Plant Gene Containment E-Book

• Herausgeber: John Wiley & Sons
• Kategorie: Wissenschaft und neue Technologien
• Sprache: Englisch

Seitenzahl: 549

Veröffentlichungsjahr: 2012

Beschreibung

Gene Containment provides a comprehensive look at genetically modified organisms and the strategies and implementation of key methods to gene containment. The book is divided into 5 parts: An Introduction that discusses the need for biotechnology and GMOs, Section 1 looks at the need for gene containment, Part II discusses varying strategies for gene containment, section III explores the assessment of gene containment approaches, and section IV covers the steps involved in implementing gene containment. Gene Containment will provide a thorough and up to date look at gene containment research and the needs for implementing new strategies in this arena.

Leseprobe

Contents

Cover

Title Page

Copyright

List of Contributors

Foreword

Section 1: Perspectives of Gene Flow

Chapter 1: Potential Environmental Impacts of Transgene Flow in Rice with a Particular View on Herbicide Resistance

Introduction

Transgene Escape and Its Potential Environmental Impacts

Herbicide-Resistant Rice

Gene Flow from Herbicide-Resistant Rice and Its Potential Impacts

Managing Potential Environmental Impacts from Transgene Flow

Conclusion

Acknowledgments

Chapter 2: Assessment and Detection of Gene Flow

Importance of Assessment and Detection of Gene Flow

Detection Methodologies

Applications in Assessment of Gene Flow

Perspectives

Acknowledgments

Chapter 3: Bt Resistance Monitoring in European Corn Borers and Western Corn Rootworms

Introduction

Resistance Monitoring

Bioassay Methods and Baseline Susceptibility

Recent Developments

Conclusions

Acknowledgments

Chapter 4: Commercialization of GMOs: the Company Perspective on Risks and Containment

Commercialization of Traits Derived from Plant Biotechnology (GMOs and Alien Chromosome Introgression)

Gene Containment During the Commercial Utilization of Crops Derived from Biotechnology

Gene Containment During Research, Development, and Seed Production

Stewardship and Gene Containment

Section 2: Strategies for Gene Containment

Chapter 5: Trait Control (Gene Switch Technology)

Components of Gene Switch Systems

Application of Gene Switch Technologies

Future Perspectives

Chapter 6: Male Sterility and Hybrid Plant Systems for Gene Confinement

Introduction

Cytoplasmic Male Sterility

Transgenic Nuclear Male Sterility

Transgenic Nuclear Male Sterility for Gene Confinement

Male Sterility for Recovery of Wide Crosses with Gene Confinement

Hybrid Plant Systems

Seed-Based Hybrid Plant Systems

Gene-Deletor System

Male and Female Sterile Lines

Total Sterility

Stable Knockout Mutations

Chapter 7: Gene-Deletor Technology and Its Potential Applications in Addressing Gene Flow and Food Safety Concerns over Transgenic Plants

Introduction

The Principle of Gene-Deletor Technology and Supporting Experimental Data

Characteristics and Requirements of Gene-Deletor Technology

Potential Applications of Gene-Deletor Technology

Conclusions and Perspectives

Acknowledgments

Chapter 8: Seed-Based Gene Containment Strategies

Seedless Crops

Seed-Based Containment

GeneSafe (Technology Protection System (TPS)) Technology

Expression Cascades for Gene Containment

Chapter 9: Transgenic Mitigation of Transgene Dispersal by Pollen and Seed

Introduction—Containers Leak

Mitigation—The Concept of Conferring a Selective Disadvantage Outside of Cultivation1

Evidence that TM Mitigates Transgene Establishment

Tailoring Mitigator Genes to Crops and Cropping Situation

Concluding Remarks

Chapter 10: Tapping RNA Silencing for Transgene Containment through the Engineering of Sterility in Plants

Introduction

Chapter 11: Control of Cleistogamy and Seed Dormancy for Biological Gene Containment in Oilseed Rape (Brassica napus L.)

Introduction

Pathways of Gene Flow in Oilseed Rape

Managing Gene Flow in Oilseed Rape

Combining Biological Measures Package for Gene Containment

Conclusion

Acknowledgments

Index

This edition first published 2013 © 2013 by John Wiley & Sons, Inc.

Wiley-Blackwell is an imprint of John Wiley & Sons, formed by the merger of Wiley's global Scientific, Technical and Medical business with Blackwell Publishing.

Editorial offices:2121 State Avenue, Ames, Iowa 50014-8300, USA The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK 9600 Garsington Road, Oxford, OX4 2DQ, UK

For details of our global editorial offices, for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at www.wiley.com/wiley-blackwell.

Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted by Blackwell Publishing, provided that the base fee is paid directly to the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923. For those organizations that have been granted a photocopy license by CCC, a separate system of payments has been arranged. The fee codes for users of the Transactional Reporting Service are ISBN-13: 978-0-8138-0349-4/2013.

Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought.

Library of Congress Cataloging-in-Publication Data

Plant gene containment / edited by Melvin J. Oliver. p. cm. Includes bibliographical references and index. ISBN 978-0-8138-0349-4 (hardcover : alk. paper) 1. Transgenic plants. 2. Plant genetic transformation--Prevention. I. Oliver, Melvin J. SB123.57.P567 2012 632′.8–dc23 2012010721

A catalogue record for this book is available from the British Library.

Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books.

Cover images: DNA strands: istock © Gregor Buir Rice plant: courtesy of Bao-Rong Lu & Wei Wang

Cover design by Matt Kuhns: [email protected]

List of Contributors

Hani Al-AhmadDepartment of Biology and Biotechnology, An-Najah National University, Nablus, PalestineRoger N. BeachyDonald Danforth Plant Science Center, St. Louis, MO, USAYongqin ChenDepartment of Biotechnology, Hubei University, Wuhan, ChinaAdam DeresienskiDepartment of Cell and Molecular Biology, University of Rhode Island, West Kingston, RI, USAHui DuanPlant Sciences, J.R. Simplot Company, Boise, ID, USAJonathan GresselPlant Sciences Department, Weizmann Institute of Science, Rehovot, IsraelSabine GruberInstitute of Crop Science, University of Hohenheim, Stuttgart, GermanyJoel HagueDepartment of Cell and Molecular Biology, University of Rhode Island, West Kingston, RI, USAKater HakeAgricultural and Environmental Research, Cotton Incorporated, Cary, NC, USAAlexandra HüskenMax Rubner-Institut, Federal Research Institute of Nutrition and Food (MRI), Department of Safety and Quality of Cereals, Detmold, GermanyAlbert P. KauschDepartment of Cell and Molecular Biology, University of Rhode Island, West Kingston, RI, USACharles KwitDepartment of Plant Sciences, University of Tennessee, Knoxville, TN, USAYi LiDepartment of Plant Science and Landscape Architecture, University of Connecticut, Storrs, CT, USAZongrang LiuUSDA-ARS, Appalachian Fruit Station, Kearneysville, WV, USABao-Rong LuMinistry of Education Key Laboratory for Biodiversity and Ecological Engineering, Institute of Biodiversity Science, Fudan University, Shanghai, ChinaRichard McAvoyDepartment of Plant Science and Landscape Architecture, University of Connecticut, Storrs, CT, USAReginald J. MillwoodDepartment of Plant Sciences, University of Tennessee, Knoxville, TN, USAHong S. MoonDepartment of Plant Sciences, University of Tennessee, Knoxville, TN, USAKimberly NelsonDepartment of Cell and Molecular Biology, University of Rhode Island, West Kingston, RI, USAMelvin J. OliverUSDA-ARS, Plant Genetics Research Unit, University of Missouri, Columbia, MO, USAM. Isabel OrdizDonald Danforth Plant Science Center, St. Louis, MO, USABlair D. SiegfriedDepartment of Entomology, University of Nebraska-Lincoln, Lincoln, NE, USAStacy D. SingerUSDA-ARS, Appalachian Fruit Station, Kearneysville, WV, USATerence SpencerDepartment of Entomology, University of Nebraska-Lincoln, Lincoln, NE, USAC. Neal Stewart Jr.Department of Plant Sciences, University of Tennessee, Knoxville, TN, USAMichael TilelliDepartment of Cell and Molecular Biology, University of Rhode Island, West Kingston, RI, USAWei WangMinistry of Education Key Laboratory for Biodiversity and Ecological Engineering, Institute of Biodiversity Science, Fudan University, Shanghai, ChinaWei WeiDepartment of Plant Sciences, University of Tennessee, Knoxville, TN, USAJaemo YangDonald Danforth Plant Science Center, St. Louis, MO, USA

Foreword

World Food Supply

J. Perry Gustafson1 and Peter H. Raven2

1USDA-ARS, Plant Genetics Research Unit, University of Missouri, Columbia, MO, USA 2President Emeritus, of the Missouri Botanical Garden, St Louis, MO, USA

Dr Norman E. Borlaug was involved in the discussions, concepts, and the first draft of the manuscript before his untimely death in 2009.

The United Nations projects that by 2050 world agriculture will need to increase food production by 70% (The United Nations Food and Agriculture Organization, 2010) in order to feed a projected world population of approximately 9 billion. Even with the increase in food production, the world will still have more than 1 billion undernourished people and more than 100 million living close to starvation. Therefore, in reality, world food production needs to be increased by more than 70% in order to have any impact on decreasing world hunger. However, the many claim that there is no longer a major food problem as evidenced by improvements in a decrease in the percentage of poor people. However, the numbers tell a different story since there will still be more than 1 billion impoverished people when the world population reaches approximately 9 billion, even though the percentage of impoverished people will have decreased. In addition, food imports will continue to dramatically increase; for example, wheat imports are projected to increase from 30 to 75 million metric tonnes (MT) by 2020 (Pingali and Rosegrant, 1998). It is clear that extraordinary improvements in world food production will be necessary.

Domestication of all the major food crops revolutionized human culture by allowing for a decrease of hunter–gatherer societies and the development of villages, then towns, cities, and finally countries. Once domesticated, plants grown for food have been continually improved and selected for increased production. The improvements in world food production have been steady over the past 40 years, ranging from ∼2.63 billon metric tonnes (BT) in 1963 to ∼7.99 BT in 2005 (including plants 6.90 BT and animals 1.09 BT) (The United Nations Food and Agriculture Organization FAOSTAT, 2010). Most importantly, this massive increase in food production was accomplished basically on the same amount of land. The dramatic increase in plant food production mainly came from improved crop cultivars, crop technology advances, and better management practices (World Bank World Development Report, 2008).

To feed the world's increasing population, we will be required, first, to not only increase food (plant) production but also do it in a manner that will improve world dietary standards. Second, the world has to start addressing the overwhelming task of equitably distributing food to all regions of the world in order to offset increasing world hunger in developing countries. We will never see a lasting solution to the world hunger problem without a strong balance between food production and distribution—in other words, social justice. Third, we will need to accomplish these objectives with a minimum, or even better, a positive impact on the world's environment. Fourth, most importantly, we will need to continue increasing world food production without expanding the land currently under cultivation. The increase in food production between the years 1963 and 2005 was accomplished without increasing the amount of arable land under production. For example, world grain yields more than doubled from 1.4 T/ha in 1961–1963 to 3.05 T/ha in 1997–1999, on approximately 56% less land (World Bank World Development Report, 2008). A 70% increase in world food production would equal approximately 23% of the current world production. Therefore, increases in world food production between now and 2040 are feasible by utilizing existing and newly developed technology to improve cultivar and management development without any further damage to our environment. However, we should keep firmly in mind that world food production increases could be subject to a number of additional undefined constraints.

This discussion suggests the potential for continued increases in world food production based on existing and newly developed technology; however, there are several limitations that could influence any increase in world food production. First, there might be limited access to technology for advancing yield to all regions of the world where food demands exist. Second, advanced technology and management inputs could easily spread into areas where environmental problems would be accelerated resulting in an adverse impact on the environment and biodiversity. Third, the public understanding of modern technology for increasing food production certainly needs to be improved. Major food production increases will certainly have to be based on the utilization of modern technology. To obtain the 7.99 BT of food produced in 2005 using 1963 cultivar/management technology (∼2.63 BT) (The United Nations Food and Agriculture Organization FAOSTAT, 2010) would have required an additional ∼1.6 billion hectare of new land brought under cultivation, which would have had a massive negative impact on existing biodiversity and our environment. Intensifying agricultural technology on existing lands, therefore, will continue to play a major role in preserving biodiversity and maintaining the sustainability of our fragile global environment.

It is clear that, to eliminate hunger, we must increase sustainable world food production. One vital need involves our continued preservation of sufficient genetic diversity in plants and their relatives to ensure that the capacity to create cultivars capable of resisting new biotic and abiotic stresses and at the same time adapting to new environmental conditions is maintained. Existing and newly developed biotechnological tools alongside traditional plant-breeding technology will play a major role in improving world food production, as did the green revolution that occurred from the 1960s through the 1980s. We will have to adapt new technologies to the needs of individual countries, industrialized and developing, so that they can effectively adapt and improve their food production without any adverse effects on the environment and biodiversity.

Modern biotechnology is capable of taking plant improvement to new heights with the potential of greatly improving food production. Recently developed successful technologies include, first, tissue culture, in which plants are broken down into cell suspensions and manipulated to regenerate plants, has bypassed some traditional approaches to seed production. Second, anther culture techniques have been successful in creating double haploid populations, greatly reducing the time required to produce cultivars. Third, modern approaches to mutation technology have been successful in creating genetic variation necessary for crop improvement. Fourth, the utilization of molecular marker-assisted selection and other molecular oriented technologies, where various types of DNA fragments, including numerous examples such as restriction fragment length polymorphism (RFLP), simple sequence repeat or microsatellite repeat (SSR), amplified fragment length polymorphism (AFLP), single nucleotide polymorphism (SNP), diversity array technologies (DArT), and so on, have been and will continue to be linked to value-added traits and have already been successfully used in cultivar improvement programs. Fifth, plant transformation technology, which involves transferring genes from one organism to another bypassing any sexual process, has and will continue to have a significant impact on the adaption of new cultivars to various biotic and abiotic stresses.

Most of the traditional and newly developed technologies have been and will continue to be adapted to a more land- and labor-intensive form of agriculture improvement. It is clear that organic and subsistence farming applications are neither capable of producing enough to feed nor improve dietary standards of our existing population, let alone the projected increase of 9 billion people by 2040. It has been estimated that organic farming applications are capable of only feeding a world population of approximately 4 billion people (Smil, 2001, 2004; Conner, 2008). The pure organic approach to feeding the world is a theory that simply is not possible and does not take into account the current scale of human suffering from malnutrition and starvation. Embracing social justice for everyone is the only way that humanity can survive and prosper.

Agriculture is certainly capable of feeding the projected world population on approximately the same amount of land currently under production (World Bank World Development Report, 2008). It will take all of our newly developed technologies and plant breeders’ skills to achieve the desired goal of satisfying world hunger. Significant progress has been made in advancing our understanding of the world we live in, which can be applied to technology for improving food production. No one knows the direction current research and breeding programs will take, but we can all assume that any application will have to be determined by economic and social factors. Only the coordinated application of all technologies will sustain the productivity of the lands and maintain our fragile environment.

All crops can be improved by traditional and biotechnological approaches to increase their yield potential. Building adaptable gene complexes from other species and even genera into the crops for the future is something that we have done in the past and must continue to do in the face of global climate change and the world's increasing population. This will require a much larger number of cultivars, with different genetic backgrounds, than in the past. It is very important that we consider improving the world production involving new varieties and management practices, including transgenic crops to the degree it will be possible to predict their impact on wild and weedy crop relatives, and the environment. Such concerns about gene contamination and environmental impacts should be carefully dealt with on an individual crop/environment basis. We must keep firmly in mind that gene complexes from other species and genera have been inserted into most of the world's major crops for more than the past 60 years. For example, Sears (1956) inserted a gene into wheat controlling disease resistance from Aegilops umbellulata, which saved wheat from a world rust epidemic. Future crop development will require that, first, we understand the manipulation of gene function and regulation in all crops. Second, we continue to explore and utilize all sources of gene complexes and technology. Third, cropping systems must be characterized to establish the genetic flexibility of various species in diverse ecological contexts, according to their breeding systems, mutation rates, genome recombination properties, and the genomic distribution and function of structural genes. Fourth, we characterize the interface between developing agricultural ecological dynamics and adaptive ecosystems in order to characterize genome evolution and the potential for gene contamination on an individual crop/location basis. In the past, when modern agriculture competed with the traditional subsistence forms of agriculture, local landrace cultivars were often discarded in favor of the new high-yielding cultivars. Massive efforts have been undertaken to preserve plant diversity, which has resulted in the retention of more old and new diversity in agriculture than existed 50 years ago. National and international seed banks are and will continue to be critically important to agriculture and the maintenance of the world's biodiversity.

The continued long-term health of world food production is one of the foundations to world security. The stable future of humanity, our environment, and our biodiversity are intimately tied to the improvement of crop production. Feeding the masses is clearly the most important challenge facing the world today and in the future.

References

The United Nations Food and Agriculture Organization (2010) Available: http://www.fao.org/fileadmin/templates/wsfs/Summit/Docs/Final_Declaration/WSFS09_Declaration.pdf (accessed: January 2012).

Pingali, P.L. and Rosegrant, M.W. (1998) Supplying wheat for Asia's increasingly westernized diets. American Journal of Agricultural Economics, 80, 954.

The United Nations Food and Agriculture Organization FAOSTAT (2010) Available: http://faostat.fao.org/site/368/Desktop Default.aspx?PageID=368#ancor (accessed: January 2012).

World Bank World Development Report (2008) Agriculture for Development. The World Bank, Washington, DC.

Conner, D.J. (2008) Organic agriculture cannot feed the world. Field Crops Research, 106, 187. Available: http://www.sfiar.ch/fileadmin/documents/recommend_dubock_field_crops_research.pdf (accessed: January 2012).

Smil V. (2001) Feeding the World: A Challenge for the Twenty-First Century. MIT Press, Cambridge.

Smil V. (2004) Enriching the Earth: Fritz Haber, Carl Bosch, and the Transformation of World Food Production. MIT Press, Cambridge.

Sears, E.R. (1956) The transfer of leaf-rust resistance from Aegilops umbellulata to wheat. Brookhaven Symposia in Biology, 9, 1–22.

Section 1

Perspectives of Gene Flow

1

Potential Environmental Impacts of Transgene Flow in Rice with a Particular View on Herbicide Resistance

Bao-Rong Lu and Wei Wang

Ministry of Education Key Laboratory for Biodiversity and Ecological Engineering, Institute of Biodiversity Science, Fudan University, Shanghai, China

Introduction

The continuous increase in human population and decrease in world arable lands and water resources have challenged the world food security. According to the statement by the Director-General Dr Jacques Diouf of the Food and Agriculture Organization (FAO), United Unions, at the “World Summit on Food Security” held in Rome, Italy, November 16–18, 2008, over 1 billion people are still fighting against hunger. He stressed the urgent need to produce food where the poor and hungry lived and to boost agricultural investment in these regions. To resolve the severe problems of the world, food security will provide a sustainable guarantee to the stabilization of human society. Scientists have proposed that the efficient application of high and new technologies, including transgenic biotechnology, in agriculture may provide an alternative solution to this problem (Huang et al., 2003). Transgenic biotechnology employs the modern genetic tools to engineer organisms; therefore, this technology provides highly effective and accurate tools for the genetic improvement of crop species. In addition, it can overcome the reproductive isolation when transferring genetic traits between distantly related species, which enable the modification of any crop species almost freely according to human design. Transgenic biotechnology with its new improvement, including cotransformation with multiple transgenes, is considered to be the most promising technology in the twenty-first century (Halpin, 2005).

Since the past 25 years or so, the research and development of transgenic biotechnology has been unprecedented. To date, gene transfer through biotechnology has been successfully achieved in more than 200 plant species, including food crops (e.g., rice, wheat, maize, sorghum, and barley), cash crops (e.g., cotton, soybean, and oilseed rape), vegetables (e.g., tomato, cucumber, leaf mustard, cabbage, and eggplant), and forestation species (Paulownia, Populus, Pinus, and Eucalyptus) (Yan, 2001). On the other hand, a large number of functional genes with practical values have been successfully explored in plant genetic engineering to develop transgenic plants. These include high protein content and unique nutritional compounds (Gura, 1999; Ye et al., 2000), disease and insect resistance (Datta et al., 2002; Huang et al., 2005; Bock, 2007), virus resistance (Shepherd et al., 2007; Vanderschuren et al., 2007), herbicide resistance (Lutz et al., 2001; Toyama et al., 2003), as well as salt and drought tolerances (Bahieldin et al., 2005; Tang et al., 2006).

The application of transgenic biotechnology and GM products has had tremendous impacts on world crop production in terms of its potential for poverty alleviation and solving the problems of malnutrition. By the end of 2009, the estimated global cultivation area of GM crops has exceeded 130 million hectares and generated about US $52 billion economic gains worldwide (James, 2009). Herbicide-resistant GM crops are among the most successful transgenic products, accounting for more than 65% of total global cultivation area of GM crops. GM crops with other traits also have played important roles in crop production. For example, the cultivation of insect-resistant GM cotton has led to reduced applications of pesticides that can harm human health and agricultural ecosystems (Huang et al., 2005; Brookes and Barfoot, 2009). A considerable decrease in regional outbreaks of cotton ball worms was associated with the extensive cultivation of GM Bt cottons (Wu et al., 2008).

The commercial production of GM crops is important for world food security by enhancing crop production. But, on the other hand, the extensive environmental release and commercial cultivation of GM crop varieties have aroused tremendous biosafety concerns and debates worldwide, including food and feed safety, environmental safety, and long-term availability of biodiversity (for a review, see Lu, 2008). Biosafety issues have already become a crucial factor in constraining the further development of transgenic biotechnology and the wider application of GM products in agriculture. Nowadays, it is not possible to circumvent biosafety issues when discussing the development and application of GM crops (Stewart et al., 2000; Pretty, 2001; Ellstrand, 2001, 2003). Therefore, it is necessary to face the challenge of the biosafety issues aroused by the cultivation of GM crops and try to close the “knowledge gap” by providing solid data from science-based research.

Transgene escape and its potential environmental impacts are among the most debated biosafety issues (Ellstrand et al., 1999; Ellstrand, 2001, 2003; Lu and Snow, 2005; Wang et al., 2006). What are transgene escape and gene flow? What are the potential environmental impacts caused by transgene flow? Is it possible to minimize transgene flow and to mitigate any negative impacts caused by transgene flow? In this chapter, we shall provide some information to address these questions using herbicide-resistant GM rice as a case study.

Transgene Escape and Its Potential Environmental Impacts

Transgene escape indicates a process in which a transgene(s) moves from a GM crop to its non-GM crop counterparts or to its wild or weedy relatives through gene flow. Transgene escape will occur and result in potential environmental and biodiversity impacts if transgene flow to non-GM crop varieties and weedy/wild populations is significant. Because transgene escape is caused by gene flow, it is therefore important to understand what gene flow is and how many types of gene flow there are.

Transgene Flow

Gene flow is a natural process that contributes significantly to the evolution of organisms (Lu, 2008). By a simple definition, gene flow indicates the movement of genetic materials (genes or alleles) from one organism to another. In population genetics, gene flow (also known as gene migration) refers to the transfer of alleles or genes from one population to another (Hartl and Clark, 1989). Through gene flow, different genes or alleles can be transferred among or within biological populations to achieve the exchange and dissemination of genetic information (Figure 1.1). Theoretically, there are two types of gene flow: (1) vertical gene flow and (2) horizontal gene flow, although the latter is preferably referred to as horizontal gene transfer (Gogarten and Townsend, 2005).

Figure 1.1 Schematic illustration showing two-directional gene-flow-mediated pollination among cultivated, weedy, and wild rice. Arrows with solid lines indicate gene flow among taxa, and arrows with broken lines indicate gene flow within a taxon.

Horizontal gene transfer occurs only among unrelated species, such as between plants and microorganisms, as well as between different microorganisms (Thomson, 2001). It is recognized as the major force for the genome evolution of some microorganism species. The frequency of horizontal gene transfer is very low (Nielsen et al., 1998). In terms of environmental biosafety, the discussion of horizontal gene transfer is based more on theory than practice, since it has never been shown to occur with transgene outside an experimentally enforced setting, even though this process is significant in the evolution of microorganisms. Therefore, this chapter will focus only on vertical gene flow that is meaningful in terms of transgene escape and its associated environmental impacts.

Gene flow can maintain plant populations at different spatial distances with a certain degree of genetic relatedness. Reproductive isolation and gene flow function as two major opposite forces in the evolutionary process, with the former promoting speciation or diversity, and the latter maintaining the same genetic identity of a species (Rieseberg et al., 2004). Human activity, such as domestication, serves as a strong isolation force that can produce, by selection and cultivation, a separate population derived from a wild plant species. Therefore, gene flow is a natural process that occurs incessantly and permanently between biologically compatible organisms and to which all genes are subject. In the case of transgene escape, gene flow serves as a medium that moves a transgene from a GM crop to its non-GM counterparts and weedy/wild relatives.

Since gene flow is defined as, for example, in plants, the movement of genes from one plant population to another, any medium such as pollen, seeds, and vegetative organs that can move genes around will lead to gene flow. Typically, there are three avenues for gene flow to be mediated: (1) either by pollen, (2) seed, or (3) vegetative propagules (Andow and Zwahlen, 2006; Lu, 2008).

Pollen-mediated gene flow occurs when pollen grains travel from a plant individual to another individual resulting in fertilization. This process can happen between individuals within the same population or among separate populations. In the latter case, wind, animals, water current, and other factors can serve as media.

Seed-mediated gene flow occurs through the natural dispersal of seeds by animals, wind, water, or other means from one population to another. Animals with long-range migration habits can transfer seeds over very long distances. Humans can also move seeds intentionally through seed-exchanging and trading within or between geographical regions, which can promote significant amounts of gene flow. The frequencies and patterns of human-influenced seed movement require sociological (seed exchange and distribution) and economic (regional and international trading) analyses and cannot be predicted using only knowledge related to plant biology.

In the case of vegetative-propagule-mediated gene flow, the movement of genes takes place through the natural dispersal of vegetative organs (e.g., tillers, roots, tubers, and rhizomes) of plant species by animals, wind, water, or other means. As for seed-mediated gene flow, the movement of vegetative organs, particularly by animals and humans, is difficult to estimate when based only on plant biology.

Pollen-mediated gene flow will be primarily determined by the intrinsic biological features, particularly the pollination biology of the plant species, such as breeding systems, outcrossing rates, amount of pollen (pollen load) produced by pollen donors, and pollen competition between donors and recipients (Rong et al., 2010). In addition, physical or environmental conditions, such as distances between pollen donors and recipients, the strength and direction of wind, temperature, light intensity, and air humidity, will also influence pollen-mediated gene flow to a great extent (Rong et al., 2010). It is therefore very important to generate such baseline biological and physical data through a science-based approach for the accurate prediction of pollen-mediated gene flow. In agricultural ecosystems, humans can play an important role in seed and vegetative-organ dispersal and migration, as would be the case of seeds or vegetative organs falling on the ground during harvesting and picking, transportation to the processing manufacturers, and trading at the local, regional, and international level. The intensity and avenues of gene flow in different crop species can vary significantly, depending on annual or perennial characteristics, the capacity for seed dormancy, the longevity of seeds or vegetative propagules during storage (under natural or artificial conditions), differences in breeding (mating) systems, the importance of such crops in national and international markets, and those parts of the crop that are consumed by humans. Given the complexity of gene movement through seeds or vegetative organs, seed-mediated gene flow and vegetative-propagule-mediated gene flow will not be discussed further in this chapter, but it is necessary to point out that these are very important avenues for gene flow in terms of evolutionary processes or GM-related biosafety issues.

Potential Environmental Impacts Associated with Transgene Flow

The most relevant questions relating to transgene flow and its potential biodiversity and environmental impacts should be scientifically addressed and analyzed. This will not only facilitate our objective understanding of the potential biosafety problems caused by transgene flow at various situations but also for the effective assessment and management of transgene flow and its impacts. Such knowledge will guarantee the further development of transgenic biotechnology and promote the safe and sustainable utilization of its products.

The environmental impacts created by transgene escape into different recipients can vary significantly in terms of categories and magnitudes. Transgene escape from GM crops to their non-GM counterparts will have completely different consequences compared with the escape to weedy and wild relative species. Even in the latter case, different types of transgenic traits will have different effects to wild populations under different environmental conditions and human influences. Therefore, the case-by-case principle should be applied rigidly to assess the environmental impacts from transgene flow, which should be dependent upon the types of recipients (e.g., crops or wild species) that may have acquired the transgenes.

Crop-to-Crop Transgene Flow

The major consequence caused by transgene flow from a GM crop to its non-GM crop counterparts is the “adventitious mixing” of GM and non-GM crop varieties (or so-called “contamination”). If the transgene becomes present in seeds or the derived products of a non-GM crop and is consumed by human or used as animal feed, such a “contamination” may arouse food and feed biosafety concerns, and cause some trading problems between regions or countries. Sometimes, such a “contamination” may even result in legal disputes among different parties. There are already a few examples where the products of nonfood GM crops have been found in mixture with food and feed crops. One of the well-known examples is the Starlink™ GM corn (transformation event CBH-351; Organisation for Economic Cooperation and Development (OECD) unique identifier ACS-ZMØØ4-3) that was only approved for use in animal feed. In 2000, the Bt (Cry9C) toxin from Starlink corn was detected in taco shells, sparking a whole-scale product recall (Heinemann, 2007). In addition, a significant amount of gene flow to non-GM crops has the potential to increase opportunities for subsequent gene movement to weedy or wild rice populations. In these cases, the level of “mixture” or “contamination” from GM crop by gene flow is crucial.

Transgene flow from a GM crop to its non-GM crop counterparts can also lead to the change of genetic diversity in traditional crops. The extensive adoption of GM crops may lead to rapid losses of traditional crop varieties because of the continuous replacement of the traditional varieties by more commercially advantageous GM varieties. For example, after only a decade of adopting GM cotton, the current cultivation area of insect-resistant GM cotton (Bt) comprises more than 70% of the total cotton cultivation area in China, and more than 65% of the total cotton cultivation area in India (Wu, 2007; James, 2009). In addition, the spread of transgenes from a GM crop variety to non-GM traditional varieties through gene flow may change the integrity of the traditional varieties if the transgenes have a selective advantage. During the process of cultivation and seed production, hybrids containing beneficial transgenes may gradually accumulate unintentionally during selection to ultimately replace the important original genotypes of the traditional varieties (Lu, 2008).

Crop-to-Weed/Wild Transgene Flow

Transgene flow from a GM crop to the weedy and wild relatives of the crop may create invasive weeds if the GM crop that is modified to tolerate herbicides or to resist diseases and pests transfers such traits to wild or weedy relatives via gene flow. Crops can also be modified with traits that allow them to grow faster (e.g., by expressing a specific growth hormone), reproduce more (e.g., by enhancing seed production), and live in new types of habitats (e.g., by enhancing drought and cold tolerance). The potential environmental impacts caused by crop-to-weedy or crop-to-wild gene flow need to be determined in the long term. There are still many biological mechanisms underlying the process of gene flow and fitness change to be understood. The following are only some of the hypothesized or predicted consequences of crop-to-wild gene flow that are commonly discussed and debated worldwide, although most have never been found or proven. Science-based studies should be conducted to test whether in reality such consequences will happen under a case-by-case situation, and to measure the magnitude of such consequences should they occur.

Many crop species coexist with their conspecific weedy types in the same agroecosystems, for example, weedy rice (Oryza sativa f. spontanea, Cao et al., 2006) in rice field and weedy Brassica types in oil rape field (Hall et al., 2000). Gene flow between crops and their weedy populations is relatively common because the weedy populations are conspecific with the crops, and usually derived from volunteers of the same crop species, or from offspring of hybrids between crops and their wild relatives (Lu and Snow, 2005). Transgenes that convey resistance to biotic and abiotic stresses incorporated into weedy populations through gene flow may create weeds with new traits, because such crop-weedy transgene flow may accentuate the characteristics of weediness, leading to great persistence and invasiveness of already existing weeds. On the other hand, a GM crop may acquire genes for weediness from weeds leading to persistence and invasiveness of a crop species or volunteers (Lu and Yang, 2009). Therefore, the concerns of gene flow with respect to weediness are mostly related to the following two aspects: (1) a wild or weedy species that invades and persists in crop fields has the ability to become a more effective and aggressive weed; and (2) a GM crop volunteer or hybrid between the GM crop and wild relatives has the ability to become a more effective and aggressive weed after incorporating transgenes that convey traits against biotic and abiotic stresses. These concerns relate to the hypothesis that a transgene from GM crops will bring a fitness advantage to the populations of crop volunteer, weeds, and wild species (Lu and Snow, 2005). For example, in the United Kingdom, charlock (Sinapis arvensis) seeds are found to persist in soil seed banks for up to 35 years, and this observation is important because the detection of hybrids between charlock and herbicide-resistant rape indicated that transgene flow could potentially make this important weed tolerant to herbicide. Transgenic traits that may influence invasiveness of weeds include enhanced fertility, vegetative vigor, tolerance to a wide range of environmental stresses, and the quality and dispersal range of viable materials.

Wild relatives of crop species that are generally viewed as valuable resources of genetic diversity for crop breeding, and usually many populations of wild relatives occur in the vicinity of crop fields (Vaughan, 1994; Ellstrand, 2003). Genetic diversity of wild relative species should be protected from population extinction and genetic “swamping,” which frequently results from a heavy influx of crop genes through gene flow. To the extreme extent, the mere presence of transgenes in the wild germplasm of crop relatives already represents a form of “contamination” or “genetic pollution.” There are two scenarios for the undesired impacts of transgene flow on genetic diversity of wild relative species. First, it is theoretically possible that the strong selection for fitness-enhancing transgenes could generate selective sweeps, in which portions of the crop genome that are linked to these transgenes displace corresponding portions of wild genomes (Ellstrand, 2003; Gepts and Papa, 2003). This process is expected to be more common in self-pollinating species than in outbreeders that have a greater potential for the mixing and dilution of crop alleles during sexual reproduction. Selective sweeps could also be favored by clonal reproduction, which might allow more vigorous transgenic crop–wild hybrids to outcompete the non-GM plants at the local level. Second, in some cases, a large influx of fitness-reducing transgenes could contribute to population declines, or even local extinction of small, isolated populations of wild plants that occur near the crop (Haygood et al., 2003). In populations with 100 individuals or more, frequencies of fitness-reducing transgenes would diminish due to the purifying force of natural selection. Therefore, current information suggests that gene flow from self-pollinating GM crops may not threaten the genetic diversity of wild and weedy relatives to a greater extent than gene flow from conventional varieties (Ellstrand, 2003; Gepts and Papa, 2003). However, the massive transgene flow from a GM crop to wild relatives through recurrent pollination may increase the threats to the survival of wild populations, particularly for outbreeders. No matter which of the scenarios become true, transgene flow will pose potential impacts on the environment by altering population dynamics.

Herbicide-Resistant Rice

Rice (O. sativa L.) is one of the world's most important cereal crops, consumed by nearly one-half of the world population (see Lu and Snow, 2005). In many developing countries, rice is the basis of food security and is intimately associated with the local ways of life. During the “green revolution” of the 1960s and 1970s, the breeding of new semidwarf and photoperiod-insensitive rice varieties provided large yield increases, corresponding to the decrease in food shortages in many regions of the world. The dramatic increase in rice productivity is obviously attributed to the genetic improvement of this crop, the intensification of fertilizer and pesticide use, as well as better irrigation (Conway, 1997). In addition, weed control with intensive labor input that characterized traditional agriculture in many developing countries at that time also played an important role in enhancing rice production. However, with the continued increase in human population that will reach 9.2 billion by 2050 (James, 2009), the world demand for rice will continue to increase and go beyond the current production capacity, which certainly requires the application of new technologies further to enhance rice productivity.

Weeds are constant constraints to rice production and cause considerable yield losses in all rice production systems and in all seasons worldwide (Zoschke, 1990). The severity of weed problems and composition of weedy species varies among rice ecosystems and even among fields, given the unique combination of physical and environmental conditions at each site (Savary et al., 1996). The traditional methods to effectively control weeds in rice fields depended on the transplanting of rice seedlings and manual weeding. However, labors for rice transplanting and hand weeding are increasingly scarce, which forces farmers to switch to direct seeding or other no-till planting modes in many rice-growing countries, thereby losing the early season advantages that the crop has under a flooded condition to suppress initial weed growth, especially for grass species (Olofsdotter et al., 2000).

With the introduction of 2,4-Dichlorophenoxyacetic acid (2,4-D), a common systemic herbicide used in the control of broadleaf weeds in 1950s, agrochemical companies have successfully developed and brought to market a wide array of selective herbicides. The severity of limited agricultural resources, especially water availability, the measures to conserve water in rice production, such as intermittent flooding and shallow water depths are generally less efficient with respect to weed control. Water and labor shortages have considerably constrained weed control in rice field, which has alternatively led to the increase in the reliance on herbicides. The application of herbicides is the most reliable and inexpensive method for weed control compared with the manual approaches, therefore, it is rapidly being adopted in agriculture production, including rice production, in developed and developing countries. To a certain extent, the application of herbicides has simplified weed management in many cropping systems (Heap, 1997).

Table 1.1 Naturally discovered and artificially synthesized herbicide-resistant genes that are used in transgenic plants

The rapid progress in life sciences and biotechnology has promoted the development and application of herbicide-resistant genes that are usually associated with a particular group of herbicides. A large number of naturally discovered and artificially synthesized herbicide-resistant genes are used for the development of herbicide-resistant GM plants (Table 1.1). To date, most of the herbicides being used are found corresponding to specific herbicide-resistance genes. In addition, the ClearfieldTM rice was intentionally created from a mutation that confers tolerance to imidazolinone herbicides. Since Clearfield rice is not a GM variety, it has been commercially cultivated in a number of countries (Zhang et al., 2006). This technology has allowed for preemergence and postemergence applications of imazethapyr (Newpath®) and postemergence applications of imazamox (Beyond®) to control weeds in rice fields, including weedy rice (commonly referred to as red). Before the advent of this technology, there were no effective options to control weedy rice in conventional rice fields. Since the commercial introduction of the first herbicide-resistant Clearfield rice in 2002, this technology has been widely adopted. For example, in 2006, Clearfield rice was grown on 33% of the rice acreage in Mississippi. The increase in Clearfield rice acres has resulted from its advantages of exceptional weedy rice control and an overall simpler weed control, when still remains reasonable rice yield (Sha et al., 2007). Although the current Clearfield rice varieties yield about 5–10% less than the predominant Mississippi rice variety (Cocodrie), farmers still chose to grow these herbicide-resistant varieties due to the cost-effectiveness of labors (Sha et al., 2007). Once the released new Clearfield rice varieties meet or exceed rice yields of high-yielding control, acres for growing the Clearfield rice are expected to increase (Sha et al., 2007).

The trait of herbicide resistance has composed the major portion of the current GM crops, reflecting that the labor and resource saving farming is appreciated globally and will potentially be the trend for development in modern agriculture. According to the statistics of James (2009), herbicide-resistant GM crops are accounted for about 65% of the GM crops that have been cultivated worldwide, in terms of their area. As one of the world's most important food crops, rice has also been genetically modified (GM) to resist to a range of herbicides. Although no single herbicide-resistant GM rice variety has been extensively cultivated for commercial production so far, the efforts to work on herbicide-resistant GM rice have never been stopped.

In the United States, two major types of GM herbicide-resistant rice are available for potential commercialization. The first type is the LibertyLink® rice that is a GM variety, containing a bar gene and resistant to phosphinic acids (such as glufosinate) developed by the Bayer Crop Science (Vermij, 2005). The LibertyLink rice has received a regulatory approval for commercial production (Annou et al., 2001). The second type is the Roundup Ready® GM rice that contains an aroA gene and is resistant to glyphosate (N-(phosphonomethyl) glycine), produced by the Monsanto Company (Lu and Snow, 2005). However, neither of these GM herbicide-resistant rice varieties has so far been used commercially in farmers’ fields. In China, several types of herbicide-resistant GM rice are under research and development by different institutions, although it is still a long way to go for the commercial production of the GM rice. Of these, most are designed for more effective weed control, and others are developed for maintaining seed purity in the seed production for high-yielding hybrid rice (Zhang et al., 1998). In addition, many other countries, such as India, Japan, South Korea, and Brazil are also involved in the research and development of various herbicide-resistant GM rice (Olofsdotter et al., 2000; Zhang et al., 2003).

Gene Flow from Herbicide-Resistant Rice and Its Potential Impacts

The extent of pollen-mediated gene flow in rice is essentially determined by its pollination biology, climate conditions, and genetic relationships of cultivated rice with wild species in the genus Oryza if the crop-to-wild gene flow is considered (Lu and Snow, 2005; Rong et al., 2010). Only the wild Oryza species containing the AA genome that is also shared by cultivated rice have the ability to outcross with the crop (Lu and Snow, 2005). All species in Oryza are wind-pollinated and have relatively short-lived pollen, although the outcrossing rates of different species can vary significantly (Oka, 1988). As a wind pollinating taxon, cultivated rice is characterized by an extremely high ratio of self-pollination. However, under the condition of overlap in flowering, cultivated rice will show a very low frequency of cross-pollination (usually, about 1%) between adjacent plants grown in the same or different fields (Rong et al., 2004; Lu and Snow, 2005). Spontaneous gene flow between different rice varieties is highly dependent on the proximity in flowering time and spatial distance, given that the outcrossing ability of the varieties is determined (Rong et al., 2010).

Figure 1.2 Weedy rice plants are commonly found in rice fields and cause yield losses if the number is significantly large.

Field experiments in Italy showed that pollen-mediated gene flow from a transgenic herbicide-resistant rice variety to the adjacent plants of a nontransgenic counterpart was about 0.05–0.53% (Messeguer et al., 2001). A large scale of field studies using a selectable transgenic marker (hy, for hygromycin resistance) to detect outcrossing also indicated extremely low frequencies (>1%) of pollen-mediated transgene flow from insect-resistant GM rice lines to their non-GM counterparts at close spacing (Rong et al., 2005), and that beyond 6 m from the transgenic lines (pollen donors) the pollen-mediated transgene flow reduced dramatically to an extremely level (>0.001%; Rong et al., 2007). Another benchmark study was conducted in the United States by rice breeders to establish isolation standards for seed purity in their breeding and seed production program, which is still in use today (see Gealy et al., 2002). The study used the glutinous endosperm marker, which allowed the evaluation of pollen-mediated gene flow frequencies in the harvested seeds. Four rice varieties were paired with the germplasm of similar maturity containing the glutinous marker. Four US Department of Agriculture (USDA) rice stations located in Beaumont (TX), Stuttgart (AR), Crowley (LA), and Biggs (CA) participated in the study, and each planted the same variety pairs for a period of 4–6 years. The outcrossing rates detected ranged from 0% to 3.4% (0.45% on average), at distances up to 1 m. All these results are consistent with the distances that are recommended for isolating and maintaining the purity of cultivated rice grown in seed nurseries. In the United States, for instance, rice plants that are grown for certified seed to be sold to farmers must be isolated from other rice varieties by only 6 m or less (Gealy et al., 2003). As discussed in the previous sections, the impacts of crop-to-crop transgene flow are mostly associated with the adventitious presence (or “contamination”) of transgenes in non-GM rice varieties. Given the low frequency of transgene flow in cultivated rice and possibility of minimize the frequency of transgene flow to a permitted (threshold) level by the management of spatial isolation, the impacts of pollen-mediated transgene flow in cultivated rice should be negligible.

Weedy rice infests fields of cultivated rice worldwide (Figure 1.2). It is widely recognized that transgenes introduced into cultivated rice will make their way into weedy rice populations through gene flow (Figure 1.3), even though the rate of cross-pollination from the crop could be very low at each generation (Gealy et al., 2003; Lu and Snow, 2005). A large number of studies have been conducted to estimate the level of pollen-mediated gene flow from cultivated rice to weedy rice (Dunand, 1988; Wheeler and TeBeest, 2002; Zhang et al., 2003; Gealy et al., 2003; Chen et al., 2004; Messeguer et al., 2004), using morphological traits (e.g., purple leaf), molecular fingerprints (e.g., microsatellites), or transgenes (e.g., herbicide resistant) as markers to detect outcrossing. Results from these studies indicated that both cultivated and weedy rice are self-pollinating taxa, and low frequencies of gene flow from cultivated to weedy rice have been detected in different studies. For example, transgene flow from herbicide-resistant GM rice to weedy rice populations ranged from 0% to 0.06% under experimental conditions in China and southern Europe (Chen et al., 2004; Messeguer et al., 2004). Likewise, studies in the United States indicated that the rates of crop–weed hybridization varied approximately between 0.01% and 1% (Dunand, 1988; Gealy et al., 2003). In recent studies, particularly involving the ClearField rice (CL 121 and CL 2551), the estimated crop-to-weed gene flow frequencies are less than 0.05% and 1%, respectively (Estorninos et al., 2002).

Figure 1.3 A close-up shot of weedy rice spikelets showing the exserted anthers and stigmas, which will result in outcrossing by alien pollen grains.

It is important to point out that gene flow can occur in both directions, that is, from cultivars to weedy rice, and from weedy rice to the crop (the taller stature of weedy rice can facilitate gene flow in the latter direction). The extent to which crop genes are passed to weedy populations can be very low when the flowering times of the crop and weedy rice are partially or completely asynchronous, as occurs in some situations (Lu and Snow, 2005). Both crop-to-weed and weed-to-crop gene flow will probably result in environmental impacts because genes encoding traits tolerant to abiotic stresses from weedy rice will strengthen the weed–crop hybrids that may turn into weedy rice. In addition, it is also possible that the crop itself can establish feral populations (Baki et al., 2000) that may possess transgenes if the crop is transgenic. However, the question of whether such “de-domestication” occurs within modern rice cultivars has not been adequately studied, and the significance of this type of gene flow is still unknown.

All the previously mentioned studies indicated that transgenes conveying herbicide resistance could easily be transferred to weedy rice populations through gene flow although the frequency could be very low for one generation. The frequency of transgenes can be accumulated by recurrent gene flow by generations because seeds of weedy rice that acquired the transgenes can survive in soil seed bank and regenerate in the next rice planting seasons. Transgene flow will make the weed control in rice fields that depend principally on herbicides more complicated because the transfer of traits with different herbicide resistance to weedy rice populations will make the control of weedy rice using herbicides almost impossible. Due to the strong competitiveness, seed shattering, and long-lived seed bank, weedy rice containing herbicide-resistant transgenes will become very difficult to eradicate soon as it infest rice fields. Modeling studies by Madsen et al. (2002) estimated that herbicide resistance may become common in weedy rice populations within only 3–8 years of continuous rice cropping. Therefore, it seems extremely likely that new genes for herbicide resistance will spread to weedy rice, especially in regions where weed management is already difficult. The heavily infested rice fields with weedy rice may become unusable, which will make the strategy of applying herbicide-resistant GM rice to control weedy rice short lived, if no effective technology to mitigate the impact of herbicide-resistant transgene is applied.

Another important but little-studied component of gene flow is the persistence and spread of crop genes following crop–weed hybridization. Just as hybrid vigor is seen when inbred, cultivated lines are crossed to produce “hybrid” rice, so weedy rice may benefit from hybridizing with the crop, if this results in greater heterosis. In Louisiana, for example, Langevin et al. (1990) reported greater vigor in crop–weed hybrids than in their weedy parents, and frequencies of crop alleles in weedy rice were observed as high as 52% after only 2 years of contact with the crop. However, in Arkansas, first-generation hybrids between cultivated and weedy rice flowered so late that they had much lower fitness than their weedy parents (Zhang et al., 2003). In a study of field performance of F1 hybrids between insect-resistant GM rice and weedy rice in China, Cao et al. (2009) found that the 1,000-grain weight and other morphological traits of the F1 hybrids were superior to the weedy rice parents, and that seed germination and seedling survival were significantly positively correlated to 1,000-grain weight. The hybrid vigor per se in the crop–weed hybrids may facilitate the persistence and spread of transgenes into weedy rice populations through introgression. Therefore, the evolutionary importance of hybrid vigor in weedy rice populations appears to be variable and should be studied more broadly in terms of its environmental impacts from crop–weed gene flow. In addition to the impacts from the F1 hybrids, studies showed that over the course of several generations, crop genes that are strongly deleterious to weedy rice, as well as other genes that are linked to deleterious crop genes, are likely to be purged from weedy populations by natural selection and by selection pressures from farmers (e.g., tall plants and long awns). On the contrary, linked genes that are associated with greater survival and reproduction are expected to increase in frequency following the incidents of hybridization. Thus, crop genes are expected to introgress and persist in weedy rice populations at rates that reflect both the frequency of hybrids and the selective effects of these genes (Ellstrand, 2003).

Transgene escape from cultivated rice to wild Oryza species and the potential ecological consequences have also aroused a worldwide concern (Lu and Snow, 2005). Earlier studies already reported the occurrence of natural hybridization among cultivated rice, perennial common wild rice (Oryza rufipogon), annual common wild rice (Oryza nivara), and their intermediate forms (Oka, 1988; Majumder et al., 1997). Recent studies based on designed field experiments indicated gene flow from cultivated rice to different wild O. rufipogon populations with significant variation in frequencies (up to 18%), which is much greater than those of crop-to-weed gene flow (Song et al., 2003; Chen et al., 2004; Wang et al., 2006). The much greater frequency of gene flow from cultivated rice to wild O. rufipogon than to weedy rice is most likely due to the high outcrossing rates that are genetically built in the wild rice species. This has been confirmed in the modeling of pollen-mediated gene flow in rice where the outcrossing ability of a pollen recipient, together with a few other biological parameters, acts as important determinant to affect gene flow frequencies (Rong et al., 2010).

Apart from biological parameters, climatic conditions (e.g., air humidity and wind speed) and physical distance between pollen donors and recipients will also affect the frequency of gene flow (Rong et al., 2010). Although the maximum distance at which pollen from rice donors can be detected was about 30 m in a pollen flow experiment (Song et al., 2004a), the most effective physical distance for gene flow from a rice cultivar to wild O. rufipogon under field experimental condition was within 30 m where approximately 95% of the crop–wild hybrid seeds was detected (Song et al., 2003). The modeling of pollen-mediated gene flow in rice also indicated that under the worst-case scenario, the attempt to reduce the frequency of pollen-mediated gene flow from cultivated rice to common wild rice to a low level (<1%) required a long spatial isolation distance (about 100 m) because the outcrossing rate of common wild rice can be as high as 40% (Rong et al., 2010). Variation in shared flowering phenology also regulates opportunities for gene flow between cultivated and wild rice. In southeastern China and Vietnam, for example, O. rufipogon usually flowers during the winter months and peaks in November and December, whereas O. sativa varieties are cultivated for two times per year in tropical and subtropical climates. Therefore, some portions of the annual cycle of rice production are likely to overlap with the flowering period of O. rufipogon in these countries. In addition, wild rice populations often have a protracted flowering period that makes the crop-to-wild gene flow possible (Cohen et al., 2008).

The persistence and spread of crop genes (including transgenes) introduced into wild rice populations can be significantly influenced by the fitness (survival and fecundity) of the crop–wild hybrids, brought by specific genes from the crop. However, little is known about the fitness change caused by a transgene in the wild rice populations under different environmental conditions. Results from our previous fitness study of F1 hybrids between cultivated rice and O. rufipogon without the involvement of any transgene showed lower pollen fertility and seed set than the wild parent although other traits such as number of panicles per plant and seed germination rates showed significantly higher values in F1 hybrids than in the wild parent (Song et al., 2004b). This partial reproductive barrier is expected to impede the rate at which crop genes introgress into wild populations, but it will not prevent introgression altogether. It was concluded from this study that rice genes might persist in wild rice populations following the crop-to-wild gene flow through vegetative and sexual reproduction. Therefore, longer-term studies of the persistence and spread of crop genes, particularly transgenes with natural selective advantages, in wild rice populations are needed to evaluate the potential for impacts from transgene flow.

The long-term environmental impacts of herbicide resistance traits following gene flow to wild rice populations are still under discussion. Theoretically, it is generally expected that without the spray of herbicides in natural habitats, where wild rice populations occur, the herbicide-resistant transgenes will not bring any natural selective advantages to wild rice. Therefore, the herbicide-resistant transgenes will only act as neutral markers in this case, and consequently the frequency of the transgenes after gene flow will not increase owing to the absence of selective advantages. However, data from experiments have shown that the function of the aroA gene is to overexpress the target enzyme 5-enolpyruvylshikimate-3-phosphate synthase (EPSPS), consequently enhancing the resistance to glyphosate for the plants that contain this gene (Comai et al., 1983). Studies also showed that the overexpression of EPSPS can increase the metabolic function (e.g., photosynthesis) of plants (Comai et al., 1985; Guyer et al., 1995). If this is true, wild rice plants acquired aroA gene through transgene flow may become more vigorous, even without the spray of herbicides, leading to environmental impacts. Our unpublished data based on preliminary field experiments indicated that weedy rice and wild rice plants containing the aroA gene exhibited considerably better fitness in a number of vegetative and reproductive traits than the plants without the aroA gene. Therefore, the environmental impacts of herbicide-resistant transgenes transferred from GM rice to wild rice species still need a thorough assessment.

Managing Potential Environmental Impacts from Transgene Flow

Given the fact that pollen-mediated gene flow from GM rice to its non-GM counterparts and populations of wild or weedy relatives at various frequencies will occur under natural conditions, effective management of such transgene flow and the resulted environmental impacts are essential for the safe and sustainable application of herbicide-resistant GM rice. There are two major groups of strategies being proposed to confine pollen-mediated gene flow: physical and biological (Committee on the Biological Confinement of Genetically Engineered Organisms, 2004), although the former strategy is extremely difficult in many cases, particularly for transgene flow from GM rice to weedy rice and to wild rice species that coexist or occur in close proximity. In the case that the confinement of transgene flow to weedy and wild rice populations is difficult, the use of transgene mitigation (TM) can also reduce the potential environmental impacts of transgene escape (Gressel, 1999).

Transgene Flow Confinement