 and 2004/27, applicable to the decentralized authorization procedure, namely articles 8 (3), 10, 10a, 10b, 11 and annex I Directive 2004/27. This mainly concerns manufacturing methods, control methods and results of physical-chemical, biological or microbiological, toxicological, pharmaceutical, and clinical testing. Moreover, the application must contain information about, and an assessment of, potential environmental risks associated with the use of the medicinal product (article 8 [3] [ca] and [g] Directive 2004/27). However, environmental effects are not considered in the risk-benefit analysis (recital 28 Directive 2004/27); they can only give rise to conditions for using the product.110 Directive 2003/63 contains modifications to the dossier which accompanies applications for various categories of biological pharmaceuticals such as human vaccines, insulin, herbal pharmaceuticals and gene therapeu109 110

Recital 34 Regulation 726/2004; Commission, Draft regulation, COM 2002, 735 final:13/14,27; Lorenz 2006:172/73; Kwizda 1998:66. It is doubtful whether a refusal of the permit is (in contrast to medicinal products for human use) possible with respect to pharmaceuticals for animal use. Article 1 Nos. 19 and 20 Directive 2004/28 expressly includes risks to the environment in the definition of the requisite risk-benefit analysis. However, article 37 Regulation 726/2004 which sets forth the prerequisites for granting the permit only speaks of “safety” of the product and recital 23 of Directive 2004/27 seems to limit the legal consequences of the environmental assessment to special requirements for limiting releases of active ingredients into the environment.

8.6 Market authorization phase

263

tics. Moreover, it generally regulates the requisite information on biological pharmaceuticals which are essentially equal to medicinal products already authorized (biogenerics) by reducing the dossier requirements (annex I part 1). However, medicinal products generated by pharming cannot be considered as essentially equal to conventional medicinal biologics already authorized since equivalence is defined both by product and process (article 10 [1] [a] [ii] Directive 2001/83).111 Genetic modification constitutes a new process and therefore the applicant cannot claim any procedural alleviations.112 Of course, equivalence may be established in appropriate cases between existing and new transgenic pharmaceuticals.113 Whether the normal authorization prerequisites and procedures adequately respond to the specific problems presented by pharming drugs may be questioned. The genetic modification of the relevant animals, especially the generation of recombinant protein, as such is tackled practically by purification of the crude bulk material. However, the generation of transgenic animals may lead to instability of the gene construct and alterations in the metabolism of these animals; the recombinant protein may have impurities and be contaminated with host cell contaminants and infectious viruses. An example in this respect is presented by the authorization procedure relating to the transgenic drug ATryn®, which was concluded in August 2006 with a positive decision.114 In this case, concerns were raised that altered glycosylation patterns in recombinant proteins produced from animals could alter pharmaceutical properties and that there was a potential of immunogenicity. However, it would seem that such problems can, as this first case of a regulatory procedure relating to a pharming drug shows, be addressed in the normal authorization procedure. The safety of the active substance and the preparation can in particular be documented by the results of the toxicological, pharmaceutical and clinical tests. In order to submit the developmental medicinal product to such tests, certain requirements relating to safety and quality must be observed in order to ensure the transferability of the tests and compliance with good clinical practice, including the ethical justification for exposing trial persons to clinical testing. In this respect, besides Directive 2001/83, as amended by Directive 2004/27, Directive 2001/20 and implementing national laws apply.115 Although fundamental differences between conventional and pharming drugs do not exist with respect to safety, certain problems have been identified in the literature that are specific to animal pharming, such as the risk of 111 112 113 114 115

See also EMEA 2005:point 4. Schmitt 2004:35,38/39. See EMEA, supra note 111. EMEA, Press Release of 2 June 2006, EMEA/203163; EMEA/CHMP, Report of the Plenary Meeting, May 2006; see also Schmitt 2004:48. See supra 8.5; Schmitt 2004:33/34,35,47/48.

264

8 Legal problems of pharming

novel infectious diseases and immunogenicity.116 Moreover, the assurance and maintenance of the quality of the medicinal product raises pharmingspecific problems. The quality of the medicinal product is an authorization prerequisite (article 6 [2] Regulation 726/2004) that is closely linked both to efficacy and safety. Article 8 [3] [c] and annex I part 2A Directive 2001/83 require the description of the composition of the product according to kind and quantity of all components. The applicant must also provide, together with the application, information about the methods of manufacture and control (article 8 [3] [d] and [h] and annex 1 part 2B Directive 2001/83). There are various guidelines issued by EMEA in this field.117 The stability of the gene construct, the purity of the recombinant protein, as well as the homogeneity of the active substance are important factors in the quality assessment. Various guidelines issued by the International Conference on Harmonization (ICH) and EMEA guidelines are applicable in this area. The stability of the gene construct must be investigated and demonstrated over several generations; variation must be within an acceptable limit. As regards purity, EMEA requires in particular sterility of the product. The producer must demonstrate the absence of non-viral and viral adventitious agents as well as Transmissible Spongioform Encephalopathy (TSE). He/she must also ensure the removal of unnecessary source protein and DNA. In the case of continuous production, the homogeneity of the active substance must be ensured to a sufficient degree. EMEA, in principle, requires in this respect batch-to-batch consistency118 which is difficult to achieve in drug manufacture based on animals, especially in the case of antibodies. The producer must also be able to detect quality failures during the whole production process. The head of pharmaceutical production is obliged to comply with the requirements and principles of good manufacturing practice for pharmaceuticals established under Community law, including GMP and GAP guidelines (articles 46 [f], 47 Directive 2001/83). There are various guidelines on GMP and GAP of international origin such as ICH and WHO guidelines which the EC has partly incorporated into its own guidelines (Directive 2005/62).119 The general regulatory problem in this respect, especially in animal pharming, is the assurance of product consistency and protection of the production process and the product against risks originating from the environment, such as infectious and viral agents and TSE that may spread by intrusion of animals into the production premises from the outside. EMEA guidance is urgently needed regarding this matter. The required measures are not necessarily such that options for cost efficient mass production available under GMO regulation, especially keeping the production herd outside with double fencing or cultivation of transgenic plants in open 116 117 118 119

See Schmitt 2004:26–29,46/47. EMEA 1995a (under revision). EMEA 1995a:No. 8. As to the various ICH guidelines see Schmitt 2004:25,34–36.

8.6 Market authorization phase

265

fields, cannot be used in practice. There may be particular cases of highly vulnerable production animals or plants, or perhaps also very risky uses of the end product, where biosecure facilities may be needed already with respect to the keeping of the production herd and the cultivation of the transgenic plants.120 However, it would seem that normally the operator can also put the emphasis on safety and control measures relating to the gaining and processing of the crude bulk material and the manufacturing of the preparation.121 What counts is that the medicinal protein is within the defined range and free of non-viral and viral pathogens and TSE. The environmental assessment required by article 6 Regulation 726/2004 in conjunction with article 8 [3] [ca] and [g] Directive 2001/83 and the relevant EMEA guidelines122 does not give rise to additional problems, since the kind of pharmaceuticals that are presently produced by pharming are not normally associated with new adverse environmental effects. However, this statement cannot be generalized. For example, if hormones were produced by recombinant DNA technology, the potential environmental effects would require careful assessment and consideration. Finally, by reference to articles 40-53 Directive 2001/83, article 6 [1] Regulation 726/2004 requires that the production of medicinal products must be subject to a production (facility and personal) license whose prerequisites are regulated by national law. They include availability of suitable and sufficient premises, technical equipment and control, and existence of qualified and sufficient staff. In the United States, under section 505 [a] Federal Food, Drug and Cosmetic Act the central authorization prerequisite for pharmaceuticals is that they are safe for use and effective in their application. This provision is based on a precautionary approach. The authorization is denied if there is insufficient information to determine safety or there is no substantial evidence on effectiveness (section 505 [d]). The guidelines on recombinant DNA derived pharmaceuticals of 1996123 address a number of typical safety and quality issues in this field, but are outdated. The applicant has to supply detailed information to support the application in a way similar to European law. The application is processed by the Center for Biologics Evaluation and Research (CBER) of the FDA in a tiered system of pre-manufacture review, preliminary procedure and main procedure involving close contacts between the applicant and the authority. Moreover, a manufacturing authorization (license for the establishment) is required although pharmaceuticals based on DNA plasmids, certain peptides and monoclonal antibodies are exempt (21 C.F.R § 314.70 [b], [c], 61 Fed. Reg. 24227, 24231 [1996]). 120 121 122 123

As to xenotransplantation and cell therapy see EMEA 2003:No. 2.2; FDA 2003:point D 1. EMEA 2006b. EMEA 2006a. FDA 1996.

266

8 Legal problems of pharming

8.6.4 Labelling A further authorization requirement is proper labelling of the medicinal product. In this field once again Regulation 726/2004 (Art. 12 [1] subparagraph 2) refers to Directive 2001/83. The authorization must be denied when the labelling proposal does not comply with the provisions about risk communication. Labelling requirements are limited to information about active ingredients, the area of application, possible adverse side effects and the kind and duration of recommended intake. Under existing law, consumer sovereignty is not a primary concern in the labelling of pharmaceuticals. Rather, supplementary to regulation on product safety, labelling serves to provide adequate safety of use, in particular by preventing inappropriate prescriptions and wrong dosage.124 The pharmaceuticals in question do not contain active transgenic ingredients. In contrast to food and feed (Regulation 1829/2003), the production process does not need to be communicated. However, one cannot rule out, at least as a matter of policy, that the information to be provided about active substances should also need to include information about their transgenic origin.125 The case for increasing product safety through extended labelling appears weak. Although pharmaceuticals derived by recombinant DNA technology may be associated with specific quality and safety problems, these problems have already been addressed in the authorization procedure. The possibility of unexpected health effects is not greater than with any other animal-based production method, which under present law does not need to be labelled. On the other hand, arguments of consumer sovereignty and patient autonomy, as well as regulatory consistency, militate in favour of information about the transgenic production method. In this respect, a real difference between food and feed, where labelling is required, on the one hand, and pharmaceuticals, where it is not, on the other, does not exist. The interposition of medical doctors as proxies does not justify a differentiation; for complex issues it is common to rely on information agents such as medical doctors and address information to them although the need for information is determined from the perspective of the ultimate beneficiary. In order to avoid overregulation and gain experience with extended labelling, one should begin with a voluntary system whereby special labelling is permissible. In addition, one should introduce an individual right of access to information about the production method at the patient’s request. 124 125

Lorenz 2006:171. See as to the fundamental issues in the general context of labelling transgenic food Cranor 2007:201–221; McGarity 2007:128–150; Streiffer and Rubel 2007:63–87, who are in favour of labelling, and Markie 2007:88–105; Peters and Lambert 2007:151–177; Wolf 2007:178–200, who are against (all in Weirich 2007).

8.7 Production phase

267

8.6.5 Institutional design Regulation 726/2004 provides that the application is processed by the Committee on Human Medicinal Products (CHMP) established as a part of the European Medicines Agency (EMEA; articles 5 and 7 Regulation 726/2004).126 The formal decision is taken by the Commission in the administrative procedure (article 10 [1] Regulation 726/2004). The Commission is not bound by the expert opinion of CHMP.127 However, it is under an obligation to motivate any deviation from the expert opinion (article 10 [1] subparagraph 3 Regulation 726/2004). Normally it follows the CHMP. This configuration raises questions that have been widely discussed under the perspectives of political accountability, delegation and institutionalized scientific expertise.128 The institutional arrangements in the field of evaluation and authorization of medicinal products manufactured by pharming do not show major particularities. EMEA is exclusively responsible for the evaluation of most medicinal products, including pharmaceuticals derived by recombinant DNA technology. EFSA and the member state authorities only play a role in the special procedure relating to medicinal products containing, or consisting of, GMOs under article 6 [2] and [3] of Regulation 726/2004. As regards developmental medicinal products and more generally the development process for such products, national central authorities competent for medicinal products have major responsibilities. Regulation 726/2004 does not contain rules on public participation relating to applications. The decision-making process is essentially of a technocratic nature. Access to information is ensured, subject to exceptions predicated on confidentiality, by Regulation 1049/2001 as amended by Regulation 1367/2006 on public access to information held by Community institutions and bodies. Moreover, in order to meet growing demands and criticism in this direction, EMEA in 2006 adopted a policy of greater transparency that includes the publication of its meeting agendas and minutes as well as rules for strengthening impartiality and independence of its expert bodies.129

8.7 Production phase After having secured an authorization for the medicinal product, the production can commence. It consists of cultivating transgenic plants or keeping a transgenic production herd, gaining transgenic crude bulk material 126 127

128 129

For details see Lorenz 2006:266 et seq. European Court of First Instance, 2003 ECR II 6053 Nos. 52/53 – Nancy Fern Olivier; see also 2002 ECR II 3305 Nos. 188 et seq. – Pfizer Animal Health; 2002 ECR II 3395 Nos. 195 et seq. – Artegodan. See, e.g., the contributions in Joerges et al. 1997. EMEA Press release of 12 June 2006, EMEA/216787/2006.

268

8 Legal problems of pharming

from the plants or animals, purifying and processing it and manufacturing the medicinal product (“downstream production”). The regulatory problems that arise in this stage are not fundamentally different from those typical for the previous stages. However, additional factors to be considered are the larger scale of the relevant operations and greater access to the environment.

8.7.1 Protection against risks to the environment by use and release of GMOs The production entails an ongoing use or release of GMOs which is not expressly regulated. As a matter of policy, since the authorization of the medicinal product does not cover the manufacturing process except for product quality, it seems clear that GMO-specific regulation must apply to the production process, either Directive 2001/18 or national law on contained use based on Directive 90/219. However, as far as production without containment is concerned, it is unclear which regime of Directive 2001/18 governs the production process. Both the regime concerning releases (part B) and placing on the market (part C) have to be considered. Directive 2001/18 in its recitals seems to reserve the release regime to development releases (recitals 23, 25). Likewise, article 6 [2] Regulation 726/2004 with respect to pharmaceuticals that contain, or consist of, GMOs only refers to such activities.130 One could conclude – and this seems to be the common understanding of the release regime – that releases that engender ongoing production activities are not covered by it. Although the notion of release as defined in Directive 2001/18 does not contain this restriction – release is any introduction into the environment unless it entails making the GMO available to third persons – this interpretation is suggested by the express provision of article 6 [9] of Directive 2001/18.131 According to this provision, material derived from GMOs which are deliberately released in accordance with part B may only be placed on the market if authorized by a part C permit. The directive does not require this material to contain GMOs. As a matter of policy, the regime relating to the placing on the market also appears to be more appropriate because, beyond the marketing in the strict sense, it includes the permanent and wide-spread cultivation of transgenic crops and keeping and propagation of transgenic animals (see articles 19 [3] [c], 20, 23 [1] Directive 2001/18). That the rules relating to coexistence are connected to marketing must also be considered. 130

131

Article 5 [1] of directive 2002/18 which rules out, under certain conditions, the application of part B (articles 6–11) to medicinal products is confined to the release of transgenic medicinal substances and compounds for direct human use and does not cover the production process. Schmieder 2005:49,51; in this sense also EFSA 2006a:8; cf. Ostertag 2006:233; contra Voß 2006:158.

8.7 Production phase

269

There are some counter-arguments. In pharming, the material derived from the release as such is not transgenic and the placing on the market of the pharmaceutical is already covered by the authorization granted under Regulation 726/2004. If one justifies the part C regime attached to the placing on the market by a resulting greater access of releases to the environment, one has also to underline that in pharming the impact of the release remains locally limited. The wording of some national laws relating to marketing is limited to the placing on the market of material that contains, or consists of, GMOs,132 so that these laws would need to be interpreted extensively or amended. Finally, both article 12 [2] Directive 2001/18 and article 6 [2] Regulation 726/2004 declare the entire marketing regime to be inapplicable to the marketing of pharmaceuticals containing, or consisting of, GMOs where an authorization has been secured under Regulation 726/2004. It is doubtful whether it would make sense to apply it to the cultivation of the transgenic plants or the keeping of transgenic animals that, as such, are not made available to third persons.133 Nevertheless, the better arguments militate for the proposition that the whole production process, where it occurs in an open system, needs to be performed under a marketing authorization. The authorization granted under article 6 [2] Regulation 726/2004 must be based on the environmental risk assessment of the release and therefore also covers the GMO-specific aspects of the production process. This is not true of the authorization for pharmaceuticals that are simply derived by recombinant DNA technology without containing, or consisting of, GMOs. There is a need for a GMspecific scrutiny of the whole production process, which is best performed under the part C regime of Directive 2001/18. Instead of an open release, a contained use under national law based on Directive 90/219 has to be considered. In selecting between the contained use and the marketing regimes, arguably the operator has a choice. Article 2 [4] 2nd indent of Directive 2001/18 exempts activities in the framework of a contained use under national law, that is equivalent to Directive 90/219, from the requirements applicable to placing on the market. Production under the contained use regime is more protective of human health and the environment than that in an open environment because a significant exposure does not take place. The finished product does not normally contain any GMOs. Other risks associated with the pharmaceutical, including the adventitious presence of GMOs, have already been fully assessed. Therefore, there are no strong arguments that militate for exclusive reliance on the part C regime under Directive 2001/18.134 132 133 134

Exception: Section 14 [1] No. 4 in conjunction with No. 2 German Act on Biotechnology. Ostertag 2006:217. In favour of the contained use regime also Voß 2006:155; EMEA 1995c.

270

8 Legal problems of pharming

However, it should be underlined that the existing framework conditions for the pharming production process are very complex and unclear. The interplay between the horizontal concept underlying Directive 2001/18 and the sectoral concept of the regulation of pharmaceuticals is not well defined.135 This creates legal uncertainties that may hamper innovation and entail unnecessary transaction costs, and in the worst case may cause regulatory gaps. The operator has a legitimate interest in legal certainty and protection of his/her investment. For the future, a clarification in Regulation 126/2004 appears desirable. In the rare case where the producer intends to market the transgenic plant developed by him/her as seed for (exclusive) use as a pharming platform, an authorization under the seed regime, especially Directives 2002/53 and 2002/55 and national law implementing the directives, is required as well. Presently, there is no Community rule in place that declares environmental risk assessment carried out under seed regulation to be equivalent to that required by Directive 2001/18. The national seed authorization depends on a preceding national authorization for a release of GMOs (article 4 [4] Directive 2002/53, article 4 [2] Directive 2002/55). An alternative is a Community authorization for the marketing of GMOs for cultivation.136 The obligations of operators to provide information and perform a risk assessment, and the prerequisites for granting the marketing authorization under part C of Directive 2001/18 (articles 4, 12-24, annexes II, III and in addition annex IV) and national law implementing the directive137 are by and large identical to those under part B. In view of the experience gained with the preceding releases, taking safety measures is no longer a permit prerequisite but can be imposed as a condition. The fact that the geographic scale might be larger does not normally play a particular role regarding pharming activities. Therefore, the operator who had developed the transgenic pharmaceutical under the release regime can largely rely on the documentation provided in the preceding release authorization procedure, and add information about experience gained so far (see article 13 [2] subparagraph 2, [3] Directive 2001/18). A major substantive difference to the release regime is the requirement of post-marketing monitoring, which is designed to detect unexpected adverse effects presented by the ongoing cultivation of transgenic plants and keeping and breeding of transgenic animals (arti135 136 137

In this sense also European Commission 2004:6,7. As to the question whether Regulation 1829/2003 covers seeds, see articles 5 [5] [1], 17 [5] [2], 18 [3] [c]; Ostertag 2006:76–84 with further references. Germany: Sections 15 [1], 16 [2]-[5] Act on Biotechnology, section 6 Biotechnology Procedure Regulation; United Kingdom: Sections 109 [5], 111 [1] Environmental Protection Act, sections 14, 16 Genetically Modified Organisms (Deliberate Release) Regulations 2002; France: Articles L533-5, R533-26 and R533-27 Code de l’environnement.

8.7 Production phase

271

cle 20 [1] of the directive and implementing national law).138 Given the relatively small geographic scale of pharming production activities, compliance with this requirement is not overly burdensome. As regards waste disposal and occupational safety and health, there are no particularities of the production phase (see sections 8.2.5 and 8.4). What is different, though, are the procedures and institutional arrangements. Under article 15 [2] Directive 2001/18, the competent national authority has exclusive competence for denying an application for placing on the market, although all other member states and the Commission must be given an opportunity to discuss outstanding issues. Where the national authority intends to grant the authorization, it can only decide on it in the absence of a reasoned objection by another member state or the Commission; otherwise the decision on the pending application is elevated to Community level. The Commission or eventually the Council decides on the application in the regulatory procedure (articles 15 [3], 18 of the directive). Under article 28 [1] of the directive EFSA is involved in the preparation of the decision; the obligations of cooperation with expert bodies of member states, established by articles 28-30 Regulation 178/2002, apply. In view of the usual lack of consensus or at least a sufficient majority on GMO issues among the member states, the Commission and EFSA occupy a key position. Specific guidelines for the risk assessment of GMOs used as production platforms for pharmaceuticals do not exist. However, EFSA is preparing guidelines for the risk assessment of genetically modified plants used for non-food or non-feed purposes (PPP guidance).139 Directive 2001/18 provides for some form of public participation in the centralised procedure, more specifically in the preliminary procedure. Under article 24 [1[of the directive, the Commission must make available to the public summaries of the dossier submitted by the applicant, as well as positive national assessment reports. There is an opportunity to make comments to the Commission. Once the central regulatory procedure has commenced, there are no further opportunities to participate in the preparation of the decision. However, the high degree of politicization of GMO policy may result in extended possibilities of the public to indirectly influence the decision through pressure exerted on national governments represented in the regulatory committee. As regards access to information, under Regulation 1049/2001 (as amended by Regulation 1367/2006) all members of the public may request access to the full documentation submitted by applicants, unless it is confidential. 138

139

Germany: Section 16c Act on Biotechnology; United Kingdom: Section 112 Environmental Protection Act 1990, section 16 [2] [g], [5] Genetically Modified Organisms (Deliberate Release) Regulations 2002; France: Article R533-32 [6] Code de l’environnement. EFSA 2008. For a concrete case of EFSA involvement in the marketing authorisation of a non-food/feed transgenic product see supra note 45.

272

8 Legal problems of pharming

There is widespread criticism of the technocratic performance of EFSA’s advisory functions. In particular, it has been claimed that EFSA does not adequately consider inputs from national governments and that expert bodies are generally biased in favour of GMOs. In response to this criticism, EFSA has tried to strengthen relationships with member states. It has referred to its rules on scientific independence which include the requirement for panel members to declare conflicts of interest. Finally, EFSA has developed a policy of openness and transparency, whereby the summaries of the GMO applications and all EFSA opinions are published and made available on the EFSA website.140 In EU practice, due to the high degree of politicization of GMO issues, central decision-making on the marketing of GMOs has become the rule. However, the geographic scale of pharming production operations is normally small, although their environmental effects will extend beyond the area of cultivation, and the resulting products do not contain, or consist of, GMOs. Therefore, it may be expected that objections by other member states or the Commission will be less frequent and limited to cases where the experimental release had already been controversial or production will occur on several sites. This may give more room for decisions being taken by the national authorities. The strategic interests of the Commission in favour of centralized decision-making may be a counterbalancing factor. In the United States, regulation of releases of GMOs under the Plant Protection Act also applies to the cultivation of pharming and other non-food plants for commercial use. Normally, a permit is needed so that the permit procedure under the applicable regulations (7 C.F.R. § 340.2) must be performed. This may include an environmental assessement. However, there is the possibility to secure a determination of non-regulated status based on past experience with a similar release, e.g. a preceding experimental release (7 C.F.R. § 340.6).

8.7.2 Coexistence between pharming and conventional and organic agriculture 8.7.2.1 The problem

Directive 2001/18 does not cover adverse socio-economic effects associated with the acceptance of transgenic agriculture. As already stated, whenever GMOs are introduced into the environment, spread beyond the place or area where they are handled cannot be ruled out. The safety measures that already need to be taken in order to manage potential risks to human health and the environment, including the agricultural environment, also provide a shield of protection against socio-economic harm. Nevertheless, plant 140

See, e.g., EFSA 2006c; see also EFSA News 2006 No. 13:1; 2007 No. 14:1; No. 16:2.

8.7 Production phase

273

pharming may lead to some out-crossing through pollen drift and other spread of GMOs into fields with traditional or organic crops and adversely affect these crops or the seeds derived from them. Moreover, although the transgenic plants used in pharming are not designed as feed, and therefore the risk of commingling appears to be low, it cannot be entirely ruled out. For example, commingling can occur in the harvesting process when the same harvesting machines are used. As a consequence of “genetic contamination” caused on these pathways, conventional and especially organic farmers may suffer economic losses due to applicable mandatory or voluntary labelling requirements (for example the loss of GM-free or organic status). They may incur increased costs when they have to take technical measures to prevent and monitor adventitious admixtures. In the case of pharmaceutical properties, marketability of crops may be totally threatened because consumers and manufacturers alerted about “genetic contamination” may refuse to buy the products. The policy of securing compatibility between GM cultivation and conventional and organic cultivation is denoted as coexistence. 8.7.2.2 Sources of regulation

Directive 2001/18 empowers the member states to regulate in this field (article 26a). The Commission Recommendation of 2003 on coexistence141 discusses the issues raised here and gives some guidance to the member states on how to shape their regulation. As yet, only few member states have coexistence regulation in place.142 One of the reasons for this is that cultivation of transgenic crops is only envisaged for the future. However, in a number of member states new regulation is under preparation. The most important laws regulations that exist are: – Germany: Sections 1, 16a, 16b and 36a Act on Biotechnology of 1990 in the consolidated version of 1993, as amended (relevant amendments in 2004 and 2008); – Denmark: Sections 9–12 Act on the Growing etc. of Genetically Modified Crops (Act No. 436 of 2004), Executive Order on the Growing etc. of Genetically Modified Crops of 2005 and Executive Order on Compensation for Losses due to Certain Occurrences of Genetically Modified Material of 2005; – Netherlands: Regulation of the Comodity Board for Arable Farming on the Coexistence of Crops of 2005; – Portugal: Decree-Law 160/2005 on the Cultivation of Transgenic Varieties Ensuring Coexistence and Order of 2006 on the Establishment of GMO-free Areas; 141 142

Supra note 57. See COM 2006, 104 final; GMO Safety 2007.

274

8 Legal problems of pharming

– Austria: sections 62c and 78k of the Law on Biotechnology of 1994, as amended (relevant amendment in 2005), and provincial laws. 143 As spelt out, for instance, in section 1 No. 2 of the German Act, the purpose of the laws and regulations on coexistence is to ensure that products, in particular food and feed, can be produced conventionally, organically or by use of genetically modified organisms and placed on the market. The structure of the laws is quite different. They may contain duties of care that are directly applicable but may also need to be specified by the government. Often they only have an enabling character, according the central government or, as in Austria, the provincial governments powers to adopt regulation. As regards the regulatory instruments used, one can especially distinguish between administrative regulation (including recommendations issued by the government), labelling and retracing requirements, mandatory information about the site and time of releases (which exists in many member states), and liability. 8.7.2.3 Confinement and protection measures

A common practice to ensure coexistence is the determination of mandatory or recommended crop-specific separation (safety) distances, to be observed in fields where transgenic crops are grown (especially for maize and oilseed rape), and sometimes combined with buffer zones and/or differentiated according to whether conventional or organic farming is to be protected. However, not all member states that possess coexistence regulation have established safety distances. Although the setting of safety distances is normally justified with scientific findings, the figures show a great degree of national variation which cannot be simply explained by differences in climatic conditions and agricultural practices.144 The Commission145 voices some concern about the more stringent segregation distances introduced or proposed by some member states on the grounds of proportionality. However, as reliable empirical evidence is missing, the scientific basis of this criticism remains unclear. A crucial question is whether the precautionary principle can also be applied with respect to the protection 143

144

145

For a detailed analysis of most of these laws see Grossman 2007a:370–388. In France, a new coexistence law (proposal No. 149, 19 December 2007, new articles 663.8 to 663.10 Code rural) which would introduce agency empowerments to require strict protection measures such as separation distances, especially for the protection of controlled cultivation zones (“appellation d’orgine contrôlée”), and establish strict liability of GM farmers is under parliamentary discussion. For instance, with respect to maize, these distances range from 15/50 meters in Sweden to 200 meters in Denmark and to 800 meters in Bulgaria and Luxembourg; regarding oilseed rape the distances are 200/400 meters in Finland, 3000 meters in Luxembourg and 4000/6000 meters in Latvia. Portugal and Germany have introduced safety distances for maize that distinguish between conventional and organic farmers. European Commission 2006:6.

8.7 Production phase

275

of the economic interests of conventional and organic farmers. Section 16b [1] and [2] of the German Act on Biotechnology German law expressly provides for this in formulating duties of care. A radical form of segregation of transgenic agriculture and conventional and organic agriculture is to declare large areas of a country as GMO-free zones. An attempt by the Austrian province Upper-Austria to declare the whole province as GM-free has been held by a recent judgement of the European Court of Justice146 to violate article 95 [5] EC Treaty, that in the case of a directive based on the legislative competence for harmonization permits the taking of more stringent national measures only under very defined circumstances. More limited segregation models are practiced in other Austrian and some Italian provinces and in Portuguese municipalities. Their compatibility with article 95 EC Treaty or, which would seem to be more appropriate, with Directive 2001/18 itself, has yet to be tested. A more general approach (which can be combined with safety distances) is the prescription of general duties of care. For example, section 16b of the German Act on Biotechnology provides that all persons who cultivate and process transgenic crops have to take precautionary measures to avoid a significant impairment of coexistence by a transfer of genetically modified properties, commingling and other dispersal of GMOs. Cultivation and processing is not permissible where, according to the circumstances of the individual case, achievement of coexistence is not ensured. In the cultivation and other handling of plants, these requirements are deemed to be complied with by observing the rules of good professional practice. The Act specifies these requirements with respect to cultivation by naming minimum distances, the selection of plant varieties, combating volunteers or use of natural pollen barriers for preventing gene spread at the time of seed and harvest, as well as out-crossing to other crops and wild relatives. Other rules of good professional practice relate to storage and transportation. Here commingling must be prevented, in particular by segregation of GM and non-GM products and cleaning of the premises, containers and means of transportation. However, as there are no sanctions provided for any violation of this duty of care, one should not overestimate its effectiveness. Its major impact lies in the field of liability. Duties of care also exist in other countries, for example relating to growing intervals, handling of seed for sowing, harvesting crops, transport of harvested plant material and handling of waste plants and plant materials. Often they are not formulated in a general fashion but are crop-specific (for instance in Denmark). These regulatory approaches are in principle also useful with respect to pharming. 146

2007 ECR I 7141 – Land Oberösterreich and Republic of Austria/Commission, confirming European Court of First Instance, 2005 ECR II 4005 – Land Oberösterreich/Commission; for a discussion of the problems associated with GMO-free zones see Grossman 2007a:364–366.

276

8 Legal problems of pharming

8.7.2.4 Labelling requirements

Regulation 1829/2003 sets forth rules for mandatory labelling of food and feed that contains, or consists of, GMOs or is produced from GMOs or contains ingredients that are produced from GMOs.147 The notion of GMO must be interpreted to the extent that it includes natural propagation of a transgenic plant or animal. Regulation 1830/2003 supplements the labelling rules by more technical provisions relating to traceability. The labelling rules are primarily designed to ensure free consumer choice, but also have a particular relevance for coexistence between transgenic cultivation on the one hand and conventional and organic cultivation on the other. The duty of care and the rules of good agricultural practice in this respect aim to avoid a significant impairment of coexistence through gene spread and commingling. The labelling thresholds are an indication of what is tolerable although, as regards organic farming, some states have introduced a more stringent regime. The general labelling threshold is 0.9 percent for the adventitious or technically unavoidable presence of GMOs in food and feed (articles 12 and 24 Regulation 1829/2003). Under article 23 [3] Regulation 324/2007, this threshold will also apply to organic food (as from 2009). National law may set stricter requirements for positive labelling as “GM-free”.148 However, in spite of the traceability rules under Regulation 1830/2003, the regulation still suffers from a lack of common analytical methods for determining non-compliance. In any case, the 0.9 percent labelling threshold is of little legal relevance in the field of pharming. It is confined to GMO traces contained in food or feed that are (directly or indirectly) authorized for use in food or feed (articles 4 [2], [4], 16 [2], [4], 47 Regulation 1829/2003). This can be ruled out with respect to pharming plants because GMOs with pharmaceutical properties are not fit as food or feed. The consequence is that there is a zero tolerance level for pharming material. Any – analytically detectable – presence of such GMOs converts food or feed into an unauthorized transgenic product. As can be derived from article 47 in conjunction with articles 4 [2] and 16 [2] Directive 2001/18, placing on the market or use of such products is illegal.149 Their sale and use can be prohibited by an administrative order, although the competent authorities have a certain margin of discretion in 147 148 149

See Grossman 2007b:32–62; Canfora 2006:170–189 (also regarding national law). For instance in Germany no GM feed may be used. Ostertag 2006:159–168,212–217,230–236; Schmidt-Eriksen 2001:97; Mecklenburg 2006:229; contra Linke 2003:154. See also the definition of GMO under section 3 No. 3 German Act on Biotechnology. Transitional provisions which set a threshold of 0.5 percent for unauthorized GMOs pending authorisation proceedings (article 47 Regulation 1829/2003) are not relevant either. In any case they have expired in the meantime.

8.7 Production phase

277

deciding whether, in case of adventitious presence of such GMOs, they go against the relevant farmers and/or sellers.150 The zero tolerance level does not violate the principle of proportionality.151 As regards seeds, there are at present no specific rules regarding the adventitious presence of GMOs. This means that there is a zero tolerance level, with the result that all detectable traces of authorized transgenic material must be labelled and the sale of seeds with unauthorized material is prohibited. Although the Commission is willing to set special labelling rules and had made various proposals in this respect, wide-spread opposition by member states and environmental and organic farming associations has led to a stalemate in the rule-making process.152 Of course, due to the limits of possible analytical detection the zero tolerance level cannot be complied with completely. For practical reasons, one normally accepts a GMO content up to 0.1 percent.153 It may also be assumed that, because of the fragmentation of the legal rules and the difficulties for the competent authorities to detect and prove violations, the degree of compliance is relatively poor. In the case of highly bioactive GMOs with pharmaceutical properties, one should aim to achieve the highest scientific limits of possible analytical detection. Accordingly, the practical confinement and protection measures may have to be more stringent than those for the cultivation of transgenic food or feed crops. A more radical approach for preventing possible adventitious gene spread would be to limit plant pharming to non-food and nonfeed crops or require total containment. Similarly, in the United States there is a zero tolerance level for adventitious traces of GMOs that are not authorized or notified. Transgenic material from pharming plants may not be introduced into the environment without the permit required under the Plant Protection Act (7 U.S.C. § 7712). Food containing such traces is considered to be adulterated and misbranded in the meaning of sections 342 [a] [1] and 343 [a] Federal Food, Drug and Cosmetic Act because there is no reasonable certainty of no harm. The competent agencies have the discretion to decide whether and under what circumstances they proceed against the relevant plant growers and/or sellers and may decide that the transgenic material does do not pose a significant risk.154 However, in the case of transgenic material with pharmaceutical properties, it would seem normal that, in addition to strengthening 150

151 152 153 154

For Germany see Section 26 Act on Biotechnology; Administrative Court of Appeal Münster, Neue Zeitschrift für Verwaltungsrecht 2001; Ostertag 2006:296–302,306–313; Schmidt-Eriksen, Mecklenburg and Linke, supra note 147 with further references. European Court of Justice 2004 ECR I 3465 Nos. 48–53 – Bellio Fratelli; contra Herdegen & Dederer 2001:11,13/14,20 et seq. See Grossman 2007a:355–360. Federal Ministry of the Environment, Umwelt 2003:479. See USDA/APHIS 2007:14650/51.

278

8 Legal problems of pharming

confinement and protection measures on the fields, action should be taken to prevent further sales of the contaminated food products. 8.7.2.5 Liability

Finally, liability has emerged as an important element in inducing behaviour that is protective of coexistence or to compensate non-GM farmers for financial losses due to gene spread.155 Two models deserve attention because of their implications for plant pharming: the German model of strict liability of transgenic farmers and the Danish model of a compensation fund. Section 36a of the German Act modifies the elements of substantiality and customariness of an interference and economic reasonableness of preventive measures that constitute the core of the strict liability rules of neighbourhood relations contained in sections 1004, 906 Civil Code.156 Under the new law, a transfer or other dispersal of GMOs from neighbouring land is deemed to be a substantial interference where, contrary to the intentions of the owner or holder of the affected land, the agricultural products may no longer be placed on the market because of the transfer and other dispersal. The products need to be labelled as containing GMOs (because the labelling thresholds are exceeded) or the products can no longer be labelled in the way that they could according to the legal provisions applicable to the production method (for example as organic or GMO-free). The wording of section 36a of the German Act militates for the proposition that this does not only cover mandatory labelling under Regulation 834/2007 on ecological/biological products but also voluntary labelling according to the more stringent rules of associations of organic food growers.157 The market-based definition of substantiality, although hardly compatible with the use-related original system of the Civil Code, provides conventional and organic farmers a higher degree of protection because they do not need to tolerate interventions caused by transgenic cultivation per se on the grounds that the mere presence of GMOs is not substantial. However, conventional, organic and transgenic farming are all declared to be customary methods of cultivation under local relevant circumstances. Thereby, a defence against the introduction of transgenic crops predicated on priority is made impossible.158 In essence, the problem is shifted to reasonableness analysis and linked to compliance with the rules of good professional practice. Under section 906 [2] Civil Code, a substantial interference, although it is customary, can be enjoined where it could be prevented or mitigated by protective 155 156 157 158

See Rehbinder and Loperena 2006:266; Grossman 2007a:97–107. See in particular Kohler 2005:566; Wagner 2007:1017. Contra Arnold 2006:16/17; Wagner 2007:1024/1025. By contrast, section 79k of the Austrian Act on Biotechnology seems to accord the neighbour under like circumstances injunctive relief.

8.7 Production phase

279

measures that are economically reasonable. Section 36a Act on Biotechnology declares measures for compliance with good professional practice to always be reasonable. Non-compliance with such rules can therefore lead to an injunction of transgenic agriculture. Nevertheless, compliance does not automatically dispense the transgenic farmer from the obligation to pay compensation. Section 906 [2] Civil Code accords a neighbour, who must tolerate an intervention because protective measures would be unreasonable, financial compensation where the customary use of his/her land or the revenue derived from it is impaired beyond a tolerable degree. Moreover, this liability for compensation also lies in the case of contributory causation. Under the circumstances where several neighbours could have caused the harm and it cannot be determined which neighbour has actually caused it, all relevant neighbours are severally and jointly liable, unless a particular neighbour can prove that he/she only caused part of the damage and estimation of the share is possible. The new law imposes on transgenic farmers a high burden of liability. Apart from criticism based on systematic grounds, some authors159 sustain that it accords organic farmers a preference, instead of ensuring coexistence, and therefore is not compatible with article 26a Directive 2001/18. However, in its report on the national implementation of this provision,160 the Commission has abstained from challenging any liability rules. It is also argued that the new law violates the principle of proportionality, at least insofar as it protects self-chosen quality claims for organic products that are more stringent than Community regulation on organic farming.161 However, in defence of the new liability rules one can state that they are justified in order to prevent a creeping “genetic contamination” and will foster cooperative arrangements between farmers and the establishment of GMO and GMO-free agricultural zones.162 As regards plant pharming, it should be noted that the duty of care and the rules of good professional practice under section 16b of the German Act on Biotechnology also apply where it is not, or not only, coexistence that is concerned but rather human health or the environment, for example in cases of unforeseen effects. Moreover, due to its general preventive effects, liability under section 36a of the Act can at least indirectly contribute to avoiding potential adverse effects on human health and the environment under such circumstances. The Danish Act on the Growing etc. of Genetically Modified Crops of 2004 has introduced a collective compensation scheme for losses suffered 159 160 161 162

Dolde 2004:219; Wolfers and Kaufmann 2004:421; Schmieder 2005:49. Supra note 141:7. See authors cited supra note 159. Palme et al. 2004:176/77; Palme 2006:76. Cooperative agreements are now expressly admitted in Germany; § 16b Act on Biotechnology (as amended in 2008).

280

8 Legal problems of pharming

by conventional or organic farmers due to the occurrence of genetically modified material in their crop originating from transgenic agriculture in the neighbourhood (sections 9–12). The scheme is funded by a levy charged on all GM farmers (section 12)163 and administered by the Plant Directorate. Insofar as compensation is paid, claims for damages that the victim may have against the person responsible for his/her loss under private law are subrogated to the state. Compensation requires that, in the same growing season and within a specified area, a genetically modified crop of the same or a related variety has been grown and has crossbred with the crop of the victim. Administrative rules have fixed the relevant distances between the fields in a very restrictive manner, namely for maize at 300 meters, for beet at 75 meters and for potatoes at 30 meters. Moreover the GM crop must be identified in the crop of the farmer suffering the loss (section 9 [1] [i] and [i]). This latter prerequisite does not mean that, in case of several potential sources, the victim has to identify the precise field from which gene spread has occurred. Rather, it suffices that the presence of transgenic material from a GM crop, grown in the perimeter fixed by the administrative rules, can be determined by analytical methods. Otherwise, securing compensation would be very difficult once transgenic agriculture becomes more common. Under a collective system of compensation funded by all transgenic farmers, the identification of the precise source of gene spread does not make sense. It would introduce into the collective compensation system an element that is geared to traditional individual liability. In the case of organic farming, compensation may be paid at the discretion of the authority independent of these limitations (the distance requirement and the obligation to identify the transgenic crop of origin) where adventitious admixtures of genetically modified seed occur in the farmer’s seed for sowing (section 9 [4]). In all other respects, conventional and organic farmers are treated equally. Compensation comprises losses caused by lower sale prices for the products on the market, the costs of sampling and analysis and losses due to requisite conversion from organic to conventional farming resulting from the genetic contamination. However, compensation will not be granted for losses caused by the presence of GMs in the crop that do not exceed a threshold to be set by the competent Minister. This threshold has been set at 0.9 percent, and in the case of seed at the level that Community legislation will provide for special GMO labelling of seed (section 2 Executive Order).164 This means that in contrast to Germany, self-chosen quality standards of organic farmers do not confer a right to compensation. In the case of pharming it is hardly satisfactory that liability should only apply 163 164

This financing method has been approved by the Commission under article 87 EC Treaty; see State aid case N 568/2004. Executive Order on Compensation of Losses due to Certain Occurrences of Genetically Modified Material 2005.

8.7 Production phase

281

when the threshold of 0.9 percent is exceeded, as this is not applicable to GMOs not authorized as food or feed.165 8.7.2.6 Special issues in animal pharming

The rules on coexistence are, in principle, also applicable to transgenic animals. States that have introduced general duties of care include the keeping of transgenic animals. Section 16b of the German Act on Biotechnology expressly provides that the obligation to ensure coexistence also applies to the keeping of transgenic animals. As an expression of good professional practice in keeping transgenic animals, their escape from, as well as the intrusion of other animals into, their premises must be prevented (section 16b [3] No. 2). In contrast, special liability for genetic contamination does not normally extend to transgenic animals. There is no urgent need to extend strict liability to transgenic animals beyond the existing law.

8.7.3 Animal protection In the production phase, there may also be animal protection requirements that have to be observed. However, there are obvious gaps of protection at European and national levels because the more specific and more protective relevant texts only apply to normal farming activities. Pharming animals held for production are normally only subject to general requirements of animal protection. Thus, the scope of application of the European Convention for the protection of animals kept for farming purposes is confined to “animals bred or kept for the production of food, wool, skin or fur or for other farming purposes”. EC Directive 98/58, which has been adopted to give effect to the principles laid down in the European Convention, uses the same language for describing its scope of application (articles 1 [2] and 2 No. 1). It would arguably go beyond the perception and intention of the contracting parties to the convention and the EC institutions, who drafted the texts, to advocate a dynamic interpretation of the notion of “farming purposes” that includes the manufacture of pharmaceuticals only because the animals used are the same as those normally used in agriculture. Certainly, the notion of farming does not have a meaning that is fixed for all times. For example, we would perhaps not hesitate to conceive the production of agricultural raw materials for the generation of renewable energy, such as biogas or biodiesel, as agriculture. However, the historic perspective is highly relevant when interpreting international agreements. This prohibits an interpretation to the extent that an activity which is carried out at the borderline between agriculture and manufacture should be deemed to be for “other agricultural purposes”. 165

A more limited compensation fund that only covers accidental gene spread and commingling exists in Portugal under the Decree-Law 160/2005; see Grossman 2007a.

282

8 Legal problems of pharming

It should be noted that this gap of protection may be carried over to national law. While there is general regulation on animal welfare which also encompasses pharming animals, the more specific and more protective special provisions on the keeping of useful animals are not normally applicable to the keeping of pharming animals, because they do not serve agricultural purposes.166 This is a regulatory anomaly that should be remedied. However, in Germany, section 10a Animal Protection Act provides that interventions in, and treatment of, vertebrate animals for the purpose of production that may be associated with pain, suffering or harm, are only permissible when certain prerequisites applicable to the authorization of animal trials are fulfilled. These prerequisites are indispensability of the intervention, its ethical justification and the 3-R concept. The operator must notify the programme of work to the competent authority, which can prohibit it if the prerequisites of permissibility are not fulfilled. However, it would seem that normal manufacturing activities in the course of pharming are not affected by this provision, since the kinds of interventions covered by section 10a of the act are not performed at this stage. Besides, the general duty of care for the keeping of all kinds of animals under section 2 Animal Protection Act must be complied with. This duty can be summarized as the requirement of compatibility with the nature of the animal.167 Similar requirements of a general nature apply in the United Kingdom and France, who do not possess specific rules relating to interventions into animals in the production phase (section 9 Animal Welfare Act 2006, article R214-17 Code rural).

8.7.4 Production-related requirements under pharmaceuticals regulation In the production phase, the producer must comply with the management and control program provided in the marketing permit and documented in the application for the permit, with respect to the keeping of the production herd or cultivation of the transgenic plants, the gaining and processing of the crude bulk material and manufacturing the end product. In particular, the quality and safety requirements with respect to consistency and the absence of nonviral and viral agents and TSE, must be observed. In the light of experience with the ongoing production, adjustments may be necessary. New information about quality problems must be communicated to EMEA and the Commission (article 16 [2] subparagraph 1 Regulation 726/2004). Where the production herd is kept in the open environment with double fencing, protection against virus epidemics is practically impossible. In order to maintain production during possible epidemics, having several production sites that are separated by significant distances may be advisable in order to avoid market disruptions. 166

167

For Germany see Federal Administrative Court, Buchholz, Entscheidungssammlung des Bundesverwaltungsgerichts 418.9 Tierschutzgesetz No. 13 (2004); for Britain: AEBC 2002:34. Lorz and Metzger 1999: § 2 No. 16.

8.8 References

283

8.8 References AEBC (Animal and Environment Biotechnology Commission) (2002) Animals and Biotechnology. AEBC, London Anderson WL, Bryson NS, Cohen WM, Davis Jr RC, Katz P, Mannix RJ, Quarles SP, Schwartz RE (2001) Biotechnology Deskbook. Environmental Law Institute, Washington Animal Procedures Committee (1997) Report of the Animal Procedures Committee for 1997. The Stationary Office, London www.apc.gov.uk (July 2008) Arnold S (2006) Die Haftung von Landwirten bei Auskreuzungen von gentechnisch veränderten Organismen im Nachbarrecht. Natur und Recht 28:5–22 Blattner O (2003) Europäisches Produktzulassungsverfahren. Das europäische Verwaltungsverfahrensrecht dargestellt an der Zulassung gentechnisch veränderter Arznei- und Lebensmittel. Nomos, Baden-Baden Blattner O (2002) Verwaltungsverfahrensfehler und deren Folgen im europäischen Arzneimittelzulassungsverfahren. Pharma-Recht 24:277–287 Boy L (2002) La place du principe de précaution dans la directive UE du 12 Mars 2001 relative à la dissémination volontaire d’organismes génétiquement modifiés dans l’environnement. Revue juridique de l’environnement 27:5–23 Boyd Group (1999) Genetic engineering: Animal welfare and ethics, Discussion Paper, September 1999 www.boyd-group-demon.co.uk (September 1999) Board on Agriculture and Natural Resources (2002) Animal Biotechnology: Science-Based Concerns. National Academy of Science, National Research Council, Washington D.C. Brand V (2004) Rechtliche Maßstäbe der Risikobewertung und des Risikomanagement in Frankreich. In: Breckling et al. (2004) pp 131–153 Brand V, Winter G (2004) Rechtliche Maßstäbe der Risikobewertung und des Risikomanagement in der Bundesrepublik Deutschland. In: Breckling et al. (2004) pp 212–279 Breckling B, Brand V, Winter G (eds) (2004) Fortschreibung des Konzeptes zur Bewertung von Risiken bei Freisetzungen und dem In-Verkehr-Bringen von gentechnisch veränderten Organismen. Berichte 3/04 des Umweltbundesamts. Erich Schmidt Verlag, Berlin Breckling B (2004) Naturwissenschaftliche Grundlagen der Gentechnik als Ausgangspunkt zur Risikoabschätzung gentechnisch veränderter Organismen. In: Breckling B et al. (2004) pp 41–83 Breyer S, Heyvaert V (2000) Institutions for Regulating Risk. In: Revesz R, Sands P, Stewart R (2000) Environmental Law, Sustainable Development, and the Economy. Cambridge University Press, Cambridge, pp 283–352 Bundesministerium für Landwirtschaft, Ernährung und Forsten (2001) Tierschutzbericht 2001. Bundestags-Drucksache 14/5712, Bonn Bundesministerium für Landwirtschaft, Ernährung und Forsten (1997) Tierschutzbericht 1997. Bundestags-Drucksache 13/7016, Bonn Bundesamt für Veterinärwesen (1994) Bericht des Bundesamtes für Veterinärwesen für das Jahr 1994. Bern Calliess C, Korte S (2006) Das neue Recht der grünen Gentechnik im europäischen Verwaltungsverbund. Die öffentliche Verwaltung 59:10–22 Canfora I (2006) Development of Organic Food Labelling Rules in the EU and in National Legislation. European Food and Feed Law Review 1:170–189 Caspar J (1999) Tierschutz im Recht der modernen Industriegesellschaft. Nomos, Baden-Baden

284

8 Legal problems of pharming

Caspar J (1997) Der vernünftige Grund im Tierschutzgesetz. Natur und Recht 19:577–583 Christoforou T (2004) The Regulation of Genetically Modified Organisms in the European Union: The Interplay of Science, Law and Politics. Common Market Law Review 41:637–709 Cranor C (2007) Different Conceptions of Food Labels and Acceptable Risks: Some Contingent/Institutional Considerations in Favor of Labeling. In: Weirich P (ed) Labeling Genetically Modified Food. Oxford University Press, Oxford, pp 201–221 Dederer H-G (1998) Gentechnikrecht im Wettbewerb der Systeme: Freisetzungen im deutschen und US-amerikanischen Recht. Springer, Berlin, Heidelberg, New York De Cock Buning T, Theune E (1994) A comparision of three models for ethical evaluation of proposed animal experiments. Animal Welfare 3:107–128 Department of the Environment/ACRE (1993) Guidance Note 1, The Regulation and Control of the Deliberate Release of GMOs, amended in 1995. London Dolde T (2004) Haftungsrisiken für Landwirte aufgrund der Koexistenz konventioneller Landwirtschaft mit benachbarten transgenen Kulturen? Natur und Recht 26:219–222 Ebersbach W, Lange P, Ronellenfitsch M (eds) (2007) Recht der Gentechnik und Biomedizin, Kommentar. Looseleaf ed. CF Müller, Heidelberg ECVAM (European Centre for Verification of Alternative Methods) (1998) The use of transgenic animals in the European Union, Report and Recommendations. ATLA 26:21–43 EFSA (2006) Fact Sheet “Independence at EFSA” www.efsa.eu.int/science/gmo/catindex_en.html (March 2008) EFSA (2006a) Guidance document of the GMO Panel for the risk assessment of genetically modified plants and derived food and feed of 28 April 2004, final edited version of 28 April 2006. EFSA Journal 99:1 http://www.efsa.europa.eu/en/science/gmo/gmo_guidance/660.html (June 2008) EFSA (2006b) Opinion of the Scientific Panel on Genetically Modified Organisms on a request from the Commission related to the notification (Reference C/SE/96/3501) for the placing on the market of genetically modified potato EH92-527-1 with altered starch composition, for cultivation and production of starch, under Part C of Directive 2001/18/EC from BASF Plant Science, The EFSA Journal 323:1–20 EFSA (2006c) Fact Sheet “Openness and Transparency” www.efsa.eu.int/science/gmo/catindex_en.html, (March 2008) EFSA (2008) Draft Opinion of the Sciientific Panel on Genetically Modified Organisms on the risk assessment of genetically modified plants used for non-food or non-feed purposes, adopted on 16 April 2008 www.efsa.europa.eu/EFSA/efsa_locale-1178620753812_11787166069288.htm (June 2008) EMEA (2007) Guidelines on Environmental Risk Assessment for Medicinal Products containing, or consisting of, Genetically Modified Organisms, EMEA/ CHMP/473191/06 December 2006 (in force as from 1 July 2007) EMEA (2006a) Guidelines on the Environmental Risk Assessment of Medicinal Products for Human Use, EMEA/CPMP/SWP/4447/00 EMEA (2006b) Quality of Biological Active Substances produced by Stable Transgene Expression in Higher Plants, CPMP/BWP/48316/06 (consultation finished), No. 4.4.2. EMEA (2005) Guidelines on Similar Biological Medicinal Products containing Biotechnology-Derived Proteins as Active Substances: Quality Issues, EMEA/ CHMP/BWP/49348/2005

8.8 References

285

EMEA (2003) Points to Consider on Xenogeneic Cell Therapy Medicinal Products, EMEA/CPMP/1199/02, April 2003 EMEA (2000) Development Pharmaceutics for Biological and Biotechnological Products, Guidance document CPMP/BWP/398/99 EMEA (1998) Preclinical Safety Evaluation of Biotechnology-Developed Pharmaceuticals, Note for Guidance, CPMP/ICH/302/95 EMEA (1995a) Production and Quality Control of Medicinal Products Derived by Recombinant DNA Techniques, Guidance document 3AB1a EMEA (1995b) Use of Transgenic Animals in the Manufacture of Biological Medicinal Products for Human Use, Guidance document 3AB7a EMEA (1995c) Guideline on Production and Quality Control of Medicinal Products Derived by Recombinant DNA Technology, 3AB1a, No. 2 European Commission (2007) Commission recommendation of 18 June 2007 on guidance for the accommodation and care of animals used for experiments and other scientific purposes, Recommendation 2007/526/EC European Commission (2006) Report on the implementation of national measures on the coexistence of genetically modified crops with conventional and organic farming, COM (2006) 104 final European Commission (2004) Report from the Commission to the Council and the European Parliamentt on the experience of member states with GMOs placed on thre market under Directive 2001/18/EC and incorporating a specific report on the operation of parts B and C of the Directive, COM (2004) 575 final European Commission (2003) Recommendation 2003/56 of 23 July 2003 on guidelines for the development of national strategies and best practices to ensure the coexistence of genetically modified crops with conventional and organic farming, O.J. 2003 No. L 189:36 European Commission (2001) Applicability of the precautionary principle, COM (2001), 1 final FDA (2003) Guidance for Industry: Source Animal, Product, Preclinical, and Clinical Issues Concerning the Use of Xenotransplantation Products in Humans, Food and Drug Administration, Center for Biologics Evaluation and Research, April 2003 FDA (2002) Draft Guidance for Industry: Drugs, Biologics, and Medical Devices Derived from Bioengineered Plants for Use in Humans and Animals, Food and Drug Administration, Center for Biologics Evaluation and Research, June 2002 FDA (1996) Guidance for Industry for the Submission of Chemistry, Manufacture and Control Information for A Therapeutic Recombinant DNA-Derived Product or A Monoclonal Antibody Product for in Vivo Use. Food and Drug Administration, Center for Biologics Evaluation and Research, August 1996 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 FDA (1986) Final policy statement. In: US Office of Science Technology Policy, Coordinated Framework, 51 Fed. Reg. 23302 Fisahn A (2004a) Das Recht der Gentechnik in Großbritannien. In: Breckling et al. (2004) pp 154–177 Fisahn A (2004b) Beschleunigung und der Schadensbegriff im Gentechnikrecht. Natur und Recht 26:145–150 Fisahn A (1998) Die Genehmigung der Freisetzung gentechnisch veränderter Organismen eine Fallstudie. In: Winter G (ed) (1998) Die Prüfung der Freisetzung von gentechnisch veränderten Organismen – Recht und Genehmigungspraxis. Berichte 4/98 des Umweltbundesamtes. Erich Schmidt Verlag, Berlin, pp 25–49 GMO Safety, Coexistence in the countries of the EU – A European patchwork http://www.gmo-safety.eu (June 2007)

286

8 Legal problems of pharming

Grossman MR (2003) Genetically Modified Crops in the United States: Federal Regulation and State Tort Liability. Environmental Law Review 5:86–108 Grossman MR (2007a) The Coexistence of GM and Other Crops in the European Union. Kansas Journal of Law and Public Policy 16:324–392 Grossman MR (2007b) Traceability and Labeling of GM Food and Feed in the European Union. In: Weirich P (ed) Labeling Genetically Modified Food. Oxford University Press, Oxford, pp 32–62 Gruber FP, Kolar R (1997) Animal test advisory commissions: ethics committees in Germany. In: van Zutphen LFM, Balls M (eds) Animal Alternatives, Welfare and Ethics. Elsevier, Amsterdam, Oxford, pp 373–376 Gruber FP (1995) Die Tierversuchskommission nach § 15 Tierschutzgesetz in dem Recht der Bundesrepublik Deutschland. In: Schöffl H, Spielmann H, Tritthorst HA (eds) Forschung ohne Tierversuche. Springer, Wien, New York, pp 233–239 Health and Safety Executive (2000) Guidance on the Genetically Modified Organisms (Contained Use) Regulations 2000. HSE, London Herdegen M (2004) Koexistenz und Haftung. Springer, Heidelberg Herdegen M, Dederer H-G (2001) Adventitious presence of GMOs in seeds – A legal analysis of legislative projects in the European Union. Springer, Heidelberg Hermitte M-A, Noiville C (1993) La dissémination volontaire d’organismes génétiquement modifiés dans l’environnement – une première application du principe de prudence. Revue juridique de l’environnement 18:391–423 Hermon C (1999) Commentaire sur la décision du Conseil d’Etat du 25 septembre 1998 et 11 décembre 1998 Association Greenpeace France. Revue juridique de l’environnement 24:564–566 Hill J (1994) The Precautionary Principle and Release of GMOs to the Environment. In: O’Riordan T, Cameron J (eds) Interpreting the Precautionary Principle. Cameron May, London Hirsch G, Schmidt-Didczuhn A (1991) Gentechnikgesetz, Kommentar. C H Beck, München Home Office (2000) Guidance on the Operation of the Animals (Scientific Procedures) Act, 23 March 2000. Home Office, London House of Lords Select Committee on Animals in Scientific Proceedings (2002) Report, volume I, London Hughes D, Jewell T, Lowther J, Parpworth N, de Prez P (2002) Environmental Law, 4th ed. Butterworth, London Hughes E (2000) Animal Welfare Law in Canada and Europe. Animal Law 6:23–76 International Conference on Harmonization (ICH) (1998a) Preclinical Safety Evaluation of Biotechnology-Developed Pharmaceuticals, Note for Guidance, CPMP/ICH/302/95 International Conference on Harmonization (ICH) (1998b) ICH 5B, Quality of Biotechnological Products: Analysis of the Expression Construct in Cells Used for Production of r-DNA Derived Protein Products, CPMP/ICH/139/95 International Conference on Harmonization (ICH) (1997) ICH Q 5A, Viral Safety Evaluation of Biotechnological Products Derived from Cell Lines of Human or Animal Origin, CPMP/ICH/295/95 International Conference on Harmonization (ICH) (1995) ICH 5C, Quality of Biotechnological Products: Stability Testing of Biotechnological/ Biological Products, CPMP/ICH/139/95 Jörgensen M, Winter G (1996) Rechtliche Probleme der Freisetzung von gentechnisch veränderten Organismen. Zeitschrift für Umweltrecht 7:293–299 Joerges C, Ladeur K-H, Vos E (eds) (1997) Integrating Scientific Expertise into Regulatory Decision-Making. Nomos, Baden-Baden Kloepfer M (2004) Umweltrecht, 3rd edition. CH Beck, München Kluge H-G (2002) Tierschutzgesetz. Kommentar, 2nd ed. Stuttgart, Kohlhammer

8.8 References

287

Kluge H-G (1994) Grundrechtlicher Freiraum des Forschers und ethischer Tierschutz. Neue Zeitschrift für Verwaltungsrecht 13:869–872 Kohler J (2005) Schadensausgleich in Fällen des § 36a Gentechnikgesetz. Natur und Recht 27:566–575 Kwizda R (1998) Zulassungsverfahren für Humanarzneimittel in der Europäischen Union. Manz, Wien Lewidow L, Carr S, Wield D (1996) Regulating agricultural biotechnology in Europe: Harmonisation difficulties, prospects, dilemmas. Science and Public Policy 23:135–157 Linke T (2003) Nochmals: Zufallsauskreuzungen und Gentechnikgesetz (GenTG). Natur und Recht 25:154–160 Lorenz M (2006) Das gemeinschaftliche Arzneimittelzulassungsrecht. Nomos, Baden-Baden Lorz A, Metzger E (1999) Tierschutzgesetz. Kommentar. CH Beck, Munich Lübbe A (1994) Hat der Tierschutz Verfassungsrang? Natur und Recht 16:469–472 Macrory R, Purdy R (1998) The Regulation for the Release of Genetically Modified Organisms in the United Kingdom. In: Winter G (ed) Die Prüfung der Freisetzung von gentechnisch veränderten Organismen – Recht und Genehmigungspraxis. Berichte 4/98 des Umweltbundesamtes. Erich Schmidt Verlag, Berlin, pp 50–75 Maisack C (2007) Zum Begriff des vernünftigen Grundes im Tierschutzrecht. Nomos, Baden-Baden Markie P (2007) Mandatory GE Labels and Consumer Autonomy. In: Weirich P (ed) Labeling Genetically Modified Food. Oxford University Press, Oxford, pp 88–105 Marx M (1997) Der Sicherheitsstandard der Betreiberpflichten im Gentechnikrecht. Peter Lang Verlag, Frankfurt McGarity TO (2007) Frankenfood Free: Consumer Sovereignty, Federal Regulation and Industry Control of Marketing and Choosing Food in the U.S. In: Weirich P (ed) Labeling Genetically Modified Food. Oxford University Press, Oxford, pp 128–150 Mecklenburg W (2006) Zur Störfallbewältigung im Gentechnikrecht – Aussaat von mit gentechnisch veränderten Organismen verunreinigtem Saatgut. Natur und Recht 28:229–233 Müller-Terpitz R (2007) The Use of Genetically Modified Animals for the Production of Pharmaceuticals – Legal Considerations from an Animal Welfare Perspective. In: Engelhard M, Hagen K, Thiele F (eds) Pharming. A New Branch of Biotechnology. Europäische Akademie, Bad Neuenahr-Ahrweiler, pp 73–86 Office of Science and Technology Policy (1986) Announcement of Policy – Coordinated Framework for Regulating Biotechnolgy. 51 Federal Register 23302. Washington, DC OECD (ed) (1986) Recombinant DNA Safety Considerations. Safety considerations for industrial, agricultural and environmental applications of organisms derived by recombinant DNA techniques. OECD, Paris OECD (ed) (2004) Challenges and Risks of Genetically Engineered Organisms. OECD, Paris Ostertag A (2006) GVO-Spuren und Gentechnikrecht. Nomos, Baden-Baden Palme C (2006) Nationaler Naturschutz und Europäisches Gentechnikrecht. Natur und Recht 28:76–79 Palme C, Schlee M, Schumacher J (2004) Das neue Recht der Grünen Gentechnik: Europarechtliche Vorgaben und fachliche Praxis. Europäisches Umwelt- und Planungsrecht 2:170–183 Palme C, Schumacher J (2007) Die Regelungen zur FFH-Verträglichkeitsprüfung bei Freisetzung oder Inverkehrbringen von gentechnisch veränderten Organismen in § 34a BNatSchG. Natur und Recht 29:16–22

288

8 Legal problems of pharming

Parliamentary Office of Science and Technology (2001) GM Animals. Postnote Number 157. London Peters P, Lambert T (2007) Regulatory Barriers to Consumer Information about Genetically Modified Food. In: Weirich P (ed) (2007) Labeling Genetically Modified Food. Oxford University Press, Oxford, pp 151–177 Pohl S (2004) Risiko und Vorsorge nach deutschem und europäischem Gentechnikrecht. In: Breckling et al. (2004) pp 280–300 Prieur M (2004) Droit de l’environnement, 5th ed. Dalloz, Paris Radford M (2001) Animal Welfare Law in Britain. University Press, Oxford Rehbinder E, Loperena D (2006) Liability for Genetic Contamination of Non-GM Crops. Environmental Policy and Law 36:265–274 Reinhardt M (2003) Materielle Entscheidungsbefugnisse im Gentechnikrecht. Neue Zeitschrift für Verwaltungsrecht 22:1446–1452 Richter D (2007) Die Würde der Kreatur. Rechtsvergleichende Bemerkungen. Zeitschrift für ausländisches öffentliches und Völkerrecht 67:321–349 Royal Society (2001) The Use of Genetically Modified Animals. Royal Society for the Prevention of Cruelty to Animals, London Rusche (2003) The 3 Rs and Animal Welfare – Conflict or Way Forward? ALTEX 20 (Supplement 1):63–76 Russell WMS, Burch RL (1959) The Principles of Humane Experimental Technique. Methuen, London Sachverständigenrat für Umweltfragen (1998) Umweltgutachten 1998. MetzlerPoeschel, Stuttgart Sachverständigenrat für Umweltfragen (2004) Umweltgutachten 2004. Nomos, Baden-Baden Scharmann W, Teutsch G (1994) Zur ethischen Abwägung von Tierversuchen. ALTEX 11:191–198 Schenek M (1995) Das Gentechnikrecht der Europäischen Gemeinschaft – Gemeinschaftliche Biotechnologiepolitk und Gentechnikregulierung. Duncker & Humblot, Berlin Schmieder S (2005) Die Neuregelung der Folgen von Auskreuzungen im Gentechnikrecht. Umwelt- und Planungsrecht 25:49–55 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. Master thesis. University of Bonn Schmidt-Eriksen C (2001) Von Irrungen und Wirrungen im Gentechnikrecht. Natur und Recht 23:492–498 Schneider I (2003) Das Kooperationsprinzip im Vorfeld der Arzneimittelzulassung. Peter Lang Verlag, Frankfurt am Main Steines JC (2002) Widerstreitende Verfahrensansätze für Freisetzung und Inverkehrbringen gentechnisch veränderter Organismen im deutschen und US-amerikanischen Gentechnikrecht. Berlin-Verlag Spitz, Berlin Streiffer R, Rubel A (2007) Genetically Engineered Animals and the Ethics of Food Labeling. In: Weirich P (ed) Labeling Genetically Modified Food. Oxford University Press, Oxford, pp 63–87 Sukopp U (2004) Der naturwissenschaftliche Umgang mit Wissenslücken bei der Risikoanalyse ökologischer Folgen der Freisetzung und des Inverkehrbringens von GVO. In: Breckling B et al. (2004) pp 84–114 Teutsch G (1995) Die Würde der Kreatur – Erläuterungen zu einem neuen Verfassungsbegriff im Bereich des Tierschutzes. Paul Haupt Verlag, Bern, Stuttgart, Wien US Department of Agriculture (2007) APHIS Policy on Responding to the LowLevel Presence of Regulated Genetically Engineered Plant Materials, Animal and Health Inspection Service, 72 Federal Register 14649–14651

8.8 References

289

US Department of Agriculture (2008) Guidance for APHIS Permits for Field Testing or Movement of Organisms Intended for Pharmaceutical or Industrial Use, Animal and Plant Health Inspection Service, February 5, 2008 US Department of Agriculture (1986) Final Policy Statement on GMOs. In: Office of Science and Technology Policy, Coordinated Framework, 51 Fed. Reg. 23302, 23313-19 Voß U (2006) Die Novelle der Freisetzungsrichtlinie. Richtlinie 2001/18/EG. Nomos, Baden-Baden Wagner G (2007) Nachbarhaftung für gentechnische Immissionen. Versicherungsrecht 58:117–134 Weirich P (ed) (2007) Labeling Genetically Modified Food. Oxford University Press, Oxford Winter G (2006) Naturschutz bei der Freisetzung von gentechnisch verändertem Saatgut. Zeitschrift für Umweltrecht 17:456–464 Winter G (ed) (1998) Die Prüfung der Freisetzung von gentechnisch veränderten Organismen – Recht und Genehmigungspraxis. Berichte 4/98 des Umweltbundesamtes. Erich Schmidt Verlag, Berlin Winter G, Mahro G, Ginzky H (1993) Grundprobleme des Gentechnikrechts. Werner Verlag, Düsseldorf Wolf C (2007) Labeling GM Foods: Rights, Interests, Enforcement and Institutional Options. In: Weirich P (ed) Labeling Genetically Modified Food. Oxford University Press, Oxford, pp 178–200 Wolfers B, Kaufmann M (2004) Grüne Gentechnik: Koexistenz und Haftung. Zeitschrift für Umweltrecht 15:321–329 Ziani A (2006) Droit et expériences animales en France. Revue juridique de l’environnement 31:425–441

9 Conclusions and recommendations

9.1 Pharming technology and its market The manufacture of biopharmaceuticals using genetically modified cultured cells and microorganisms is an established and successful industry. Using genetically modified whole animals and plants as a production platform is a more recent development that has sprung from radical innovations in the genetic manipulation of plants and animals, and reproductive technology in animals, over the past 10 to 20 years. The first two pharming products have gained market authorization. The world market for biopharmaceuticals is large, as is an important subset of biopharmaceutical products, including antibodies that could be produced by pharming. For certain proteins, pharming may be the most cost-effective means of production and could increase the availability of valuable medicines. It may also be the only way of producing useful quantities of a particular protein, thereby allowing the development of entirely new medicines. Experience will continue to be gained regarding the suitability of the different production platforms for producing particular proteins. Technical advances in pharming continue to be made across a broad front, both in the industry and in academia. Currently available evidence indicates that pharming is capable of competing on equal terms with other methods of manufacturing pharmaceutical proteins. It has advantages due to the possibility of producing large, complex proteins with appropriate patterns of post-translational protein modification, and the comparatively low scale-up costs. However, these advantages will be relative to the development of the competing technologies, for example the increasing availability of mammalian cell cultures and the possibility of adding post-translational modifications to yeast-produced proteins. Pharming is a viable and potentially competitive means of producing important biopharmaceuticals. In some cases, pharming may be the most efficient, or even the only, way of producing a particular protein. Intellectual property rights have had a major impact on access to key technologies and potential products. While some patents covering valuable resources are due to expire, intellectual property restrictions are likely to

292

9 Conclusions and recommendations

be very important in the short to mid term. Practices such as the granting of excessive and protectionist claims by national governments are a significant negative influence that stifles innovation. The pharming industry in Europe requires a transparent, rational and stable regulatory environment in which to operate if it is to attract the longterm investment necessary for economic viability. Until now legislation has been made in reaction to rapid scientific advance and has been motivated mainly by a need to manage potential risk. The regulatory framework is fragmented and in important respects uncertain. Uncertainties regarding the application of the existing regulation, as well as their future development, are a significant disincentive to the development of new products. It is now timely to consolidate the regulatory framework and decrease the level of uncertainty. The outcome would also lead to equality of competition for pharming with other forms of manufacturing biopharmaceuticals. A clearer regulatory framework for pharming is needed in order to decrease economic risks and secure equality of competitive conditions.

9.2 Public attitudes and moral evaluation 9.2.1 Attitudes Potential controversy on pharming revolves mainly around the means, not the goals. The perception of the means -genetic modification of plants and animals- has two main components, one of a cognitive nature (basic knowledge of genetics) and the other of an evaluative character (views and feelings about purposely changing the blueprint of plants and/or animals). At present, the dominant profile of perceptions of biotechnology in general and pharming in particular is heavily dependent on the second component. Lack of knowledge, misunderstandings and misrepresentations of basic genetic concepts are observed in most advanced societies and particularly in Europe. A higher level of knowledge does not automatically translate into more positive attitudes, but at a minimum will certainly help to remove unfounded fears and empower individuals to make informed choices about accepting or rejecting pharming. The acceptance or rejection could also become less holistic, giving way to a more fine-grained evaluation, depending on the specifics of the goals and the means, such as the type of plant or animal to be used and the purpose of the biomedical application. Special attention should be paid to offering unbiased basic information about genetics and about pharming procedures to the public. The scientific community and the policy makers should foster a climate of open dialogue with society at large, without taking for granted that any intervention at the genetic level of life has to be perceived by the public as a “good

9.2 Public attitudes and moral evaluation

293

thing” and, even less, that all forms of opposition or cautious stand on the part of the public are a mere function of ignorance about the means and the risks. Public‘s views, values and preferences about nature and its genetic modification for different goals should be taken seriously by the scientific community, companies and regulators, and these actors should be better equipped to understand and engage in a productive debate with the public and the organizations concerned with nature preservation and animal defence. A higher level of salience, independence, transparent regulatory procedures, openness and accountability of these agencies and public bodies could significantly increase the level of public trust in the authorization of specific pharming applications (both processes and products). It could decrease the immense cognitive demands on individuals for personally evaluating the uselfulness, risks and implications of advances in pharming. In doing so it could reduce the potential both for adversarial and promotional campaigns on pharming mounted by single-issue organizations and lobbies. The scientific community and the policy makers should foster transfer of knowledge of biology and genetics to the public, a climate of open dialogue with society at large, and a higher level of salience, independence, transparency, and accountability of regulatory agencies.

9.2.2 Moral evaluation In assessing the moral concerns expressed in public attitudes and in the bioethical literature, the goal of the chapter on the moral evaluation of pharming has not been to search for true or objective answers. Rather, the aim is to come up with recommendations for a morally acceptable, but also socially accepted development of the field of pharming. The focus has been on conflicts about pharming caused by discrepancies between far-reaching medical and economic hopes, public attitudes, and moral concerns regarding amongst other things the moral status of animals and plants, the naturalness and 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 pharming have been analysed. Having taken into account both the argumentative strength of the various criteria proposed for the moral evaluation of pharming and their potential for mastering conflicts on pharming, it is proposed that i) in developing pharming, more weight should be given to avoiding the infliction of suffering on animals, compared with concerns regarding the unnaturalness of pharming or the infringed integrity of pharming animals and plants, ii) in evaluating pharming projects one should take into account the specific aims for which the animals and plants are used, and iii) in view of the scientific uncertainties in performing a systematic risk-benefit assessment of

294

9 Conclusions and recommendations

both animal and plant pharming, there should be a careful case by case analysis of pharming projects using the precautionary principle. The moral evaluation of pharming should consider: – the potential suffering inflicted on pharming animals; – the purpose of, and need for, the specific products for which the animals or plants are used; – the precautionary principle in view of the available scientific evidence and its limitations, and the difficulties in performing a systematic risk-benefit assessment of both animal and plant pharming.

9.3 The assessment and management of risks associated with pharming 9.3.1 Principles 9.3.1.1 Case by case

A clearer regulatory framework for pharming is needed, not only in order to decrease economic risks but also in order to effectively manage risks to humans, animals and the environment. The striking problem in risk assessment is that pharming plants and animals lead to new combinations of trait (transgene), organisms, and environment, so that it is hardly possible to draw on existing experience. Assessment and management of risks associated with plant or animal pharming have different emphases – welfare in the case of animals, risks to the environment in the case of plants. This is reflected in the sections below. In plants, unless the transgene modifies traits associated with reproduction - which is rarely the case - the mode of gene dispersal will not be changed within a given species, and therefore the dispersal of pharming plants is identical to that of non-GM plants and other GM plants. However, the fact that proteins are intentionally produced in very high concentrations in pharming plants makes it much more complicated to evaluate the probability of possible unwanted effects. This is also true of transgenic animals, leading to animal welfare concerns. Therefore, although general recommendations on risk assessment of pharming plants and animals would probably fail to address all pharming plants and animals in an appropriate way, we will below propose some general measures in addition to the case by case approach necessitated by the diversity of scenarios. The risk assessment of pharming requires a case by case approach. However, we also propose some general measures for handling high levels of uncertainty.

9.3 The assessment and management of risks associated with pharming

295

9.3.1.2 Risk-benefit evaluation

Risk assessment of pharming plants and animals can be quite uncertain. This uncertainty calls for a broad regulation of GM pharming plants and animals, in which ethical and social considerations regarding the production should accompany the traditional natural science-based risk assessment. This is in particular true of highly bioactive products. Both with respect to experimental releases and cultivation of pharming plants and experimental releases and the keeping of pharming animals, the grant of authorization should be conditional not only on a positive risk assessment considering proportionality between risks and costs, but also on a positive risk-benefit evaluation. Such an evaluation is an expression of the principle of “ecological proportionality”. It is already recognized in the regulation of pharmaceuticals, pesticides, biocides, and particularly dangerous chemicals for general use, and should also be introduced with regard to pharming activities. Risk-benefit evaluation – not only a risk assessment – should be made a precondition of authorization, both with respect to the experimental and the production phase. 9.3.1.3 Independent risk assessment research

Presently, the authorities base their risk assessment mainly on experimental results provided by the GM producer. Allocation of more funds to independent risk assessment research is necessary; presently these research activities are minimal. In order to provide an independent assessment and research on potential risks associated with pharming, researchers who are not dependent in any way on transgenic plant or animal producers or associated parties should have better access to material on pharming GM plants and animals. Procedures regarding the conditions of access and protection of trade secrets must be worked out based on a public discussion. Access to information necessary for independent risk assessment research should be improved and better funding should be made available for this research. 9.3.1.4 Transparent procedures and independence of risk assessment bodies

Although existing law affords concerned citizens an opportunity to comment on a proposed release and gives them access to information relating to the risk assessment, the present level of transparency in the risk assesment

296

9 Conclusions and recommendations

procedures is not sufficient to ensure informed and effective participation. The quality of information on the risk assessment procedures available to the public should be higher. There should be access not just to monitoring plans, but also to monitoring results. Arguments on which the evaluation of risk and decisions on risk management are based, including assumptions regarding science, values, world views etc. should be made public. Avenues that the public have for expressing their opinion should be improved. An open register giving sites of releases and production should be available in all EU member states. It should be mandatory to inform adjacent farmers about planned releases. However, trade secrets must be protected. This is in particular necessary in the experimental phase. Confidential gene constructs deserve protection throughout the development and production phases. In view of concerns about trust in the institutions, members of the EU and national risk assessment bodies should be required to declare their associations with private companies (all collaborative projects, committee memberships, etc.) when assessing particular products, and if advisable, be prohibited from making particular assessments. Procedures ensuring such higher levels of transparency should be worked out for the risk assessment bodies based on a public discussion. There should be stronger requirements for transparency in risk assessment procedures.

9.3.2 Product safety and information 9.3.2.1 Measures to prevent contamination and ensure product quality

The development and production of pharming medicinal products does not only require the protection of the environment from risks associated with the release of GMOs, but also the protection of the production process from risks originating from the environment, for example viral and other infectious agents, or pesticides. The requisite safety measures for ensuring an appropriate safety and quality of developmental and final pharmaceutical products are not necessarily such that an open release with confinement cannot be used under pharmaceuticals regulations. The producer should be allowed to lay the emphasis on safety and control measures regarding the gaining and processing of the crude bulk material and the manufacture of the medicinal product. Strict containment may in some, but not all, instances be necessary. From economic and animal welfare aspects, containment would be a disadvantage.

9.3 The assessment and management of risks associated with pharming

297

Other potential quality and safety problems associated with the use of transgenic plants and animals for the production of pharmaceuticals, such as instability, immunogenicity, impurities and host cell contaminants can be tackled by applying the normal authorization prerequisites and procedures. Adequate safety measures must be taken in order to prevent the contamination of pharming products by environmental factors, such as viral agents. If production is not performed with containment, special care is required regarding the gaining and processing of the crude bulk material and the manufacture of the finished pharmaceutical. 9.3.2.2 New guidelines on pharming medicinal products and European Medicines Agency (EMEA) committee on pharming products

In view of more recent scientific and economic developments in the field of pharming, the existing EMEA guidelines are outdated. To ensure product safety and quality also of pharmaceuticals that do not consist of, or contain, GMOs but are derived from recombinant protein, the adoption of modern guidelines is warranted. A special committee on pharming should be established within the agency in order to ensure a greater familiarity of the EMEA with the specific problems of pharmaceutical quality and safety (for example aspects of immunogenicity) presented by pharming products. New EMEA guidelines on pharming products are needed. A special committee on pharming should be established within the agency. 9.3.2.3 Labelling and consumer information

To afford patients and doctors a free choice of pharmaceuticals within the constraints of the public health system, two measures are advisable. Firstly, labelling containing information on the relevant production method should be permissible and encouraged by the authorities, as it is unclear whether this can be done under existing law. Secondly, there should be a general individual right to information on the production method against the manufacturer. Voluntary labelling should be permissible and encouraged. Individuals should have the right to information on the production method.

298

9 Conclusions and recommendations

9.3.3 Risks to the environment and food and feed chains 9.3.3.1 Experiments and cultivation with containment, and deliberate releases

Two legal avenues can be used for the production and use of genetically modified organisms: experiments and cultivation with containment under the contained use directive, or deliberate release under the release directive. In case of production in the open environment it should be clarified if a part B procedure under the release directive (normally used for field experiments) or a part C procedure (an authorization for placing on the market) must be secured. Compared with plants, accidental escape of large animals (excluding fish) is unlikely, and there are good chances of retrievability. Therefore, the rest of this section is concerned with plants. In most cases some degree of containment or confinement is imposed for experimental releases or regular cultivation of GM plants. A residual risk of transgene leakage remains in all cases. This means that dispersal of plants or plant parts (for example gametes and seeds) is always a possibility. However, the different containment, confinement and isolation measures offer differing degrees of safety. As regards risks to the environment, including risks to human health, there is a clear choice between development and production in strict containment (bioreactors, special glass-houses or closed animal units) or in the open environment provided that adequate genetic containment, or confinement, and protection measures are taken. In some cases of particularly high risk or great uncertainty, development and production under strict containment may be warranted. However, the imposition of strict containment is not warranted on a general basis. It will always be the combination of the transgenic character, the recipient plant and the receiving environment that determines, together with the principle of proportionality, to what extent, and which type of, containment or confinement should be applied from the point of view of ecological risks. All types of genetic or physical plant containment, confinement and isolation systems may be leaky at some point. The different containment, confinement and isolation measures can offer differing degrees of safety. The choice between completely avoiding release or applying more or less stringent containment, confinement and isolation measures should be guided by factors such as the degree of risk and uncertainty, the safety that can be achieved and proportionality. It is necessary to be extra cautious particularly during the first experimental releases of a pharming plant, as knowledge about the interaction between the pharming plant and the environment is probably limited at

9.3 The assessment and management of risks associated with pharming

299

this point. During the release such information should be obtained, provided that this does not jeopardize the confinement measures. Especially analysis of non-target effects of exposure to the release site and environment should be measured. As knowledge about gene flow can be gained from equivalent non-transgenic lines and applied to the GM plant, gene dispersal aspects do not have to be studied in releases, and large isolation distances or other types of confinement can be applied to the field trial. The risk of commingling should be minimized, until risks have been thoroughly evaluated. Staff involved in pharming plant releases should be required to receive special training in how to minimize potential personal and environmental risks. More and longer monitoring should be a recommendation for both field trials and production fields. Monitoring should continue after termination of the release. In this way, experience can be accumulated on which better risk assessment and management can be built. Extra strict confinement may, if technically possible, include the use of crops not used for food and feed. This confinement strategy will reduce the risk of both commingling and gene flow, especially if taxonomic relationships are weak between the pharming crop and food or feed crops and wild relatives in the area. Whether this non-food and non-feed strategy should apply only to the production of highly bioactive pharming products or should be a general precautionary measure for all pharming plants can be debated. Another confinement strategy may be zoning, an extreme variant of physical isolation by distance. During deliberate pharming plant releases for development and regular production, extra high safety standards should be enforced including: – education of personnel; – extra strict confinement (for example zoning); – total genetic containment in some cases; – more and longer monitoring; – research on the interaction between the pharming plant and the environment (where applicable). The present guidelines for GMOs were not developed to apply to pharming but only to food and feed crops. As all containment or confinement measures may be leaky at some point, the authorization of experimental releases and regular production entails a decision on the acceptability of risks associated with pharming. Even where the competence for taking this decision is vested in the member state authorities, common guidelines are advisable to protect health and the environment because pharming presents novel problems of risk assessment and risk management. Guidelines could collate the scattered experiences with pharming and put them at the disposal of all relevant authorities as well as researchers and industry.

300

9 Conclusions and recommendations

The guidelines should aim for minimum harmonization only, leaving the member state authorities an ample margin of discretion to take more stringent precautions. Common guidelines for pharming plant experimental releases and regular cultivation should be developed, providing a harmonized basis for risk assessment at national and EU levels. 9.3.3.2 Coexistence

The potential economic harm that organic and conventional farmers in the vicinity of fields with transgenic crops, including pharming crops, may suffer due to gene spread and commingling does not differ whether they are exposed to an experimental release or to a permanent cultivation. Therefore the rules on coexistence, which presently only apply to permanent cultivation, should be extended so as to also cover experimental releases. Moreover, as with the risks to health and the environment associated with releases, common EU guidelines for ensuring coexistence between organic and conventional agriculture and pharming are advisable. However, member states should retain an ample margin of discretion for considering particularities of their country, such as soil and climatic conditions, agricultural practices and structure of land holdings, as well as for deciding on the appropriate level of protection of the economic interests involved. Common EU guidelines for ensuring coexistence between organic and conventional agriculture and pharming are advisable.

9.3.4 Risks to animals in pharming Risks to animals used in pharming can be divided into those that arise during the experimental phase, and those that arise during the production phase, where offspring are used to produce the protein for commercial use. Good laboratory practice in the experimental phase, and good farming practice throughout, are already legal requirements and include attention to species-adequate housing and management conditions, priority for noninvasive options for laboratory procedures, and trained personnel. Nevertheless, and although this is not inevitable and not the intention, generating and using animals for pharming purposes potentially causes them to suffer. In view of the moral principle of avoiding animal suffering wherever possible, which is advocated in this book, emphasis should be placed on animal welfare risk assessment and management, and on the current lack of knowledge with regard to the effects of transgenesis on ani-

9.3 The assessment and management of risks associated with pharming

301

mal welfare. To gain more knowledge in the experimental phase, welfare parameters must be included in phenotyping protocols in order to carefully document unintended side-effects of transgenesis, including subtle dysfunctions. Continued epidemiological research (lasting into the production phase) is warranted in order to detect potential long-term consequences of transgenesis. The risk assessment carried out before embarking on an animal pharming project should take into account not only potential welfare problems related to the experimental phase, but also those related to the envisaged production. In line with the above mentioned general principle of also considering the benefits, it should be noted that the legal status of cost-saving benefits in risk-benefit evaluation and the ethical review of animal trials and intervention in production animals is unclear. It should be clarified whether considerable cost savings for patients and collective social and private insurance institutions, as well as a speedier mass production of medicinal products, are factors that may be considered to be benefits in risk-benefit evaluations and the ethical review. The current lack of knowledge with regard the effects of transgenesis on animal welfare should be taken into account in: – precautionary measures, for example the animal welfare risk assessment carried out before an animal pharming project, which should also consider the potential welfare problems related to the production phase; – facilitation of animal welfare research – including during the experimental and production phases – which is needed to fulfil even current animal protection legislation. The production phase of animal pharming should be covered by the regulations on the keeping of animals used for agricultural purposes, because animal pharming is similar to the farming of animals for leather and fur and can therefore be defined as an agricultural activity. Animal pharming in the production phase should be covered by the regulation on keeping of animals for agricultural purposes.

Glossary

3’ end, 5’ end

Single DNA strands are asymmetric, having a 3’ and 5’ end. These terms are commonly used to denote the orientation of DNA elements within or close to a gene.

Agamospermy Asexual seed formation Amniocentesis or allantocentesis Procedures to obtain samples of cells shed by a developing foetus into the surrounding fluids. Aneuploid Having a chromosome number that is not a multiplum af the haploid chromosome number for the species. Anthropocentrism The position that only humans, having certain specific traits (e.g. rationality), qualify as bearers of moral entitlements or as having moral value in themselves. Bacillus thuringiensis A Gram-positive, spore-forming soil bacterium. The gene for a toxin from Bacillus thuringensis is effective against insect pests. The gene has been inserted into several species of plants by genetic engineering. Biocentrism The view that not only sentient beings but all forms of life have a certain value. Biopharmaceuticals Pharmaceutical proteins that are too complex to be produced by conventional chemical synthesis, and are either isolated from biological material (i.e. blood plasma) or produced in living cells (e.g. monoclonal antibodies or blood clotting factors). Blastocyst A pre-implantation mammalian embryo of about 150 cells, comprising a ball of cells enclosing a fluid-filled cavity and an inner cell mass that later forms the embryo.

304

Glossary

Blastoderm

An early stage of avian development in which the embryo is a f lat disc of cells. Blastomere An individual cell of an early stage embryo. Cell cycle The sequence of stages between one cell division and the next. These are: gap 1 (G1) DNA synthesis (S), gap 2 (G2) and mitosis (M). Chimera An organism composed of a mixture of genetically distinct cells that originate from two individuals. Chromatin The natural form of genomic DNA within the nucleus. A highly structured multi-coiled fibre composed of a complex of DNA, RNA and proteins. Clone

a) Molecular clone: an isolated DNA fragment propagated artificially, e.g. in a bacterial plasmid. b) Cell clone: a group of genetically identical cells descended from a single individual. c) Animal or plant clone: an animal or plant whose genetic information is identical to that of an individual from which it was created.

Cloning vector A DNA molecule used to propagate and manipulate fragments of molecular cloned DNA. Vectors are frequently derived from plasmids or bacteriophages. Coding sequence

The portion of mRNA that is translated into protein. Codon A group of three bases in an mRNA sequence that specifies a particular amino acid when translated into a protein sequence. Complementary DNA (cDNA) A DNA copy generated from an RNA template molecule using the enzyme reverse transcriptase. Confinement The use of physical or agricultural management measures to prevent the unwanted escape of an organism in an open space e.g. plants in a field – see containment.

Glossary

305

Contained use

Activities that entail the use of GMOs with containment, i.e. physical barriers or a combination of physical with chemical and/or biological barriers that are capable of limiting the contact with, and providing a high level of safety for, the general population and the environment. Containment The use of closed spaces (greenhouses, indoor growth-facilities or laboratories, mines) to control the spread of an organism – see confinement. Genetic containment of the transgenic donor means that genes control dispersal (abort pollen production, abort flower production, provide seed sterility etc.), so that sexual reproduction is impossible or strogly reduced. Cre-lox recombination A system of enzyme mediated site-specific recombination derived from the bacteriophage P1 that can be used to delete selected DNA regions. Cytokines A group of hormones that act as signals between cells and affect cell behaviour and properties, e.g. the interleukins in the immune system. Cytoplasm The portion of a eukaryotic cell that is outside the nucleus. Deoxyribonucleic acid (DNA)

A long chain double-helical nucleic acid that contains the genetic information in the cell and provides a template for RNA synthesis. Development (developmental) medicinal product A medicinal product that is under development but must already comply with certain requirements with certain pharmaceutical regulations. Differentiation

A process by which a cell takes on a more specialised role or function. Enhancer A genetic element which positively affects the transcription of a gene. Enucleation Removal of the nucleus from a cell, e.g. an oocyte in preparation for nuclear transfer. Epididymis The duct through which sperm pass from the testes.

306

Glossary

Epigenetic regulation

Regulation of gene expression by modification of DNA not involving changes to the base sequence, e.g. by methylation of bases. Ethics Method for the critical examination of morals. It is one of the main tasks of ethics to work out rules for moral discourse and to test if these rules are adequate for mastering moral conflicts. Exon Part of the transcribed region of a gene that is retained after RNA splicing. Fibroblast A cell type involved in the production of connective tissue. Fluorescent in situ hybridization A technique to identify and locate complementary regions of DNA or RNA within a cell or chromosome by hybridisation with fluorescently labelled DNA or RNA labelled probe molecules. Founder A first-generation transgenic organism that can be used to found a line. Gamete An egg (oocyte) or sperm Gene targeting A technique to engineer specific changes in a particular DNA region in situ by homologous recombination. Genome The total genetic content contained in a haploid set of chromosomes (in eukaryotes). Mitochondria and chloroplasts also have small amounts of DNA. Genomic DNA Usually refers to DNA isolated from the nuclear chromosomes. Germ cell A cell of the germ line: oocytes, sperm and their precursors Glycoprotein A protein with carbohydrate molecules (sugars) that are covalently attached. Glycosylation The addition of carbohydrate molecules to a protein.

Glossary

307

Green biotechnology

Metaphor commonly used to denote biotechnology applied for agricultural and environmental purposes, e.g. genetic modification of plants make them tolerate spraying with herbicides. Haematopoietic Involved in the formation of blood and blood cells. Hemizygous

Having a single copy of a particular transgene. Heterochromatin Regions of condensed chromatin, generally transcriptionally inactive. Heterozygous Having two different forms of a particular gene or DNA region. Holism The position (sometimes also called ecocentrism or physiocentrism) assumes that, in addition to animals and plants, also anorganic structures like mountains or rivers have value not merely as a means for human aims, but have value in themselves. Homologous recombination Reciprocal exchange of strands between different DNA molecules at regions of identical (or close to identical) sequence. Homozygous Possessing two identical forms of a particular gene or DNA region. In vitro

In an artificial environment, e.g. in a laboratory dish or test tube, as opposed to in vivo.

In vivo

In the living organism, as opposed to in vitro.

Insulator A DNA element that protects a DNA region from the influence of its chromatin environment. Integrity Integrity draws on notions of ‘wholeness’ and ‘completeness’ of animals, their ‘species-specific balance’, and their ‘capacity to maintain themselves independently’. The aim of this position is to develop an ethical principle going beyond the (pathocentristic) evaluation of an animal’s well-being.

308

Glossary

Interspecific

Among different species Intra-cytoplasmic sperm injection (ICSI) Artificial fertilization of an oocyte by microinjection of a sperm or sperm head into the cytoplasm. Intraspecific Within the species Introgression The stable incorporation by sexual crosses of genetic material from one gene pool into another, differently composed gene pool. Intron Portions of genomic DNA that are transcribed but later spliced out and do not form part of the mRNA. Lactoferrin A multifunctional protein with antimicrobial activity. Lactoferrin is found in milk and other mucosal secretions Lentiviral transduction Viral transduction using a type of retrovirus termed a lentivirus. Locus A defined position or location of a gene in the genome. Meiosis

Cell division during formation of gametes that results in reduction of the normal chromosome complement by half.

Mesenchymal stem cell

A stem cell from the developing connective tissue. Metaphase A stage of cell division in which chromosomes are condensed, ready to be apportioned to the daughter cells. Minigene A recombinant construct in which one or more introns of a gene are combined with cDNA to partially reproduce the genomic structure, usually to improve expression. Moral A set of convictions, norms and recommendations to act that is subject to multiple cultural and historical changes.

Glossary

309

Mosaic

An individual composed of two or more types of genetically different cells. mRNA The spliced mature RNA transcript containing the protein coding sequence that is transported out of the nucleus to be translated. N, C terminal N and C terminals refer to the opposite ends of a polypeptide chain. Naturalness Arguments drawing on naturalness are ubiquitous in academic literature and deeply entrenched in everyday language. One should distinguish between genetic naturalness referring to the process of formation of the entity that we qualify as natural or unnatural and qualitative naturalness referring to the current composition or appearance of that entity. Northern hybridization A technique by which RNA species are size separated by gel-electrophoresis, transferred and bound to a membrane and their complementary regions identified by hybridization to labelled DNA or RNA “probe” molecules. Nuclear transfer A technique that combines an enucleated egg and a cell nucleus to make an embryo. Nutraceutical A food product with medical properties. Oligosaccharide A chain of sugar molecules. Ooplasm

The cytoplasm of a mammalian oocyte. Pathocentrism The position that the capacity to feel – and not the fact of being human or having specifically human traits – is the decisive criterion for having moral entitlements or having moral value in themselves. Peptide

A short chain of amino acid residues, typically 10-30 in number. Perivitelline space The narrow fluid filled space between the cell membrane and the Zona pellucida of a mammalian oocyte.

310

Glossary

Pharming

The use of higher organisms (plants and animals) for the production of recombinant biopharmaceuticals. Phytomanufacturing Recombinant protein production utilizing plant-based platforms (plant factories). Plant-made industrial product (PMI)

Non-pharmaceutical enzyme, biopolymer or other recombinant compound with industrial properties produced by transgenic plants. Plasmid A circular DNA molecule capable of autonomous replication, distinct from the main genome of bacteria. Pluripotent

The ability of a single stem cell to generate all cell types of the body. Pluripotent cells cannot make extra-embryonic structures, e.g. the amniotic sac or the placenta. Polyadenylation The addition of multiple ‘A’ ribonucleotide residues to the 3’end of mRNA, which takes place during mRNA maturation. Necessary for mRNA stability. Polymerase chain reaction A technique for generating multiple copies of a piece of DNA in vitro. Reverse transcriptase PCR (RT-PCR) is a variant in which an RNA molecule is first converted to cDNA and then copied. Polypeptide A chain of amino acid residues. Referring to the amino acid chain of a protein. Position effect Variation in the pattern and level of transgene expression between animals or plants carrying the same transgene integrated at different genetic loci. Precautionary principle Principle of political and administrative decision making whereby regulatory action can be taken before there is secure scientific evidence that an activity will cause harm to health or the environment, especially if the harm could be very serious or irreversible. The precautionary principle requires, at a minimum, that there is some scientifically plausible reason to believe that such harm may occur. It is laid down in article 174 (2) of the EC Treaty.

Glossary

311

Promoter

A sequence close to the 5’ end of a gene that is required for initiating transcription. Pronuclei Two structures formed from the sperm and oocyte genetic material following fertilisation, and which later fuse to form the embryo nucleus. Propagule

Any plant material used for the purpose of propagating the plant: it can be a seed, a bud, a tuber, a runner etc. Protozoans

Single-celled eukaryotes which are commonly mobile and display heterotrophy (requires organic substrate for growth and development).

Recombinant

A gene or protein that is not derived from the species in which it is expressed. Red biotechnology Metaphor commonly used to denote biotechnology for production of pharmaceuticals or for medical purposes; e.g. genetic modification of rice to produce human lactoferrin. Reporter gene A gene that provides a convenient means of identifying its expression, e.g. by producing fluorescence, light, or a stainable product of enzyme activity. Reprogramming A radical change in the repertoire of genes expressed by a nucleus in response to a change in cytoplasmic factors, e.g. after transfer of a nucleus from a somatic cell into an oocyte. Reverse transcriptase An enzyme that produces a DNA copy from an RNA template. Ribonucleic acid (RNA) A single stranded nucleic acid similar to DNA. Messenger RNA (mRNA) is formed by transcription on a DNA template and carries genetic information from the cell nucleus to the cytoplasm where protein synthesis takes place. Risk assessment A process based on natural science to determine the risk presented by an activity. It consists of four steps: hazard identification, hazard characterisation, exposure assessment and risk characterisation.

312

Glossary

Risk management

Evaluation of acceptability of risk and decision on the taking of measures to exclude the risk or to reduce it to a degree which is deemed to be acceptable. Risk Probability of future harm expected to be caused by an activity due to its intrinsically hazardous properties and exposure of people or the environment. Risk-benefit evaluation A process of weighing the risk presented by an activity with the potential benefits to be derived from it in order to decide on acceptability of the risk. Ribonucleic acid (RNA) A single stranded nucleic acid similar to DNA. Messenger RNA (mRNA) is formed by transcription on a DNA template and carries genetic information from the cell nucleus to the cytoplasm where it is translated into protein. RNA splicing Processing of the primary RNA transcript to remove intron sequences. Somatic cell Cells of an organism other than the germ line Southern hybridisation A technique by which DNA fragments are size separated by gel-electrophoresis, transferred and bound to a membrane and their complementary regions then identified by hybridisation to labelled DNA or RNA “probe” molecules. Splice sites

Sequences on the primary RNA transcript that define the regions to be removed during splicing.

Stem cell A cell with the ability to divide indefinitely and also give rise by differentiation to more specialised cells. Tapetum

Nutritive layer of cells that lines the inner wall of the pollen sac

Tetraploid

Having four copies of each chromosome, twice the normal number.

Glossary

313

Transcription

The first stage in the expression of a gene in which an RNA copy, the primary transcript, is made from DNA. This transcript contains exon and intron sequences. Transfection

Loose term covering a variety of methods of introducing nucleic acids into cells.

Transgene A foreign gene or DNA region (artificially produced or from another organism) artificially inserted into the genome of an organism. Transgenesis The process of producing a transgenic organism. Tropism:

For viruses – the range of cells that can be infected, host cell specificity. Undifferentiated Relating to a cell that is capable of differentiation, but has not generated specialised structures or proteins characteristic of a specialised cell type. Viral supernatant

Culture medium containing viral particles.

Viral transduction Transfer of a DNA fragment from one cell to another by means of a virus particle. White biotechnology Colour metaphor commonly used to denote biotechnology applied for industrial use; e.g. genetic modification of potatoes to produce special types of starch. Zona pellucida A transparent, shell-like membrane formed around a mammalian oocyte as it develops in the ovary.

Appendix: Examples of GM pharmaceutical crops and animals

I.

Production of molecular farmed human intrinsic factor (rhIF) in potato (Solanum tuberosum)

Cobento Biotech A/S is a Danish company that has developed technology for molecular farming in potato tubers and in Arabidopsis. The inserted gene encodes for the B-12 binding human intrinsic factor (rhIF). The human intrinsic factor is necessary for uptake of vitamin B-12. It is estimated that B-12 deficiency is a problem for 10-15 % of the population above 60 years of age. Vitamin B-12 is found naturally in animal but not in plant products. The recommended daily dose is 2 micrograms. The reason for the low daily dose is that the body recycles the vitamin, as it is excreted from the gall bladder to the intestine and taken up again. B-12 is very difficult to absorb from the intestine. There do not seem to be any adverse effects if an excess of vitamin is ingested, as only small amounts (1 %) of excess vitamin B-12 will be absorbed. The current production of rhIF is based on tissue from animal stomachs, and the product is injected into patients. This production form is quite inconvenient and, therefore, the strategy to produce rhIF in plants seems quite attractive, especially so because the plant parts can be used directly as tablets for oral intake, and the price will be reduced compared to the current situation. Analysis: Molecular farming of human intrinsic factor, rhIF, in potato tubers: – Production platform: Production can occur in potato tubers, which may be conveniently harvested, transported and stored for future processing. Tubers will be harvested by dedicated machinery, homogenized and the protein fraction recovered and purified using standard procedures. The plant tissues can also be used directly without extraction of the protein. Molecular farmed potatoes will be produced under contract. Varieties that are easily distinguishable morphologically from conventional cultivars can be used. – Scale of production: The company estimates that meeting the entire market demand for human intrinsic factor would require a production area of approximately 100 ha.

316

Appendix: Examples of GM pharmaceutical crops and animals

– Risk of gene transfer: The risk is very low. Potato has no sexually-compatible wild relatives in Europe (but they do in South America). Outcrossing and viable seed set are generally low for potato, and propagation for commercial potato production is vegetative, even though some plants will flower. Tubers and berries are sensitive to frost. In the case of food crops, 20 m of isolation distance is considered sufficient to reduce pollen flow between fields to very low levels (Danish regulation on co-existence, Departmental order, no. 220 31.03.2005). Machinery should be cleaned and volunteers controlled. – Risk to humans, animal health and environment: It as assumed by specialists in the uptake of vitamin B-12 (pers. com. Dr. Ebba Nexø, Aarhus University Hospital, Aarhus, Denmark) that unintended intake of the GM potatoes will have no adverse health effects, as the excess rhIF will be excreted. This has been shown in humans. The company has tested effects (dose-response curves) on rats, and there were no effects observed, and excess rhIF was excreted. Impacts on other non-target organisms should be assessed, however. – Rating and mitigating measures: There is a minimal potential for effects on the environment. May require limited physical confinement and no genetic containment.

II. Production of Molecular Farmed human lactoferrin (rhLf) in rice (Oryza sativa) Ventria Bioscience (formerly Applied Phytologics) has developed rice varieties that produce pharmaceutical proteins (more conveniently named “value-added proteins for human consumption” by some). From 1997–2004 Ventria has grown its pharmaceutical rice in field tests in the northern Central Valley of California (except perhaps in 1999 and 2002). Ventria proposes to add human lactoferrin to yoghurt, meal replacements and performance beverages and in nutritional supplement drinks. Human lactoferrin is claimed to have beneficial antimicrobial activity. As human lactoferrin can be used as a supplement to food products, it must be assumed to have a slow degradation. Analysis: Molecular farming of human lactoferrin (rhLf) in rice: – Production platform: Cultivated rice is used as the production platform. The rice should be harvested by dedicated machinery and stored in special facilities. Rice grains can be stored until extraction of the lactoferrin is needed. Molecular farmed rice will be produced under contract. – Scale of production: The total size of Ventria’s field trials was 37 ha in 2003, but it must be expected that if commercial production is initiated, the area will be much larger.

III. Production of antithrombin in goats’ (Capra hircus) milk

317

– Risk of gene transfer: Rice is mainly self-pollinating, but a low percentage of outcrossing takes place by wind. In the area where Ventria had their field releases, there is a substantial production of conventional rice, so Ventria’s GM rice might have outcrossed with this, especially as the isolation distance to conventional rice was less than 70 m. Also found in the release area is the weedy rice (Oryza rufipogon), a strongly competitive plant that is difficult to eradicate. Recently, the weedy red rice (an eco-type of Oryza sativa) has also been introduced to the area. This rice is considered a serious risk to the Californian rice industry. The cultivated rice and weed Oryza rufipogon have a rate of spontaneous intercrossing of 1–3 %1; hybridization frequencies between cultivated rice and red rice are usually below 1%. The grains of wild rice ripen before those of the cultivated rice and are extremely prone to shattering before harvest. Ventria’s field tests were apparently not netted or fenced. As birds, mice and other rodents feed from rice grains, these herbivores would represent risks of grain dispersal. – Risk to humans, animal health and environment: There is a possibility that human lactoferrin is an allergen, as it shows homology to bovine lactoferrin, which is considered to be allergenic. Ventria refers that chicken fed with the rhLF rice had improved health and growth rates compared to control2. If this is a general phenomenon, wild birds that feed from the rhLF fields may increase their population sizes and possibly affect the ecosystem. It has been reported that certain pathogenic microbes can feed from the iron in the lactoferrin, and in that way animals with mild infections of these pathogens may be severely affected. The rice material must be segregated from food and feed supplies. Effects to the wider environment are largely unknown, e.g. the impact on wild rice populations and herbivores has not been investigated. – Rating and mitigating measures: Preferably lactoferrin rice should be produced in zones where no other rice is found. All herbivores should be fenced away from the production area, and the production area burned to destroy all remaining seeds. In conclusion, production of human lactoferrin will require strong physical confinement and preferably also zoning.

III. Production of antithrombin in goats’ (Capra hircus) milk GTC Biotherapeutics Inc. (Framingham, Massachusetts, USA) is a company that specialises in animal pharming. Its lead product is ATryn®, which contains recombinant human antithrombin III (antithrombin alpha, ATIII), purified from the milk of transgenic goats. 1 2

Chen et al. 2004. Humphrey et al. 2002.

318

Appendix: Examples of GM pharmaceutical crops and animals

Antithrombin is a blood protein that prevents blood-clotting. The prevalence of congenital antithrombin deficiency is about 1 in 3,000 to 5,000 in the general population. Substitution with therapeutic antithrombin is considered for the affected individuals in risk situations such as surgery or childbirth, and for treatment of thrombosis. Antithrombin may also be indicated for patients with acquired antithrombin deficiency. Human antithrombin has been difficult to produce with the traditional recombinant protein production methods, and has therefore previously been available only from donor blood. It can be produced relatively cheaply in large quantities in the goats. GTC Biotherapeutics has initiated procedures for authorization of the product in the USA and has completed phase III clinical trials there. Since 2006, ATryn® has been authorized in Europe for the prophylaxis of venous thromboembolism for patients with congenital antithrombin deficiency undergoing high-risk surgery3. The present authorization holder in Europe is LEO Pharma, Ballerup, Denmark, who is currently starting to market the product. LEO Pharma is also developing ATryn® for other indications and is conducting a clinical phase II study using ATryn® to treat an acquired antithrombin deficiency, disseminated intravascular coagulation4. Analysis: Production of antithrombin III in goats’ milk: – Production platform: Production takes place in the mammary glands of dairy goats. The method of gene transfer is pronuclear microinjection. GTC Biotherapeutics’ facilities contain buildings for the experimental phase of producing transgenic lines, as well as for kid-rearing, general animal housing, milk collection, and processing. – Product: The frozen milk is shipped to another facility for purification (Cambrex BioScience, MA, USA), and the formulated bulk of drug is shipped to MedImmune BV, Nijmegen, The Netherlands, from where the final product is supplied to LEO Pharma. – Risk of gene transfer: The risk of vertical gene transfer is very low, as it is unlikely that goats will escape, not be caught, and subsequently survive and breed in the wild. To avoid such incidences, the goats are kept contained. Very little is at present known about potential horizontal gene transfer. – Risk to humans and environment: Risks to the environment would be related to potential horizontal gene transfer, about which there is very limited knowledge. The risk to consumers of milk and milk products is negligable, as the milk is kept separate from other dairy, and it is not likely that it would inadvertently be consumed by humans. According to the 3 4

For a detailed documentation of the product and authorization process, see documents at www.emea.europa.eu/humandocs/Humans/EPAR/atryn/atryn.htm. www.gtc-bio.com; www.leo-pharma.com.

III. Production of antithrombin in goats’ (Capra hircus) milk

319

Food and Drug Administration (FDA)’s guidelines5, nontransgenic animals associated with the herd, and animals in which transgenesis has failed, can be used for food or feed consumption in accordance with Food Safety Inspection Service (FSIS, the public health agency in the US Department of Agriculture) regulations. Transgenic animals can potentially be used as food or feed after a food safety assessment coordinated by the FDA’s Center for Veterinary Medicine (CVM). The ATryn®-producing goats are certified scrapie-free, and additions to the production herd are closely monitored with regard to disease. GTC Biotherapeutics keeps a “Transgenic Bank” comprised of founder males, their offspring and semen, to ensure stability of the product supply. Glycosylation patterns are not similar to human, but relatively close. ATryn® is not recommended for patients with known allergies to goats or goat products. For further details regarding the pharmacological riskbenefit analysis, see detailed documentation by the European Agency for the Evaluation of Medicinal Products (EMEA)6. – Risks to the animals: To ensure high levels of health and welfare, the GTC Biotherapeutics farm is accredited by the American Association for the Accreditation of Laboratory Animal Care International7. However, no published data are available on the health and welfare of the goats, or on the details of their housing conditions. As the product is in the production stage and the company keeps the “Transgenic Bank”, it is unlikely that experimental stage procedures need to be carried out to a large extent any more. At the experimental stage, with pronuclear microinjection, there were probably many “excess” animals that did not, or not adequately, express the transgene. Some were probably used by the company for other purposes (other product development, cross-breeding). In the process of testing whether first generation offspring have integrated the transgene, lactation was induced hormonally before sexual maturity. – Rating and mitigating measures: There is a minimal potential for effects on the environment. The production therefore requires standard limited physical confinement. There is some potential for animal suffering and a high number of excess animals, at the experimental stage, which is now probably largely completed. Little is known about the health and welfare characteristics or housing cicumstances of the production herds or “Transgenic Bank” animals; data on this would be useful for evaluation, and for potential suggestions that might improve animal welfare. 5 6

7

FDA 1995. For a detailed documentation of the product and authorization process, see documents at www.emea.europa.eu/humandocs/Humans/EPAR/atryn/atryn. htm (June 2008). Their guidelines are based on Institute of Laboratory Animal Research et al. 1996.

320

Appendix: Examples of GM pharmaceutical crops and animals

The clinical studies and market authorization have served to demonstrate the non-inferiority to plasma-derived antithrombin in preventing thromboembolisms. This is thought to be an important step for the further development of pharming. The need for the product itself seems to be more limited.

References

321

References Chen LJ, Lee DS, Song ZP, Suh HS, Lu BR (2004) Gene flow from cultivated rice (Oryza sativa) to its weedy and wild relatives. Annals of Botany 93:67–73 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) 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 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

List of authors

This book is written, discussed and integrated by the entire group. As a starting point, “seeding” texts have been provided by the members of the group. The authors who take primary responsibility for the different chapters are as follows: Chapter 1: Chapter 2: Chapter 3: Chapter 4: Chapter 5: Chapter 6: Chapter 7: Chapter 8: Chapter 9:

all M. Engelhard (2.1, 2.2), A. Schnieke (2.3) R. B. Jørgensen (3.1), K. Hagen (3.2) K. Hagen R. Pardo Avellaneda F. Thiele F. Thiele, E. Rehbinder E. Rehbinder all

Rehbinder, Eckard, is Professor em. Dr. jur. of Economic Law, Environmental Law and Comparative Law at the University Frankfurt/Main, and Co-Director of the Research Centre for Environmental Law at the same university. He studied law at the University Frankfurt/Main and the Freie Universität Berlin; LLD in 1965 at the former university with a doctoral thesis on “Extraterritoriale Wirkungen des Gesetzes gegen Wettbewerbsbeschränkungen”. Habilitation in 1968 at the University Frankfurt/Main with a comparative treatise on “Konzernaußenrecht und allgemeines Privatrecht”, published in 1969; venia legendi for Civil, Commercial and Economic Law, Private International Law and Comparative Law. From 1969 to 1972 he was Professor of Civil, Commercial and Economic Law at the University Bielefeld, between 1970 and 1972 also one the deputy rectors of that university. Since 1972 he has been Professor of Economic Law, Environmental Law, Civil Law and Private International Law at the University Frankfurt. In 1988 – following a reorientation of his research interests – the nomination of his chair was changed into Economic Law, Environmental Law and Comparative Law. In the academic year 1981/82 he was dean of the law faculty. Between 1987 and 2000 he was a member, between 1996 and 2000 also the chairman of the Council of Environmental Advisors (Sachverständigenrat für Umweltfragen). Since 1996, he has been a member of the Europäische Akademie Bad Neuenahr-Ahrweiler GmbH (Euro-

324

List of authors

pean academy for the research of consequences of scientific and technological advance) and participated in two further research projects of the academy. He has been a member of several national and international bodies in the field of environmental law. Between 2000 and 2006 he was Secretary General of the International Court of Environmental Arbitration and Conciliation. He was visiting professor at the Universities of Michigan (Ann Arbor), University of California (Berkeley) and several times at the European University Institute (Florence). Apart from the doctoral thesis and the habilitation treatise, he has published a number of books and numerous essays and was co-author of books and commentaries both in the field of economic law and environmental law; since the eighties, the emphasis has to an increasing extent been laid on environmental law. His interest has focused on general environmental law such as principles, economic and flexible instruments, environmental liability, judicial protection and comparative law, and regarding special areas in particular on toxic substances, genetically modified organisms, air pollution and protection of nature. He was awarded the Prix Elizabeth Haub in 1978 and the Bruno H. Schubert Prize in 2004. Postal address: Speckerhohlweg 3, 61462 Königstein, Germany Engelhard, Margret, Dr. phil., Dipl.-Biol., studied biology in Marburg and Edinburgh and graduated 1997 in micro- and molecularbiology at the Phillipps-Universität Marburg and the Max-Planck Institute for Terrestrial Microbiology. For this work she collected probes during an expedition to Nepal and discovered a novel kind of bacteria in symbiosis with rice plants that are, under certain conditions, able to fertilize this important crop with nitrogen. For her Ph.D.-thesis with Professor Boller at the University of Basel she continued to work on agriculturally relevant plants that are able to live in symbiosis with nitrogen fixing bacteria, and focused on the molecular crosstalk between these two organisms. For the genetic aspects of this work she stayed for half a year at the University of Geneva. In 2004 she was conferred a doctorate. Parallel to university, she started to work in science journalism and gained insights into pharmaceutical industry when she was admitted to the Swiss program “Women into industry”. In January 2004, Dr. Engelhard became member of the scientific staff of the Europäische Akademie and coordinated the project “Incentives for Organ Donation” from 2004 to spring 2006. Thereafter, she initiated and then worked for the project on pharming. Postal address: Europäische Akademie zur Erforschung von Folgen wissenschaftlich-technischer Entwicklungen Bad Neuenahr-Ahrweiler GmbH, Wilhelmstraße 56, 53474 Bad Neuenahr-Ahrweiler, Germany

List of authors

325

Hagen, Kristin, Ph.D., studied biology, philosophy and agricultural sciences at the University of Tromsø, Norway, and was subsequently a postgraduate student with Professor Broom at the Dept. of Clinical Veterinary Science, University of Cambridge, where she obtained her Ph.D. in 2001 with a thesis entitled “Emotional reactions to learning in cattle”. She went on to do postdoctoral research on cattle housing, behaviour and welfare at the University of Veterinary Sciences, Vienna, and at the Freie Universität Berlin. In 2006, she joined the Europäische Akademie. Postal address: Europäische Akademie zur Erforschung von Folgen wissenschaftlich-technischer Entwicklungen Bad Neuenahr-Ahrweiler GmbH, Wilhelmstraße 56, 53474 Bad Neuenahr-Ahrweiler, Germany Jørgensen, Rikke Bagger, Ph.D., is senior scientist and project leader at Risø National Laboratory for Sustainable Energy, at the Danish Technical University. Her main research areas are gene flow and introgression in plants, risk assessment and co-existence of transgenic plants, and effects to plants from a changing climate. She has a Master of Science and subsequently a Ph.D. from the University of Copenhagen in population genetics and evolution. During the first part of her carrier as at Risø National Laboratory she has had projects on plant breeding and genetic modification of plants. After this she got a position as head of the Office for Gene Technology and Variety Testing at the Danish Plant Directorate. When she returned to Risø in 1991 she became leader of the group for risk assessment and co-existence of GM plants. She is an author of more than 65 peer reviewed publications in international journal (for example in Nature and Molecular Ecology), several book chapters and a monograph. She has supervised 15 students on topics within risk assessment of transgenic plants and been responsible for a suite of different externally funded projects, and has participated in four EU projects (METALLOPHYTES, CONFLOW, ANGEL, SIGMEA) on gene flow and risk assessment of transgenic plants. Presently she is editor of Environmental Biosafety Research, appointed member of the Danish Ethical Council (by the Minister of Science, Technology and Innovation) and The Nordic Committee on Bioethics. Current projects deal with management of GM volunteer plants and effects of multifactor environmental stress (climate change) on plant fitness and gene expression. Postal address: Risø National Laboratory, Biosystems-309, Frederiksborgvej 399, 4000 Roskilde, Denmark Pardo-Avellaneda, Rafael, is Professor of Sociology and has been the General Director of the BBVA Foundation since 2000. From 1996–2000 he was Professor of Research at the National Council for Scientific Research (CSIC), and Professor of Sociology at the Universidad Pública de Navarra from 1993–96. He has held appointments at Stanford University as Visiting Professor (1998) and Visiting Scholar (1990–91, 1996), and as Vis-

326

List of authors

iting Scholar at the Massachusetts Institute of Technology (1987, 1988). He has been an advisor to General Directorate XII (Science & Technology) of the European Commission and to Intel Corporation in San José, USA. From 1994–96, he chaired the National Evaluation Commission for Social Science Research Projects, part of the Spanish National Agency for Evaluation and Science Policy. His major research and publications deal with organization studies, innovation, scientific and technological culture, social dimensions of Artificial Intelligence, cloning, and environmental culture and values. Postal address: Fundación BBVA, Paseo de Recoletos, 10, 28001 Madrid, Spain Schnieke, Angelika, is Professor of livestock biotechnology at the Technical University of Munich. She gained a Diplom in bioengineering at the Fachhochschule Hamburg and her Ph.D. for a thesis entitled “Cell-mediated transgenesis in livestock” from the University of Edinburgh. Her research interest is the genetic manipulation of mammals to understand and combat human disease. Her early work was with Prof. Rudolf Jaenisch, first at the Heinrich-Pette Institute, Hamburg and later at the Massachusetts Institute of Technology, and focussed primarily on retroviral vectors for gene therapy and insertional mutagenesis in mice. During this time she produced the first model of a human disease - a lethal disorder arising from collagen dysfunction and later an accurate model of human osteogenesis imperfecta type 1 (brittle bone disease) by a dominant negative mutant transgene. She subsequently joined Colorado State University, where her research extended to the production of transgenic livestock. From 1992–2003 she worked with the biotechnology company PPL Therapeutics in Edinburgh, becoming Assistant Director of Research in 2001. Her research at PPL centered on the production of pharmaceutical proteins in the milk of transgenic large animals and generation of xenotransplantation donors. Here she developed key technologies, most notably somatic cell nuclear transfer – Dolly the sheep, in collaboration with Ian Wilmut of the Roslin Institute. In 1997 she reported the first transgenic animal produced by nuclear transfer – a sheep carrying human clotting factor IX, for which she was awarded paper of the year by the journal “Science”. This was followed shortly by the first gene-targeted large animal. Current research activities include the generation of large animal models of serious human disease and the development of novel techniques for genome engineering in livestock species: particularly animal stem cells, artificial chromosomes and RNA inhibition. Postal address: Lehrstuhl für Biotechnologie der Nutztiere TU München, WZW Weihenstephan, Hochfeldweg, 1, 85354 Freising, Germany

List of authors

327

Thiele, Felix, Dr. med. M.Sc., has been Deputy Director of the Europäische Akademie since 1999. He studied medicine in Hamburg and Heidelberg and received a Dr. med. from the University of Heidelberg for experimental work in the field of high blood pressure research. He subsequently obtained an M.Sc. in Philosophy and History of Science at the London School of Economics. From 2002–2007 he was member of the Junge Akademie an der Berlin-Brandenburgischen Akademie der Wissenschaften und der Deutschen Akademie der Naturforscher Leopoldina. He teaches philosophy and medical ethics at the Fernuniversität Hagen and the University Duisburg-Essen. At the Europäische Akademie, he has previously coordinated the project “Human Genetics. Ethical Problems and Societal Consequences“. Postal address: Europäische Akademie zur Erforschung von Folgen wissenschaftlich-technischer Entwicklungen Bad Neuenahr-Ahrweiler GmbH, Wilhelmstraße 56, 53474 Bad Neuenahr-Ahrweiler, Germany

Index

adult stem cells 48 et seq. adversity 230 Agrobacterium mediated transformation 15 et seq. Agrobacterium rhizogenes 23 allergic responses 14, 64, 86 alteration in metabolism 263 animal cloning 43, 108, 112 animal closeness 148 animal death 51, 104, 108, 113, 115 animal husbandry 34, 49, 62 et seq., 101, 104, 108, 188, 191 animal pharming 27 et seq., 235 animal protection 241, 282 animal reproduction 108 animal rights 135 animal right to life 148 animal trials 242 animal welfare 101 et seq., 183, 188, 242, 243, 245, 294, 296, 300 animal welfare, assessment protocols 111 animal welfare, concept 104 animal welfare, housing 106, 109, 116 animal welfare, indicators 105 animal welfare, risks 102, 248 anthropocentrism 185 antibodies 53, 55, 58, 217 antithrombin 3, 54, 56, 189, 317 artificial chromosomes 29, 33 artificial insemination 38, 108 ATryn® 3, 56, 263, 317 attentive public 122, 128 attitudes 127 attitudes to pharming 121 et seq. attitudes to science 126 attitudinal structure 128

authorization 226, 249, 261 awareness 131 et seq. banana 22, 80 batch-to-batch consistency 25, 264 beltsville pigs 103 bioactive proteins 56 et seq. bioactivity 61, 79, 93, 110, 133 biocentrism 186 biologic, well-characterized 63 biological weapons 137, 163 biopharmaceuticals, cost of production 11, 26 biopharmaceuticals, definition 10 biopharmaceuticals, market of 1 biopharmaceuticals, produced in animals 27 et seq. biopharmaceuticals, produced in plants 11 et seq. biopharmaceuticals, purification 24, 61 biosimilars 64 blood 11, 29, 53, 58 caesarean delivery 49, 113, 114 callus 17, 18, 19 chicken 12, 36, 39, 41 et seq., 52, 58 et seq., 79, 94, 106, 107, 317 chicken eggs 11, 28, 58 et seq., 110 chimera 41, 42, 48 chinese hamster ovary (CHO) cells 2, 9 chromatographic purification 25 coding sequence 27 co-existence 73 et seq., 80, 237, 272, 300 co-existence, GM plants, legal aspects 77

330 compensation 280 competence for authorization 235 competence for denying an application 271 complementary DNA (cDNA) 27 confinement 88, 91, 274, 299 conflict, mastering of 181, 293 consistency, principle of 183 consumer sovereignty 266 contained use 269 containment 89 et seq., 218, 222, 298 cost-benefit analysis 251 cross compatibility 84 crude bulk material 260 deficit model 124 development medicinal products 257 development phase I 213 et seq. development phase II 241 et seq. development phase III 255 et seq. development phase IV 257 et seq. diabetes 1, 10 donor animals 31, 43, 46, 114 duties of care 275, 282 ecocentrism 186 edible vaccine 21, 22 EFSA 74, 113, 214, 224, 231, 235, 267, 271 EFSA, Animal Health and Welfare Panel 113, 187 embryonic germ cells 42 et seq., 51 embryonic stem cells 39 et seq., 45, 47 EMEA 267 equivalence to nature 232 Escherichia coli 9, 10, 26, 31 ethical justification 246 ethical theory 181 ethics 180 et seq. ethics, task of 184 ethics and public attitudes 184 ethics of pharming 179 et seq. evaluative angles 134 exon 9 expectations 129, 152

Index expectations about techno-scientific developments 145/146 exposure, quantification 79 facets angles 134 fitness, GM plants 85 flanking sequences 18, 19 foetus 113 foster mother 43, 49, 114 founder animal 4, 34, 47, 49, 59 et seq., 103, 112, 115, 196 frames 129, 132 gene 27 gene construct 13, 27, 231, 244, 263, 264, 296 gene construct, instability of 263 gene dispersal 80 et seq., 196, 223, 294 gene, eukaryotic 9 gene flow 80 et seq., 196, 299 gene gun 16, 19 gene spread 80 et seq., 224, 230, 300 gene targeting 28, 30, 40, 50 genetic code 9 genetic contamination 273 genetic engineering of animals 27 et seq. genetic engineering of plants 13 et seq. glycosylation 2, 10, 15, 52 et seq., 59, 64, 263 GMO-free 278 GMO-free zones 275 good farming practice 87 green biotechnology 5 guidance document, GM food and feed 75 guidelines 260 hairy roots 22 et seq., 26 holism 186 homologous recombination 39 et seq. human growth hormone 11 Humulin® 10 indispensability 247 informed public 128, 297

331

Index innovation 201, 207 insertion mutation 40, 75, 91, 111 institutions involved in pharming 214 insulator element 28 insulin 2, 10, 12, 92 integrity 191 et seq. intron 9, 28 labelling 266 et seq., 275, 297 lactation, induction of 33, 112, 115 large offspring syndrome 113 legal aspects, GM plants 74 et seq., 213 et seq. legal regulations involved in pharming 214 lentivirus 36 liability 278 male sterility, GM plant 89 market authorisation phase 258 microinjection 30 et seq., 43, 60, 114, 115 micromanipulator 32 milk 27, 45, 53 et seq., 61, 106. 108, 110, 112, 116, 141, 235, 240 milking 108, 112 mitigating measures 87 et seq. modelling, gene flow 87 monoclonal antibodies 2, 265 moral 180ff moral concerns 184 et seq., 293 moral concerns, aims and means of using animals and plants 192 et seq. moral concerns, integrity 191 moral concerns, naturalness 189 et seq. moral conflict, see conflict moral considerations 134 moral status of plants and animals 185 et seq. naturalness 189 et seq. need for a particular new medicinal product 262 Newcastle disease 11 nuclear transfer 43 et seq., 109, 112, 114, 115

occupational safety and health 255 particle bombardment mediated transformation 17 et seq. patents 201 et seq., 291 patents, extent of protection 205 patents, mandatory licenses 206 patents, moral justification 201 patents, regulation framework 202 patents and innovation 201, 207 pathocentrism 186, 187 et seq., 293 perception of pharming 136 perception of usefulness 135 perception, similarity between animals and human beings 147 pharming, definition of 1 physiocentrism 186 plant cell culture 22 plant pharming 11 et seq., 73, 134, 184, 214 plant production platform, European crops 82 plant virus 20 pollen dispersal 83 et seq., 231, 233 pollen flow 83 polymerase chain reaction (PCR) 33 position effect 28, 111 post-translational modification 2, 10, 14 et seq., 61 precautionary principle 197, 218, 222, 226, 229, 293 principle of justification 244 prion diseases 15, 61, 63 production phase 267 promoter 9, 14, 28, 60, 201, 216 protein excretion 26 protein purification 24, 26, 53, 61 public attitudes 121 et seq., 234, 293 public participation 236, 271 public understanding of science 124, 125 public views 121 et seq. rabbit 31, 38, 39, 47, 52, 56, 94, 106 recombinant protein 9 red biotechnology 5, 88

332 release of GMO 78, 81, 213 et seq., 268 reproductive system, gene flow 81 reproductive technologies 112, 114 reservations 129, 152 retrovirus 30, 34 et seq. risk 195 risk assessment 193 et seq., 195, 223 risk assessment, principles of 74 et seq. risk management 233 risk perception 126, 134, 149, 152 risk-benefit analysis 102, 193 et seq., 196, 227, 230, 261, 293, 294, 295 risks originating from the environment 258, 264 Saccharomyces cerevisiae 9 schemas 129, 131, 132 scientific literacy 123, 124 seeds 11, 22, 84, 204, 231, 270 seed dispersal 84, 231, 233 seminal fluid 57 separation (safety) distance 274 sequencing 13, 61 silencing 14, 20, 29, 37, 60 somatic cell nuclear transfer 45, 112 special regime 259 specific attitudes 128 sperm 38 spermatogonial stem cells 47, 48 subcellular targeting 14 summated scale 129 terminator technology 89 tissue-specific expression 28 tobacco 11, 12, 21, 139, 166

Index tobacco mosaic virus 20 trade-off between means and goals 134 transgene analysis 111 transgene construct 13, 27 et seq. transgene expression 14, 28, 45 transgene loci 36, 60 transgenic animals 29 et seq. transgenic plants 11 et seq. transient expression 20 translation 9 transplastomics, GM plant 89 trust 134, 152 trust in scientific community and regulators 126, 150 unintended exposure, GM plants 78 United States 225, 228, 230, 233, 254, 260, 265, 272, 277 urine 56 views of animals 126 views of nature 126 viral transduction 34 waste disposal 238 worldviews 126, 134, 146, 152 zero tolerance level 89, 277 3-R concept 244

Further volumes of the series Ethics of Science and Technology Assessment (Wissenschaftsethik und Technikfolgenbeurteilung): Vol. 1: Vol. 2: Vol. 3: Vol. 4: Vol. 5: Vol. 6: Vol. 7: Vol. 8: Vol. 9: Vol. 10: Vol. 11: Vol. 12: Vol. 13: Vol. 14: Vol. 15: Vol. 16: Vol. 17: Vol. 18: Vol. 19:

A. Grunwald (Hrsg.) Rationale Technikfolgenbeurteilung. Konzeption und methodische Grundlagen, 1998 A. Grunwald, S. Saupe (Hrsg.) Ethik in der Technikgestaltung. Praktische Relevanz und Legitimation, 1999 H. Harig, C. J. Langenbach (Hrsg.) Neue Materialien für innovative Produkte. Entwicklungstrends und gesellschaftliche Relevanz, 1999 J. Grin, A. Grunwald (eds) Vision Assessment. Shaping Technology for 21st Century Society, 1999 C. Streffer et al., Umweltstandards. Kombinierte Expositionen und ihre Auswirkungen auf den Menschen und seine natürliche Umwelt, 2000 K.-M. Nigge, Life Cycle Assessment of Natural Gas Vehicles. Development and Application of Site-Dependent Impact Indicators, 2000 C. R. Bartram et al., Humangenetische Diagnostik. Wissenschaftliche Grundlagen und gesellschaftliche Konsequenzen, 2000 J. P. Beckmann et al., Xenotransplantation von Zellen, Geweben oder Organen. Wissenschaftliche Grundlagen und ethisch-rechtliche Implikationen, 2000 G. Banse, C. J. Langenbach, P. Machleidt (eds) Towards the Information Society. The Case of Central and Eastern European Countries, 2000 P. Janich, M. Gutmann, K. Prieß (Hrsg.) Biodiversität. Wissenschaftliche Grundlagen und gesellschaftliche Relevanz, 2001 M. Decker (ed) Interdisciplinarity in Technology Assessment. Implementation and its Chances and Limits, 2001 C. J. Langenbach, O. Ulrich (Hrsg.) Elektronische Signaturen. Kulturelle Rahmenbedingungen einer technischen Entwicklung, 2002 F. Breyer, H. Kliemt, F. Thiele (eds) Rationing in Medicine. Ethical, Legal and Practical Aspects, 2002 T. Christaller et al. (Hrsg.) Robotik. Perspektiven für menschliches Handeln in der zukünftigen Gesellschaft, 2001 A. Grunwald, M. Gutmann, E. Neumann-Held (eds) On Human Nature. Anthropological, Biological, and Philosophical Foundations, 2002 M. Schröder et al. (Hrsg.) Klimavorhersage und Klimavorsorge, 2002 C. F. Gethmann, S. Lingner (Hrsg.) Integrative Modellierung zum Globalen Wandel, 2002 U. Steger et al., Nachhaltige Entwicklung und Innovation im Energiebereich, 2002 E. Ehlers, C. F. Gethmann (eds) Environmental Across Cultures, 2003

Vol. 20: Vol. 21: Vol. 22: Vol. 23: Vol. 24: Vol. 25: Vol. 26: Vol. 27: Vol. 28: Vol. 29: Vol. 31: Vol. 32: Vol. 33: Vol. 34: Vol. 35:

R. Chadwick et al., Functional Foods, 2003 D. Solter et al., Embryo Research in Pluralistic Europe, 2003 M. Decker, M. Ladikas (eds) Bridges between Science, Society and Policy. Technology Assessment – Methods and Impacts, 2004 C. Streffer et al., Low Dose Exposures in the Environment. DoseEffect Relations and Risk-Evaluation, 2004 F. Thiele, R. A. Ashcroft (eds) Bioethics in a Small World, 2004 H.-R. Duncker, K. Prieß (eds) On the Uniqueness of Humankind, 2005 B. v. Maydell, K. Borchardt, K.-D. Henke, R. Leitner, R. Muffels, M. Quante, P.-L. Rauhala, G. Verschraegen, M. Žukowski, Enabling Social Europe, 2006 G. Schmid, H. Brune, H. Ernst, A. Grunwald, W. Grünwald, H. Hofmann, H. Krug, P. Janich, M. Mayor, W. Rathgeber, U. Simon, V. Vogel, D. Wyrwa, Nanotechnology. Assessment and Perspectives, 2006 M. Kloepfer, B. Griefahn, A. M. Kaniowski, G. Klepper, S. Lingner, G. Steinebach, H. B. Weyer, P. Wysk, Leben mit Lärm? Risikobeurteilung und Regulation des Umgebungslärms im Verkehrsbereich, 2006 R. Merkel, G. Boer, J. Fegert, T. Galert, D. Hartmann, B. Nuttin, S. Rosahl, Intervening in the Brain. Changing Psyche and Society, 2007 G. Hanekamp (ed) Business Ethics of Innovation, 2007 U. Steger, U. Büdenbender, E. Feess, D. Nelles, Die Regulierung elektrischer Netze. Offene Fragen und Lösungsansätze, 2008 G. de Haan, G. Kamp, A. Lerch, L. Martignon, G. Müller-Christ, H. G. Nutzinger, Nachhaltigkeit und Gerechtigkeit. Grundlagen und schulpraktische Konsequenzen, 2008 M. Engelhard, K. Hagen, M. Boysen (eds) Genetic Engineering in Livestock. New Applications and Interdisciplinary Perspectives, 2009 E. Rehbinder, M. Engelhard, K. Hagen, R. B. Jørgensen, R. Pardo Avellaneda, A. Schnieke, F. Thiele, Pharming. Promises and risks of biopharmaceuticals derived from genetically modified plants and animals, 2009

Also the following studies were published by Springer: Environmental Standards. Combined Exposures and Their Effect on Human Beings and Their Environment, 2003, Translation Vol. 5 Sustainable Development and Innovation in the Energy Sector, 2005, Translation Vol. 18 F. Breyer, W. van den Daele, M. Engelhard, G. Gubernatis, H. Kliemt, C. Kopetzki, H. J. Schlitt, J. Taupitz, Organmangel. Ist der Tod auf der Warteliste unvermeidbar? 2006