Agronomy
D VA N C E S I N
VOLUME 95
Advisory Board Paul M. Bertsch University of Georgia
Ronald L. Phillips University of Minnesota
Kate M. Scow University of California, Davis
Larry P. Wilding Texas A&M University
Emeritus Advisory Board Members John S. Boyer University of Delaware
Kenneth J. Frey Iowa State University
Eugene J. Kamprath North Carolina State University
Martin Alexander Cornell University
Prepared in cooperation with the American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America Book and Multimedia Publishing Committee David D. Baltensperger, Chair Lisa K. Al-Amoodi Kenneth A. Barbarick
Hari B. Krishnan Sally D. Logsdon Michel D. Ransom
Craig A. Roberts April L. Ulery
Agronomy D VA N C E S I N
VOLUME 95 Edited by
Donald L. Sparks Department of Plant and Soil Sciences University of Delaware Newark, Delaware
AMSTERDAM • BOSTON • HEIDELBERG • LONDON NEW YORK • OXFORD • PARIS • SAN DIEGO SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO Academic Press is an imprint of Elsevier
Academic Press is an imprint of Elsevier 84 Theobald’s Road, London WCIX 8RR, UK Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands Linacre House, Jordan Hill, Oxford 0X2 8DP, UK 30 Corporate Drive, Suite 400, Burlington, MA 01803, USA 525 B Street, Suite 1900, San Diego, California 92101-4495, USA
First edition 2007 Copyright ß 2007, Elsevier Inc. All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the Publisher. Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (þ44) (0) 1865 843830; fax (þ44) (0) 1865 853333; e-mail:
[email protected]. Alternatively you can submit your request online by visiting the Elsevier web site at http://elsevier.com/locate/permissions, and selecting Obtaining permission to use Elsevier material. Notice No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made.
ISBN: 978-0-12-374165-3 ISSN: 0065-2113 For information on all Academic Press publications visit our website at books.elsevier.com
Printed and bound in USA 07 08 09 10 11
10 9 8 7 6 5 4 3 2 1
Contents CONTRIBUTORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PREFACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
xi xiii
THE EMERGING GLOBAL WATER CRISIS: MANAGING SCARCITY AND CONFLICT BETWEEN WATER USERS William A. Jury and Henry J. Vaux, Jr. I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Signs of the Coming Crisis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Population and Food Production Trends . . . . . . . . . . . . . . . . . . C. The Global Freshwater Resource . . . . . . . . . . . . . . . . . . . . . . . . D. Pollution and Human Health . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Challenges in Optimizing Water Use . . . . . . . . . . . . . . . . . . . . . II. The Present Global Water Situation . . . . . . . . . . . . . . . . . . . . . . . . A. Water Use by Sectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Water-Scarce and Water-Stressed Countries . . . . . . . . . . . . . . . . C. Drinking Water, Sanitation, and Waterborne Disease . . . . . . . . D. Chemical Contamination in Water . . . . . . . . . . . . . . . . . . . . . . . E. Water for Ecosystems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Groundwater Overdraft . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III. Population Trends and Water Stresses . . . . . . . . . . . . . . . . . . . . . . . A. Water-Short and Water-Stressed Countries . . . . . . . . . . . . . . . . B. Urbanization Trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Industrial and Municipal Water Demands . . . . . . . . . . . . . . . . . D. Transboundary Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Projected Water Deficit Under Business as Usual Practices . . . . F. Threats to Ecosystem Health . . . . . . . . . . . . . . . . . . . . . . . . . . . G. The Wild Card of Climate Change . . . . . . . . . . . . . . . . . . . . . . . IV. Dimensions of Water Scarcity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Water Savings Through Conservation . . . . . . . . . . . . . . . . . . . . B. Expansion and Improvement of Irrigation . . . . . . . . . . . . . . . . . C. Productivity Improvements in Rainfed Agriculture . . . . . . . . . . D. Economic Methods for Water Supplementation in Developing Countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Desalination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Improvements Through Institutional Changes . . . . . . . . . . . . . . V. Paths to Sustainability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Ending Unsustainable Practices . . . . . . . . . . . . . . . . . . . . . . . . . B. Management Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Agriculture and Water Management in the Developing World. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Societal Changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
2 3 4 6 7 7 8 9 10 12 16 22 24 26 26 30 32 34 38 40 42 43 44 45 50 51 53 55 57 58 62 66 67
vi
CONTENTS VI. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
69 69
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE Richard A. Dixon, Joseph H. Bouton, Brindha Narasimhamoorthy, Malay Saha, Zeng-Yu Wang and Gregory D. May I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Sequenced Genomes, Model Systems, and Comparative Genomics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. A. thaliana . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Rice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Poplar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Medicago truncatula and Lotus japonicus . . . . . . . . . . . . . . . . . . F. Genetic Resources for Functional Genomics . . . . . . . . . . . . . . . III. Transcriptomics, Proteomics, and Metabolomics . . . . . . . . . . . . . . . A. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Approaches for Transcript Profiling . . . . . . . . . . . . . . . . . . . . . . C. Proteomics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Metabolomic Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV. Molecular Markers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Marker Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Molecular Genetic Maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Genomics for Generation of Molecular Markers . . . . . . . . . . . . D. Metabolomic-Based ‘‘Markers’’ . . . . . . . . . . . . . . . . . . . . . . . . . E. Advantages of Marker-Assisted Breeding . . . . . . . . . . . . . . . . . . V. Transgenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Transgenesis as a Tool for Functional Genomics . . . . . . . . . . . . B. Current Approaches to the Generation of Transgenic Plants . . . C. Strategies for Overcoming Recalcitrance of Crop Species to Genetic Transformation . . . . . . . . . . . . . . . . . . . . . . . D. Transgenesis for Trait Integration and Commercialization . . . . . E. Virus-Induced Gene Silencing as an Alternative to Stable Transformation for Functional Genomics . . . . . . . . . . . . . . . . . F. TILLING as an Alternative to Transgenesis for Gene Knockdowns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VI. Case Studies for Alfalfa Improvement . . . . . . . . . . . . . . . . . . . . . . . A. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Improvement of Aluminum Tolerance . . . . . . . . . . . . . . . . . . . .
78 80 80 80 82 83 83 85 87 87 87 95 96 103 103 104 105 108 108 110 110 112 114 116 117 119 119 119 120
CONTENTS
vii
C. Gene Discovery and Metabolic Engineering for Forage
Quality Enhancement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Issues for Molecular Development of Alfalfa . . . . . . . . . . . . . . .
VII. The Future: Bridging the Gap from Models to Crops . . . . . . . . . . . VIII. The Future Technologies, Opportunities, and Challenges . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
124 132 132 134 138 139
THE MOLECULARIZATION OF PUBLIC SECTOR CROP BREEDING: PROGRESS, PROBLEMS, AND PROSPECTS Sangam L. Dwivedi, Jonathan H. Crouch, David J. Mackill, Yunbi Xu, Matthew W. Blair, Michel Ragot, Hari D. Upadhyaya and Rodomiro Ortiz I. Introduction to Global Food Production and Major Breeding Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Development of Markers for Assisting Selection . . . . . . . . . . . . . . . A. Genetic Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Genomic Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Genetic Linkage Map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Marker-Trait Associations from Analysis of Diverse Germplasm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III. Marker Validation and Refinement . . . . . . . . . . . . . . . . . . . . . . . . . A. Markers for Simply Inherited Traits . . . . . . . . . . . . . . . . . . . . . . B. QTL Marker for Complex Traits . . . . . . . . . . . . . . . . . . . . . . . . IV. Successful Applications of Marker-Assisted Genetic Enhancement in Public Sector Breeding Programs. . . . . . . . . . . . . . A. Resistance to Biotic Stresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Tolerance to Abiotic Stresses . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Agronomic and Seed Quality Traits . . . . . . . . . . . . . . . . . . . . . . D. Specific Challenges for Alien Gene Introgression . . . . . . . . . . . . V. Successful Application of Marker-Assisted Genetic Enhancement in Private Sector Breeding Programs . . . . . . . . . . . . . . . . . . . . . . . . VI. Impact of Marker-Assisted Genetic Enhancement . . . . . . . . . . . . . . A. Enhanced Selection Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Reduced Cost, Increased Feasibility, Time Savings, and Parental Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Overview of Products from Molecular Breeding . . . . . . . . . . . . . VII. Approaches to Enhance the EYciency and Scope of Molecular Breeding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Studying the Molecular Basis of Heterosis . . . . . . . . . . . . . . . . .
165 171 171 179 192 200 204 205 206 217 217 221 228 233 235 239 239 242 245 250 250
viii
CONTENTS B. Fine-Mapping, Cloning, and Pyramiding of QTL Associated
with Improved Agronomic Traits . . . . . . . . . . . . . . . . . . . . . . . . C. Expression QTL Mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Simulation and Modeling of MAS . . . . . . . . . . . . . . . . . . . . . . .
VIII. The Role of Computational Systems in Molecular Breeding Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Germplasm Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Managing Breeding Populations . . . . . . . . . . . . . . . . . . . . . . . . . C. Genetic Map Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Identifying Marker-Trait Associations . . . . . . . . . . . . . . . . . . . . E. Marker-Assisted Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. GEI Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G. Breeding Design and Simulation . . . . . . . . . . . . . . . . . . . . . . . . . H. Information Management and Integrated Tools . . . . . . . . . . . . . IX. Future Prospects for the Molecularization of Public Crop Improvement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
254 257 259 261 262 266 267 267 269 270 270 271 273 278 278
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE D. D. Stuthman, K. J. Leonard and J. Miller-Garvin I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Importance of Reliable Disease Resistance in Major Crops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Causes of Plant Disease Epidemics (The Disease Triangle) . . . . C. Examples of Plant Disease in Natural Populations . . . . . . . . . . . II. Concepts of Resistance to Plant Disease Useful in Breeding EVorts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. A Pragmatic Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Classification of Types of Resistance from a Breeder’s Perspective. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III. What Makes Disease Resistance Durable? . . . . . . . . . . . . . . . . . . . . A. Resistance in Wild Plant Species. . . . . . . . . . . . . . . . . . . . . . . . . B. Impact of Agriculture on Resistance. . . . . . . . . . . . . . . . . . . . . . C. EVorts to Delay Breakdown of Inherently Transient Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Examples of Durable Monogenic Resistance . . . . . . . . . . . . . . . E. Durability of Polygenic Resistance . . . . . . . . . . . . . . . . . . . . . . . IV. Examples of EVective Polygenic Resistance . . . . . . . . . . . . . . . . . . . A. Maize . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
320 320 322 327 328 328 332 335 335 336 338 340 342 343 344
CONTENTS B. Wheat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Barley . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Potato . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ix
V. Classical Breeding Approaches. . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Recurrent Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Pedigree Breeding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Perennial Species . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VI. Molecular Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Marker-Assisted Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Genetic Transformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VII. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
346 349 350 352 352 354 355 357 357 358 360 361
INDEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
369
See Color Plate Section in the Back of this Book
Contributors Numbers in parentheses indicate the pages on which the authors’ contributions begin.
Matthew W. Blair (163), Centro Internacional de Agricultura Tropical (CIAT), AA6713, Cali, Colombia Joseph H. Bouton (77), Forage Improvement Division, Samuel Roberts Noble Foundation, 2510 Sam Noble Parkway, Ardmore, Oklahoma 73401 Jonathan H. Crouch (163), International Maize and Wheat Improvement Center (CIMMYT), Apdo 0660 Mexico, D.F., Mexico Richard A. Dixon (77), Plant Biology Division, Samuel Roberts Noble Foundation, 2510 Sam Noble Parkway, Ardmore, Oklahoma 73401 Sangam L. Dwivedi (163), Agricultural Science Center at Clovis, 2346, SR288, Clovis, New Mexico 88101 William A. Jury (1), Department of Environmental Sciences, University of California, Riverside, California 92521; Department of Agriculture and Natural Resources, University of California, Berkeley, California 94720 K. J. Leonard (319), Cereal Disease Laboratory, USDA-ARS, St. Paul, Minnesota 55108 David J. Mackill (163), International Rice Research Institute (IRRI), DAPO, Box 3777, Metro Manila, Philippines Gregory D. May (77), National Center for Genome Resources, 2935 Rodeo Park Drive East, Santa Fe, New Mexico 87505 J. Miller-Garvin (319), Department of Agronomy and Plant Genetics, University of Minnesota, St. Paul, Minnesota 55108 Brindha Narasimhamoorthy (77), Forage Improvement Division, Samuel Roberts Noble Foundation, 2510 Sam Noble Parkway, Ardmore, Oklahoma 73401 Rodomiro Ortiz (163), International Maize and Wheat Improvement Center (CIMMYT), Apdo 0660 Mexico, D.F., Mexico Michel Ragot (163), Syngenta Seeds Inc., Stanton, Minnesota 55018 Malay Saha (77), Forage Improvement Division, Samuel Roberts Noble Foundation, 2510 Sam Noble Parkway, Ardmore, Oklahoma 73401 D. D. Stuthman (319), Department of Agronomy and Plant Genetics, University of Minnesota, St. Paul, Minnesota 55108 Hari D. Upadhyaya (163), International Crops Research Institute for the SemiArid Tropics (ICRISAT), Patancheru 502324, Andhra Pradesh, India Henry J. Vaux, Jr. (1), Department of Environmental Sciences, University of California, Riverside, California 92521; Department of Agriculture and Natural Resources, University of California, Berkeley, California 94720 Zeng-Yu Wang (77), Forage Improvement Division, Samuel Roberts Noble Foundation, 2510 Sam Noble Parkway, Ardmore, Oklahoma 73401 Yunbi Xu (163), International Maize and Wheat Improvement Center (CIMMYT), Apdo 0660 Mexico, D.F., Mexico xi
Preface Volume 95 contains four cutting-edge reviews in the agronomic sciences. Chapter 1 is a thought-provoking and timely review on ‘‘The Emerging Global Water Crisis: Managing Scarcity and Conflict Between Water Users.’’ The authors, William A. Jury and Henry J. Vaux, Jr., discuss signs of the coming crisis, the present global water situation, population and water stresses, dimensions of water scarcity, and paths to sustainability. Chapter 2 is a comprehensive review entitled ‘‘Beyond Structural Genomics for Plant Science.’’ Topics that are covered include sequenced genomes, model systems, and comparative genomics, transcriptomics and metabolomics, molecular markers, and transgenesis. Chapter 3 discusses the molecularization of public sector crop breeding and addresses progress, problems, and prospects. Chapter 4 deals with breeding crops for durable resistance to disease. Successes in durable resistance to multiple diseases of maize and to leaf rust and stripe rust of wheat are discussed. I thank the authors for their first-rate reviews. DONALD SPARKS University of Delaware Newark, Delaware
xiii
THE EMERGING GLOBAL WATER CRISIS: MANAGING SCARCITY AND CONFLICT BETWEEN WATER USERS William A. Jury1,2 and Henry J. Vaux, Jr.1,2 1
Department of Environmental Sciences, University of California, Riverside, California 92521 2 Department of Agriculture and Natural Resources, University of California, Berkeley, California 94720
I. Introduction A. Signs of the Coming Crisis B. Population and Food Production Trends C. The Global Freshwater Resource D. Pollution and Human Health E. Challenges in Optimizing Water Use II. The Present Global Water Situation A. Water Use by Sectors B. Water‐Scarce and Water‐Stressed Countries C. Drinking Water, Sanitation, and Waterborne Disease D. Chemical Contamination in Water E. Water for Ecosystems F. Groundwater Overdraft III. Population Trends and Water Stresses A. Water‐Short and Water‐Stressed Countries B. Urbanization Trends C. Industrial and Municipal Water Demands D. Transboundary Issues E. Projected Water Deficit Under Business as Usual Practices F. Threats to Ecosystem Health G. The Wild Card of Climate Change IV. Dimensions of Water Scarcity A. Water Savings Through Conservation B. Expansion and Improvement of Irrigation C. Productivity Improvements in Rainfed Agriculture D. Economic Methods for Water Supplementation in Developing Countries E. Desalination F. Improvements Through Institutional Changes V. Paths to Sustainability A. Ending Unsustainable Practices B. Management Strategies 1 Advances in Agronomy, Volume 95 Copyright 2007, Elsevier Inc. All rights reserved. 0065-2113/07 $35.00 DOI: 10.1016/S0065-2113(07)95001-4
2
W. A. JURY AND H. J. VAUX C. Agriculture and Water Management in the Developing World D. Societal Changes VI. Conclusions References
For the first time in human history, human use and pollution of freshwater have reached a level where water scarcity will potentially limit food production, ecosystem function, and urban supply in the decades to come. The primary reason for this shortage is population growth, which has increased at a faster rate than food production for some years and will add up to 3 billion more people by the middle of the twenty‐first century, mostly in poor and water‐short countries. Water quality degradation has also contributed significantly to a number of problems of global concern, including human drinking water supply and species survival. As of today, some 1.1 billion planetary inhabitants do not have access to clean drinking water, and 2.6 billion do not have sanitation services. Water pollution is a leading cause of death worldwide, and transmits or supports numerous debilitating diseases to populations forced to drink contaminated water. Agriculture is by far the leading user of freshwater worldwide, accounting for almost 85% of global consumption. Because of growing demand, we will need to raise food production by nearly 50% in the next 50 years to maintain our present per capita supply, assuming that the productivity of existing farmland does not decline. Further, we will have to increase it by much more if we are also to alleviate malnutrition among the poorest members of our current population. For a variety of reasons, feasible expansion of irrigated agriculture will be able to accommodate only a portion of this increased demand, and the rest must come from an increase in the productivity of rainfed agriculture. In the absence of coordinated planning and international cooperation at an unprecedented scale, the next half century will be plagued by a host of severe water‐ related problems, threatening the well being of many terrestrial ecosystems and drastically impairing human health, particularly in the poorest regions of the world. The latter portion of this chapter discusses ways in which this emerging crisis may be mitigated. # 2007, Elsevier Inc.
I. INTRODUCTION A century ago, the rivers of the world all ran wild and discharged the bulk of their contents into the seas. Groundwater use was limited to manual extraction from wells that only tapped the near surface, and crops were grown mostly with rainwater. Wetlands existed wherever nature intended them to be, and provided both habitat for waterfowl and a host of water regulation services. Water pollution was caused mainly by disposal of human sewage, added in small enough quantities that only the immediate zones surrounding the emissions were adversely aVected. The oceans were thriving
THE EMERGING GLOBAL WATER CRISIS
3
with life, and species reproduced rapidly enough to balance any losses from human consumption. A little more than 1.5 billion humans inhabited the planet, less than one quarter of today’s population. And except in extremely arid zones, they had plenty of water. The next 100 years will be quite diVerent than the last century, as another 3 billion or so humans join the current population of 6.5 billion. Without immediate action and global cooperation, a water supply and water pollution crisis of unimaginable dimensions will confront humanity, limiting food production, drinking water access, and the survival of innumerable species on the planet.
A. SIGNS OF THE COMING CRISIS This dire forecast is based on an extrapolation of current activities and trends on the planet. First, unlike estimates of the global supply of scarce minerals or underground fuels which are surrounded by uncertainty, planetary supplies of water are relatively well characterized. There are no large groundwater deposits awaiting human detection in readily accessible locations, so that any new resources discovered will be very expensive to develop. Second, many vital human activities have become dependent on utilizing groundwater supplies that are being exhausted or contaminated. Third, much of the population growth projected for the next century will occur in areas of greatest water shortage, and there is no plan for accommodating the increases. Finally, global economic forces are luring water and land from food production into more lucrative activities, while at the same time encouraging pollution that impairs drinking water quality for a large and ever‐ growing segment of the population. These and other signs indicate that we are heading toward a future where billions of people are forced to live in locations where their needs for food and potable water cannot be met. This is not the first time that modern civilization has faced an impending food crisis. In 1950, the world produced 630 million tons of grain for its population of 2.5 billion humans, a yield that was insuYcient to prevent starvation in certain regions. Most notably, China suVered a massive famine at the end of the decade that killed as many as 30 million people (Smil, 1999), prompting talk that the global population might have reached or exceeded the maximum number of people who could be fed by existing resources. But the Green Revolution changed the earth’s productivity dramatically through a combination of crop breeding strategies, fertilization, pest control, and irrigation (Borlaug, 2002). By 1990, grain production had risen to 1.77 billion tons to feed a population of 5.3 billion, an increase of 2.8 times the 1950 yield to provide for less than 2.1 times the number of people. This dramatic increase in productivity has had the eVect of both assuaging fears of global
4
W. A. JURY AND H. J. VAUX
famine and giving the world public a sense that human ingenuity would always be able to produce the technology it needed to survive and prosper. In the nearly half century since the great Chinese famine, global agriculture has steadily produced impressive increases in crop yield. Today, irrigation, synthetic fertilizers, and pesticides are in widespread use in all but the poorest parts of the world. China and India, each with over a billion inhabitants, are both able to feed their huge populations. Yet, there is as much concern today about exceeding the planetary carrying capacity as there was in the days immediately prior to the Green Revolution. The reason for the concern is that the global population has vaulted upward to 6.5 billion in 2006, and for some time has been increasing at a rate which outpaces gains in food production. The best current population forecasts are that the world will have 7.9 billion people by 2025 and 9.2 billion by 2050. Thus, to maintain our present per capita supply we will need to raise food production by nearly 50% in the next 50 years, assuming that the productivity of existing farmland does not decline. Further, we will have to increase it by much more if we are also to alleviate malnutrition among the poorest members of our current population. Meeting future food demand will be a significantly more challenging task than the world faced prior to the Green Revolution when agricultural eYciency was low everywhere.
B. POPULATION AND FOOD PRODUCTION TRENDS Food production and population have both been increasing steadily since the dawn of the Green Revolution, but the latter has been rising more rapidly for decades. One way to visualize the relative growth of these two dynamic variables is to look at the global grain yield (wheat, rice, and coarse grains) per person as a function of time since the Green Revolution began (Fig. 1). This index peaked in the early 1980s and has gradually declined since, reaching a low of 15% below its maximum value in 2003 before rebounding in 2004–5. At the same time, the ratio of global grain stocks to annual consumption has fallen steadily during the last decade to an all time low (Fig. 2). The cause and significance of the declining grain yield per capita are a matter of debate. To some, it indicates that a crisis in food production is looming which threatens to make many countries that are currently self‐ suYcient into food importers, fighting for a declining supply of surplus (Brown, 2004). To others, the decline has been caused mainly by market forces and is not indicative of a limit to yield potential (FAO, 2003). Regardless of the explanation for the slowing of grain yield increases relative to population growth, the trend is a cause for concern if only because population growth has not ceased and significant increases in global crop yield will be necessary to avert food shortages in the future.
THE EMERGING GLOBAL WATER CRISIS
5
Per capita grain yield (kg)
350 325 300 275 250 225 1950
1960
1970
1980
1990
2000
2010
Year Figure 1 Total grain yield per person (data taken from FAS, 2006).
Grain stock (% of Annual consumption)
40 35 30 25 20 15 1980
1985
1990
1995
2000
2005
Year Figure 2 Ratio of grain stocks to annual consumption as a function of time (data taken from FAS, 2006).
The optimists among those who predict the future of food production have many facts to bolster their arguments. Crop yields in numerous poor countries are far below maximum attainable levels that have been reached elsewhere (FAO, 2003). Substantial additional land is available for agricultural expansion (Greenland et al., 1998). Introduction of irrigation technology to areas with marginal rainfall for crops can produce substantial benefits (Postel, 1999). And genetic alteration of plant species could greatly improve the productivity of agriculture (Hoisington et al., 1999).
6
W. A. JURY AND H. J. VAUX
However, the pessimists are no less able to find support for their contention that the future world will be challenged to provide the food it needs to survive and prosper. A significant part of the world’s agricultural land is being managed unsustainably, and cannot continue to be farmed indefinitely (Eswaran et al., 2001). Market forces in developing countries are driving the conversion of agricultural land to urban or industrial use. Loss of topsoil from water or wind erosion is decreasing the fertility of many soils. And perhaps the most compelling of all the arguments made by those looking with trepidation at the next 50 years on earth is simply that we may be running out of freshwater.
C. THE GLOBAL FRESHWATER RESOURCE About 97.4% of the water on the planet is in the oceans, and is too saline for beneficial use without treatment. Most of the rest of the water (about 2%) is also unavailable because it is locked up in polar ice or glaciers. Humans and all other terrestrial life must subsist on the remaining 0.6%. The global freshwater (nonsaline) resource that is potentially available for human use is divided into groundwater or surface water in rivers, lakes, and reservoirs, which together total about 475 million km3 (Shiklomanov, 1997). This is a staggering amount, but focusing on the global freshwater storage resource alone is misleading because much of the water is inaccessible. For that reason, it is more sensible to consider humanity’s freshwater resource as consisting of three sources: rainfall used to grow crops, accessible groundwater, and surface water. Falkenmark and Rockstrom (2004) divide this resource into two categories which they call blue water and green water. Blue water is the liquid resource remaining after evaporation, and green water is the water originating as rainfall that subsequently returns to the atmosphere after evaporation or transpiration. Transpired rainwater is clearly a vital part of the resource for food production, and must be figured into estimates of present or future water shortages. The blue water resource of global runoV potentially accessible to humans is diYcult to estimate, and has considerable uncertainty (Postel et al., 1996). A frequently quoted value is 42,700 km3 (Shiklomanov, 1997). Global runoV is not evenly distributed over the planet’s surface, so that there are some regions with excess water and others with chronic shortage. In regions with excess water, much of the volume flow of rivers and streams reaches the ocean without being used by humans, although it serves important environmental purposes. For example, 20% of average global runoV flows through the Amazon River, where it is mostly unutilized by the indigenous population (Gleick, 1998). Also, substantial flow reaches the Arctic Ocean from six major Eurasian rivers that are scarcely touched by humans
THE EMERGING GLOBAL WATER CRISIS
7
(Peterson et al., 2002). In contrast, large areas of the globe receive low rainfall and are water deficient. Regions experiencing the greatest shortfall of freshwater are the Middle East, significant portions of Africa, and some parts of Europe and Southeast Asia (Postel, 1997).
D. POLLUTION AND HUMAN HEALTH Not all of the freshwater resources are fit for human consumption. The World Health Organization estimated in 2000 (WHO/UNICEF, 2000) that 1.1 billion people on the planet lacked access to safe drinking water (Table I), and 2.6 billion did not have sanitation services. Indeed, water pollution is a leading cause of death worldwide, and transmits or supports numerous debilitating diseases to populations forced to drink contaminated water. Because of continued population growth and rapid economic development in a number of countries with little or no water quality monitoring or regulation, water pollution from industrial, municipal, or agricultural sources is growing worse in many regions, and threatens to further reduce the supply of usable water in countries experiencing scarcity.
E. CHALLENGES IN OPTIMIZING WATER USE Determining whether a country has a suYcient water supply to serve both its present and future population is a complex matter because the relationship between population and water demand is contingent on many factors. A nation with enough wealth to import the food it needs to feed its population has a greatly reduced water demand compared to one that must grow its own nourishment. Moreover, a nation that can grow crops using only rainfall has a very diVerent water budget than one relying on irrigation. Eating habits are also very important in determining water demand, particularly when meat is a significant portion of the diet.
Table I Population in Millions Lacking Access to Safe Drinking Water in 2000 (Pacific Institute, 2003) Region Africa Asia Latin America and Caribbean Europe World
Rural
Urban
Total
256 595 49 23 926
44 98 29 3 173
300 693 78 26 1099
8
W. A. JURY AND H. J. VAUX
Because human needs for water will take precedence over all other water demands, providing the water needed to maintain ecosystem protection in the presence of conflicting water demands will pose perhaps the greatest challenge of all, given the high market value of water for industrial and municipal uses, and the obvious priority water‐short countries will place on providing food and water services for their population. Substantial damage has been done to the world’s ecosystems in the last century by modifying natural water courses, for example, by draining wetlands, constructing dams, excessive pumping of groundwater, or exhausting the flow of rivers. Countless other ecosystems have been stressed by water pollution originating from human activity. Optimizing water use among the agricultural, industrial, municipal, and environmental sectors cannot be achieved without consensus agreement on a clear set of priorities and a commitment to the long‐term well being of the planet. Numerous global trends are pointing to a growing water crisis that could have devastating consequences for human health, economic stability, and ecosystem protection by the middle twenty‐first century, if not sooner. Averting this crisis will require international cooperation on an unprecedented scale, using a mix of technological and institutional procedures designed to utilize water more eYciently and optimize its benefits to humans and the environment.
II. THE PRESENT GLOBAL WATER SITUATION Estimates of the annual surface water flow that is potentially available for human use vary considerably, but the figure of 12,500 km3 year1 used by Falkenmark and Rockstrom (2004) suYces for purposes of illustration. This volume is considerably less than the figure of 42,700 km3 year1 quoted for global surface water runoV, but a substantial portion of the latter is either geographically unavailable, necessary for groundwater recharge, or temporally unavailable (Postel et al., 1996). Although the global freshwater supply in the aggregate is more than suYcient to meet all current and forecasted demands for consumptive use, this statement fails to represent the status of the planet’s freshwater resource because there are many regions where indigenous supplies are completely inadequate to support sustainable food production and other consumptive uses. Global water consumption by humans is increasing each year as population rises and developing countries increase their degree of urbanization and industrialization. Figure 3 shows the results of a comprehensive analysis of human water consumption for years up to 1995, with projections to 2025 (Shiklomanov, 1997).
THE EMERGING GLOBAL WATER CRISIS
9
6000
Annual consumption (km3 year−1)
5000 4000 Projection 3000 2000 1000 0 1900
1925
1950
1975
2000
2025
Year Figure 3 Global freshwater consumption in km3 year1 with projections to 2025 (Data taken from Shiklomanov, 1997).
Although the use projected for 2025 is still less than 25% of the surface water supply, humans cannot utilize all of the available surface water without destroying ecosystems that depend on water for survival. Riparian ecosystems, for example, require that a substantial fraction (e.g., 30%; Falkenmark and Rockstrom, 2004) of the annual flow volume must be maintained for adequate ecological health. This volume is not being provided in a number of riparian ecosystems today, and future demands will surely place even greater stress on the environment. Postel et al. (1996) estimated that about 18% of all available water in 1990 was used directly by humans, and an additional 34% is necessary for proper ecosystem function. They projected that these two needs could comprise as much as 70% of available runoV by 2025. Viewed in that light, it is easier to see that human freshwater use, even averaged on a global basis, is going to be a significant fraction of the available resource in the coming century.
A. WATER USE BY SECTORS In 1995, about 3800 km3 of freshwater was withdrawn from surface water or groundwater supplies for human use (Shiklomanov, 1997). Of that amount, some 2100 km3 was consumed, and thus removed from the supply base. Table II summarizes how that water was distributed among the four major use categories. Several facts on this table are worth noting. First, agriculture is overwhelmingly the dominant consumer of freshwater, accounting for nearly 85% of all water lost. Most of this is due to plant
10
W. A. JURY AND H. J. VAUX
Table II Annual Global Withdrawal or Consumption of Freshwater in km3 year1 by Use Category in 1995 (Data taken from Shiklomanov, 1997) Sector or use category Agriculture Municipal Industry Reservoir losses Total
Annual withdrawal
Total withdrawal (%)
Annual consumption
Total consumption (%)
2504 344 752 188 3788
66.1 9.1 19.9 4.0 100.0
1753 50 83 188 2074
84.5 2.4 4.0 9.1 100.0
transpiration of irrigation water. Second, although industry and municipal withdrawals of water are significant (29% of the total), much of their withdrawal is returned to the supply base, leaving only 6.4% that is actually consumed. Finally, the second highest consumption category is due to reservoir losses from evaporation and leakage. Quality reductions in the water returned to the freshwater supply are not reflected in the figures given in Table II. For example, the 2504 1753 ¼ 751 km3 of water withdrawn by agriculture but not consumed includes a substantial component of agricultural drainage water that is higher in salt, nutrient, and pesticide concentration than the supply water originally withdrawn. In addition, domestic and industrial water returned to the source in developing countries is often highly polluted, and thus is not only unusable for many purposes but also degrades the remaining supplies.
B. WATER‐SCARCE AND WATER‐STRESSED COUNTRIES The distribution of freshwater around the globe is highly uneven, leading to regional shortages or excesses that are not apparent from the global average figures. Moreover, the amount of water a country needs depends mostly on whether it grows or imports the food to feed its population, and how much rain it receives. There are also substantial diVerences in household and industrial water consumption between poor and wealthy countries. The global average water requirement for food production has been estimated as 1200 m3 year1 per person (Rockstrom et al., 1999). This is about 70 times more than the estimated 18.2 m3 year1 per person that represents average per capita household use (Falkenmark and Rockstrom, 2004). Not unexpectedly, the per capita consumption of water for nonagricultural use (domestic, service, and industry) is much higher in developed countries. The United States averages about 366 m3 year1, Europe 232 m3 year1, and Africa only 25 m3 year1 (Falkenmark and Rockstrom, 2004).
THE EMERGING GLOBAL WATER CRISIS
11
Classification of the degree of water security of a given country may be done in a variety of ways. The most commonly used index is the Falkenmark Stress Indicator (FSI), which classifies a country in diVerent categories of water shortage based on per capita liquid water resource availability (PWR) (surface water flow or groundwater recharge). This index has been divided into three regions for classification purposes: (1) PWR > 1700 m3 year1, which is regarded as the amount required for water self‐suYciency, allowing a country to grow the food it needs to feed its population, and to provide all services needed for human and ecosystem health; (2) 1000 < PWR < 1700 m3 year1, which indicates water stress; and (3) PWR < 1000 m3 year1, which denotes chronic water scarcity. A PWR of 500 m3 year1 or less is considered to be a water barrier, below which a country depending on irrigation cannot avoid salinization problems and progressive loss of agricultural land. As indicated previously, the figure of 1700 m3 year1 is comprised largely of the 1200 m3 year1 per person required to produce food. Thus, the stress index primarily indicates whether a country relying on irrigation has suYcient water to grow the food it needs to feed its own population. Although this index is arbitrary, it does allow an objective assessment of regional water availability. Table III summarizes the number of countries experiencing water stress or scarcity in 1995 according to this index. Of the 18 water‐scarce countries, 9 are in the Middle East, and 6 in Africa, primarily in the extreme north. The FSI is only one of several diVerent ways of representing water scarcity, and at best provides a qualitative measure of a country’s present or future degree of food and water security. As an alternative, Raskin et al. (1997) defined water scarcity in terms of the total volume of water withdrawn annually as a percentage of a country’s annual water resources. This study, sponsored by the United Nations Commission on Sustainable Development, classified a country as water scarce if its annual withdrawals exceeded 40% of its total resource. Seckler et al. (1998) used this index together with
Table III Population and Numbers of Countries Experiencing Water Stress or Scarcity in 1995 According to FSI (Data adapted from Population Reports, 1998)
Category Water scarce Water stressed Water scarce or stressed a
Annual water resources (m3 year1 per person)
Countries
Population (millions)
PWR < 1000 1000 < PWR < 1700 PWR < 1700
18 (12)a 11 29
166 (65) 294 460
Number in parentheses indicates countries below water barrier of PWR < 500 m3 year1 per person.
12
W. A. JURY AND H. J. VAUX
an estimate of the projected percentage increase in withdrawals between 1995 and 2025 to place countries in five diVerent groups of water availability. Group 1, with both indices above 50%, was deemed the most problematic. Although the Falkenmark, Raskin, and Seckler water stress indices have some common elements, they do not produce the same classification when applied to the countries of the world.
C. DRINKING WATER, SANITATION, AND WATERBORNE DISEASE There are surprisingly few sources of pristine water remaining in the modern world. Even rainfall can contain substantial amounts of chemicals arising from air pollution or agricultural emissions. Rivers, streams, and lakes become contaminated from a variety of industrial, agricultural, or municipal sources, as well as from individual septic tanks or other household waste disposal practices. The nature of the pollution varies with the level of development of the region, and depends as well on whether the host country has waste control policies and cleanup procedures. In modern wealthy countries, water sources are monitored and either treated or isolated from human contact if harmful levels of pollution are present. These societies regard safe drinking water and adequate sanitation as basic rights granted to all their citizens. Yet for a significant part of the world these services are woefully inadequate. The World Health Organization estimated that 1.1 billion people lacked access to clean drinking water in 2002, and 2.6 billion did not have sanitation services. The problem is particularly bad in rural parts of Africa and Asia, where the majority of the citizens have no sanitation or freshwater access. Table IV shows the percentage of the population with drinking water and sanitation services in various regions of the world. These numbers show clearly that the poorer regions of the world lag far behind industrialized nations in water supply and sanitation access. Table IV Percentage Water and Sanitation Coverage by Region (Data taken from WHO/UNICEF, 2000) Region Africa Asia Latin America and Caribbean Oceania Europe North America World
Water supply
Sanitation
62 81 85 88 96 100 82
60 48 78 93 92 100 60
THE EMERGING GLOBAL WATER CRISIS
13
The 1.1 billion poor people in the world forced to drink contaminated water in order to survive are subjected to a host of debilitating and even fatal diseases that are virtually unknown in countries with safe drinking water and adequate sanitation services. The most widespread of the waterborne diseases are those arising from human or animal waste contamination. The World Health Organization reported that, of the 51 million deaths worldwide in 1993, about one‐third (16.4 million) were caused by infectious and parasitic diseases. In developing countries these totals are even higher, with infectious and parasitic diseases accounting for 44% of all deaths and 71% of deaths in children (World Development Report, 1993). There is insuYcient data to determine how much of global infectious disease is waterborne, although estimates of up to 80% have been given (Clarke, 1993). The United Nations and the World Health Organization were suYciently concerned about the water problems of poor nations that they designated the 1980s as the International Drinking Water Supply and Sanitation Decade, whose stated goal was to ‘‘Provide every person with access to water of safe quality and adequate quantity, along with basic sanitary facilities, by 1990.’’ Although the goal was not reached, the UN/WHO action focused attention on the problem and greatly increased funding to address global deficiencies. As a result, rural water supply increased by 240% and sanitation access grew by 150% in rural areas between 1980 and 1990. Although urban water supply and sanitation also increased by 150% as a result of the eVort, there was a net decrease during the decade in percentage access because of the large rise in urban population. Service provision eVorts have continued to increase globally in the decades since, although some of the poorest countries have not been able to increase water supply and sanitation services as fast as population has grown. The percentage of the world population with access to water supply increased from 76% to 82% between 1990 and 2000, although the absolute number of people without service remained constant at about 1.1 billion. At the same time, the percentage with access to sanitation increased from 55% to 60%, but again the numbers without services changed little and remained at about 2.4 billion. The United Nations created the Millennium Development Goals, which were adopted in 2000 by all the world’s governments as a blueprint for building a better world in the twenty‐first century. One target of the environmental sustainability goal was a plan to halve by 2015 the proportion of people without sustainable access to safe drinking water and basic sanitation. The assessment of the 2006 Progress Report was that the world was unlikely to reach this target in the designated time frame (UN, 2006). The major source of pollutants in developing countries is by direct disposal of domestic and industrial wastewater into rivers, lakes, or on land. Emerging Asia, published by the Asia Development Bank in 1997, identified water pollution as the most serious environmental problem facing the continent
14
W. A. JURY AND H. J. VAUX
(ADB, 1997). The majority of the world’s poor people reside within Asia, and over half of the total global population. Although significant eVorts were made across the continent to improve both drinking water quality and access to sanitation services during the 1980s, population growth during the same time period erased much of the progress. As of 1990, some 850 million people in Asia had no access to fresh drinking water, and 2.1 billion had no basic sanitation facilities. In Asia and the Pacific, fecal pollution is one of the most serious problems, aVecting both surface water and groundwater and causing a host of waterborne diseases such as cholera, typhoid, and hepatitis. Currently, over 80% of river lengths in the Hai and Huai basins in China are classified as very highly polluted and cannot meet any designated beneficial uses. Estimates of the increase in water pollution loads in high‐growth areas of Asia over the next few decades are as high as 16 times for suspended solids, 17 times for total dissolved solids, and 18 times for biological pollution loading (UNIDO, 1996). Parasites in water or insects breeding in water cause a host of illnesses, the most serious of which are fatal. Table V lists the most serious of these water diseases, their morbidity, and the deaths they cause. These diseases kill over 5 million people per year, and incapacitate even more. In addition, other diseases that are not generally fatal can cause a variety of incapacitating injuries. For example, onchocerciasis and trachoma are responsible for over 6 million cases of blindness or equally severe complications. Diarrhea is caused by a number of diVerent bacterial, viral, and parasitic organisms present in contaminated water, and is a major cause of death for children who do not have access to clean water. It has been estimated that diarrhea causes 4% of all deaths and 5% of disability (WHO/UNICEF, 2000). Malaria is caused by four species of Plasmodium parasites. It does not infect humans through water contact, but rather is transmitted by mosquitoes which breed in stagnant water. The disease is among the five leading causes of death in children under 5 years of age in Africa. In many regions Table V Estimated Morbidity and Mortality for Various Water‐Related Diseases (Data from Gleick, 2002 and WHO, 2004) Disease Diarrhea Malaria Schistosomiasis Trypanosomiasis Intestinal helminths Dengue fever Onchocerciasis
Annual cases (millions)
Annual deaths (thousands)
1000 400 200 0.27 1500 1750 18
3300 1500 20 130 100 20 40
THE EMERGING GLOBAL WATER CRISIS
15
where malaria is present, the natural habitat is wet enough to provide the breeding ground for mosquitoes. However, the development of irrigation systems, dams, and reservoirs in regions lacking a natural mosquito habitat has caused the disease to spread. In other regions, for example, the Central Asian republics, malaria has returned because of the deterioration of water management facilities. Schistosomiasis is an infection caused by three diVerent species of flatworm that develop in freshwater snails. It infects humans who contact contaminated water by ingestion of the flatworms through the skin. At least 600 million people are at risk of infection and 200 million currently have schistosomiasis, about 80% of which are in sub‐Saharan Africa. Of those infected, some 20 million have a severe and potentially fatal form of the disease. Water resource schemes for power generation and irrigation have resulted in a tremendous increase in the transmission and outbreaks of schistosomiasis in several African countries. In northern Senegal, an area without intestinal schistosomiasis before the building of the Diama dam in 1986, virtually the entire population had become infected by 1994. Trypanosomiasis, known as sleeping sickness, is an insect vector‐borne parasitic disease caused by protozoa transmitted to humans by tsetse flies, which breed along rivers, streams, and lakes. The disease occurs only in sub‐ Saharan Africa, in regions where tsetse flies are endemic. It currently threatens over 60 million people in 36 countries of sub‐Saharan Africa. In certain provinces of Angola, the Democratic Republic of Congo, and southern Sudan, the prevalence of trypanosomiasis is between 20% and 50% of the population, making it the first or second leading cause of death in those regions. Intestinal helminths are parasitic worms that cause intestinal infections. It is estimated that 133 million people suVer serious complications from these parasites, such as massive dysentery, anemia, or brain damage. Ascariasis, caused by the Ascarias worm, is one of the most common human parasitic infections. Up to 10% of the population of the developing world is infected with intestinal worms—mainly by Ascaris. Worldwide, severe Ascaris infections cause 60,000 deaths per year, mostly children. Infection occurs with greatest frequency in tropical and subtropical regions, and in any areas with inadequate sanitation. Dengue fever is a mosquito‐borne infection causing a severe, flu‐like illness that aVects infants, young children, and adults but rarely causes death. Dengue hemorrhagic fever (DHF) is a potentially lethal complication and is today a leading cause of childhood death in several Asian countries. It is characterized by high fever, hemorrhage, liver enlargement, and circulatory failure in the most severe cases. Dengue has spread dramatically through the world in recent decades, and is considered a major health threat today. Globally there are an estimated 50–100 million cases of dengue fever and 500,000 cases of DHF each year. The disease is now found in more than
16
W. A. JURY AND H. J. VAUX
100 countries in Africa, South and Central America, the Eastern Mediterranean, South and Southeast Asia, and the Western Pacific. A substantial fraction of the world’s population is at risk from water‐ related diseases, and the regions where the risk is greatest also are among the most rapidly growing in population. It is likely that most of the 2 billion people added to the world due to growth in population between 2000 and 2025 will be at risk for the diseases in Table V unless significant improvements are made in global sanitation and drinking water quality.
D. CHEMICAL CONTAMINATION IN WATER Although the most advanced of the industrialized countries have largely controlled biological contamination of water, many of them have seriously polluted their surface water and groundwater supplies with agricultural, municipal, or industrial releases of toxic chemicals. Even in wealthy countries committed to remediation, cleanup of badly contaminated sites has been extremely time consuming and expensive. In 1994, the Congressional Budget OYce estimated that it could take as much as US$75 billion to clean up the remaining 4500 non‐Federal sites on the Superfund list. Similarly, the overall extent of the financial public and private environmental clean up liability risk in Germany is thought to be between 200 and 500 billion euros, mostly due to contamination in the former East Germany (Freshfields, 2003). Indeed, the evidence shows that it is almost always cheaper to prevent pollution than to remediate it once it has been released to the environment. In contrast to pathogenic contamination, chemical pollution tends to be localized and reflective of the land use around it. Agricultural nutrients and pesticides seep to groundwater below cropped fields or concentrated feedlots, or accumulate in surface waters receiving agricultural runoV from irrigated fields. Industrial contamination occurs from leaks, accidents, or deliberate dumping, and depending on the industrial processes can contain a host of persistent toxic metals or organic compounds. Municipal waste might include untreated sewage in a developing country or toxic stormwater runoV in a developed nation. Chemical contamination is diYcult to monitor in the subsurface and expensive to analyze, so that much of it is uncharacterized. In the following sections, the major chemical pollutants found in groundwater and surface water will be briefly discussed. 1. Agricultural Nutrients Agricultural nutrients can cause significant changes in aquatic systems. Eutrophication is the term used to describe the process through which surface waters are enriched with nutrients. There are natural eutrophication
THE EMERGING GLOBAL WATER CRISIS
17
processes which cause lakes and streams to evolve ecologically. There are also simpler human‐induced eutrophication processes which are driven by the residues of fertilizers and other nutrient‐rich materials that trigger artificial and sometimes unstable changes in aquatic ecosystems. Phosphates and nitrates are often the limiting factors in the growth of algae. Addition of the limiting chemicals can trigger significant algal blooms which are followed by a die‐oV and consumption of algal biomass by bacteria. This latter process can consume oxygen to levels that are toxic to fish and other aquatic organisms. Eutrophication can cause a number of undesirable eVects, including: increase in production and biomass of phytoplankton and algae, shift in habitat characteristics, replacement of desirable fish by less desirable species, production of toxins by certain algae, lowering of oxygen levels by microbial respiration, and loss of functionality of the water resource (Ongley, 1996). The principal agricultural nutrients with potentially harmful environmental consequences are nitrogen and phosphorus. Both are added in large quantities in modern fertilized agriculture, and tend to be more of a problem in developed countries with farmland under intense cultivation. Phosphorus binds tightly to soil particles, preventing it from moving deep into the soil with drainage water. Thus, it is seldom seen in groundwater. However, it is readily transported with sediment during lateral runoV, allowing it to reach streams and lakes where it can trigger an explosive growth of algae. Phosphorus does not reach high concentration levels in water, and there is no health risk associated with exposure to phosphorus in the natural environment. Nitrogen undergoes a number of reactions in soil, and under normal conditions of adequate oxygen culminates in formation of the stable nitrate ion, which is very soluble, does not bind to stationary soil particles, and is extremely mobile in soil. If not taken up by plants, nitrates can seep below the root zone to groundwater, and move laterally to surface water with runoV. In surface water, nitrogen contributes to eutrophication of receiving bodies, and can alter the aquatic ecology through weed proliferation and algae growth. Water containing high concentrations of nitrate can have adverse health eVects. Infants under 6 months of age are most sensitive to elevated levels of nitrates in drinking water. A baby fed water high in nitrates may develop a condition called methemeglobinemia, in which the blood is unable to properly carry oxygen. The condition can be fatal, if oxygen deprivation is severe and lengthy enough. For this reason, the public health limit for nitrate in drinking water has been set at 10 mg liter1 NO3‐N (45 mg liter1 NO3) in the United States and slightly higher (50 mg liter1 NO3) in Europe. Deaths from methemeglobinemia are extremely rare in the United States and Western Europe, and those that have occurred were generally in rural areas where drinking wells had been contaminated to high levels of nitrate by septic tanks or other sources of concentrated N emissions. Under high‐ risk conditions, such as intensely fertilized agricultural fields containing
18
W. A. JURY AND H. J. VAUX
well‐drained soil above shallow groundwater, groundwater concentrations of nitrate can rise above the public health limit. Nolan et al. (2002) reported that 26% of the wells sampled in high‐risk areas of the Midwest and western United States had concentrations above 10 mg liter1. For the period 1992– 1996, over 65% of the rivers in the European Union had average annual nitrate concentrations exceeding 1 mg liter1, and 15% had concentrations over 7.5 mg liter1. The highest levels were found in Northwest Europe, where agriculture is intensive. A limited number of studies have suggested other health eVects linked to nitrates, including spontaneous abortions (Centers for Disease Control, 1996), bladder cancer (Weyer et al., 2001), and non‐Hodgkin’s lymphoma (Law et al., 1999).
2.
Agricultural Pesticides
The global use of chemical pesticides has undergone three stages of evolution. Until the early 1900s, inorganic toxins such as arsenic, copper, and lead were used around the world to kill a variety of pests. However, these chemicals were toxic to all organisms they contacted, and persistent in the environment. They were eventually restricted from use as pesticides. The second stage of pesticide use began in the 1940s when synthetic organic compounds, consisting of either chlorinated hydrocarbons, carbamates, or organophosphates, were introduced for pest control. The chlorinated hydrocarbons, such as DDT, dieldren, and lindane, are nerve toxins that act on any organism with a central nervous system. They are persistent and bioaccumulate readily in the environment. The organophosphates and carbamates are less persistent, but much more toxic to humans. They also are significantly more expensive to use than the chlorinated hydrocarbons. The most recent stage of pesticide management is the use of natural toxins in connection with a suite of procedures known as integrated pest management. Natural biocides, such as the bacterial toxin bacillus thuringiensis, are easily degraded into nontoxic forms. They are also much narrower in their toxicity and attack a specific target organism. Modern pest management techniques, when widely adopted, could have substantial beneficial impacts on water quality. The Western world has begun making the transition into the third stage of pesticide use. However, developing countries still use large quantities of toxic and persistent chemicals because they are significantly less expensive to employ. As a result, pesticides have become a major health threat in the developing world, largely due to handling and exposure during application. An estimated 3 million reported cases of pesticide poisoning occur annually, resulting in 220,000 deaths (WHO, 1990). About 99% of these occur in the
THE EMERGING GLOBAL WATER CRISIS
19
developing world, despite the fact that developing countries account for only 20% of global pesticide use. The extent of pesticide contamination of surface water or groundwaters around the globe is largely unknown. It is extremely expensive and laborious to measure pesticide concentrations in soil or water, and only a few comprehensive surveys have been conducted. In the United States, the US EPA’s National Pesticide Survey found that 10.4% of community wells and 4.2% of rural wells contained detectable levels of one or more pesticides (US EPA, 1992). More than 68,000 groundwater wells in 45 states were sampled in this survey, and pesticides were detected in nearly 25% of the wells and in 42 states. Analysis was carried out for a total of 605 pesticides and related compounds, of which 265 were detected at least once. Of the pesticides detected, 28 are no longer in use in the United States, and regulatory restrictions have been placed on 54. In a study of groundwater wells in agricultural southwestern Ontario (Canada), 35% of the wells tested positive for pesticides on at least one occasion (Lampman, 1995). The Netherlands National Institute of Public Health and Environmental Protection concluded that groundwater was threatened by pesticides in all European states. They reported that the EC standard for the sum of pesticides (0.5 mg liter1) will be exceeded on 65% of all agricultural land, and that the standard will be exceeded by more than an order of magnitude on 25% of the area (RIVM, 1992). The National Water Quality Assessment (NAWQA) program of the US Geological Survey (USGS) represents the most comprehensive national‐scale analysis to date of pesticide occurrence and concentrations in streams and groundwater of the United States (Gilliom et al., 2006). This decade‐long survey from 1992 to 2001 conducted assessments of 75 pesticides and 8 degradation products in surface water, groundwater, and sediments in 51 US major river basins and aquifer systems. At least one pesticide was detected in water from all streams studied, and pesticide compounds were detected throughout most of the year in water from streams with agricultural, urban, or mixed‐land‐use watersheds. Organochlorine pesticides (such as DDT) and their degradation byproducts were found in fish and bed‐sediment samples from most streams in agricultural, urban, and mixed‐land‐use watersheds, and in more than half the fish from streams with predominantly undeveloped watersheds. Pesticides were less common in groundwater than streams. They were found most frequently in shallow groundwater beneath agricultural and urban areas, where more than 50% of wells contained one or more pesticide compounds. Detections were often at low concentrations, and NAWQA personnel estimated that less than 10% of their monitored stream sites and about 1% of wells surveyed had concentrations greater than levels deemed to be high enough to aVect human health.
20
W. A. JURY AND H. J. VAUX
Use of pesticides in developing countries is quite variable, ranging from none in much of Africa to extremely heavy use in intensive agricultural areas of Brazil and plantations of Central America. Pesticide use in the Brazilian state of Parana´ is typical of developing countries undergoing rapid expansion of agriculture. Andreoli (1993) reports that Brazil had become the third largest user of agricultural pesticides by 1970, but conducted very little monitoring of their dispersal through the environment. One major study was conducted over the period 1976 and 1984 in the Parana´ River basin, and showed that over 91% of in situ samples contained at least one pesticide or residue.
3.
Industrial Emissions of Chemicals
Water is used by industry in a large variety of ways, many of which degrade its quality. The total withdrawal from surface water or groundwater by industry is usually much greater than the amount of water that is actually consumed, so the industrial eZuent has potential for beneficial use if its quality is not impaired. The residual is commonly returned either by direct injection into a water body, disposal to a sewer, or disposal after treatment by an on‐site wastewater treatment plant. In some cases, industrial eZuent is recycled or reused directly on‐site, either before or after treatment. When industrial eZuent is discharged directly into a water body without adequate treatment, a number of toxic chemicals can enter the water cycle. If the water is contaminated with heavy metals, they can attach to suspended particles and contaminate lake or stream sediment. Injected water that has a high level of organic matter can cause a rapid growth of algae, bacteria, and slime, followed by a depletion of the level of oxygen in the water. Whenever polluted eZuent is injected into a water body, it contaminates a much larger volume of water as it mixes with the surroundings. Industries and water quality regulators in some places still rely on dilution to disperse contaminants by mixing with unpolluted water until the levels of contaminant drop below harmful levels. This short‐sighted policy is problematic for many reasons. Dilute levels of contaminant can bioaccumulate in the food chain, reaching toxic concentrations at higher trophic levels. Unregulated discharge by multiple sources can rapidly pollute large bodies of water to harmful levels, as well as cause oxygen depletion through organic matter additions. In international river basins, intentional discharge or industrial accidents and spillages by one country can cause severe damage to downstream users in another country. In 1986, a fire in a pesticide‐manufacturing plant in Basel Switzerland resulted in firefighters washing 30 t of pesticides and dyes into the Rhine River. Damage to the river ecosystem was extensive, and it
THE EMERGING GLOBAL WATER CRISIS
21
interrupted water use of the river by the four countries adjoining it for months (Capel et al., 1988). Many municipalities around the globe have their drinking water supply impacted by industrial pollution, raising water treatment costs for the water supply utility. If irregular eZuent discharges produce highly variable water quality, the water treatment plant may not be able to cope adequately with the contaminants. Industrial pollution may also indirectly aVect water supplies by leaching of chemicals from solid wastes and by atmospheric deposition. A study of 15 Japanese cities, for example, showed that 30% of all groundwater supplies were contaminated by chlorinated solvents from industry. In some cases, the solvents from spills had traveled as far as 10 km from the source of pollution (UNEP, 1996). Many streams, rivers, and lakes in Europe are more acidic than they would naturally be, due to acidic deposition. In Scandinavia, for example, hundreds of lakes still suVer from acidification, and will take a long time to recover (European Environment Agency, 1997). Exposure to heavy metals has been linked with developmental retardation, various cancers, and kidney damage. Exposure to high levels of mercury, gold, and lead has also been associated with the development of autoimmune disease, in which the immune system starts to attack its own cells, mistaking them for foreign invaders (Glover‐Kerkvliet, 1995). Several studies have shown that exposure to lead can significantly reduce the IQ of children (Goyer, 1996). In some countries, heavy metal emissions are falling as a result of the removal of lead from petrol, improvements in wastewater treatment and incinerators, and improved industrial technologies. Significant further improvements could be achieved if the available technologies were more widely applied (European Environment Agency, 1998).
4. Natural Toxics One of the greatest water quality challenges to manage is the accumulation of toxic chemicals that are dissolved out of native soil or rock material. The element posing the greatest threat to humans is arsenic. Arsenic is a natural part of the earth’s crust in some parts of the globe, and may be found in groundwater underneath arsenic‐rich rocks. Long‐term exposure to arsenic via drinking water causes cancer of the skin, lungs, urinary bladder, and kidney, as well as other skin problems such as pigmentation changes and thickening (WHO, 1993). A public health limit of 10 mg liter1 has been established by the World Health Organization and subsequently adopted by a number of countries, based on evidence from chronic exposure in arsenic‐rich areas of the world. Although concentrations of arsenic in rivers are generally low, they can be found at high levels near geothermal activity or
22
W. A. JURY AND H. J. VAUX
through discharge from arsenic‐rich groundwater (Smedley and Kinniburgh, 2005). A number of large aquifers in various parts of the world with arsenic levels at concentrations above 50 mg liter1 have been linked to health problems. Regions suVering from high arsenic levels include parts of Argentina, Bangladesh, Chile, northern China, Hungary, the West Bengal region of India, Mexico, Romania, Taiwan, and parts of the Southwest United States. The problem is most severe in Bangladesh, where over 25% of the wells tested have revealed levels of arsenic above 50 mg liter1. It has been estimated that up to 77 million inhabitants of Bangladesh are at risk from drinking arsenic‐contaminated water (Smith et al., 2000). Ingestion of excess fluoride in drinking‐water can cause fluorosis, which aVects the teeth and bones. Moderate exposure will cause dental complications, but long‐term ingestion of large amounts can lead to potentially severe skeletal problems (WHO, 1993). Since some fluoride compounds in the earth’s upper crust are soluble in water, fluoride is found in both surface waters and groundwater. In surface water, fluoride concentrations are usually low, but levels in groundwater can rise to more than 35 mg liter1 depending on aquifer conditions. Fluorosis is endemic in at least 25 countries across the globe. The total number of people aVected is not known, but could number in the tens of million or higher. All states of India except in the northeast have reported cases of fluorosis, and 25–30 million people are estimated to be exposed to high fluoride intake, of which half a million suVer from skeletal fluorosis (UNICEF, 1999). In China, 300 million people are living in endemic areas of fluorosis, of whom 40 million have dental fluorosis and 3 million suVer from skeletal changes (Li and Cao, 1994). Selenium is another natural constituent of certain rock and soil material that can be dissolved by percolating water. In 1983, incidents of mortality, congenital deformities, and reproductive failures in aquatic birds were discovered at Kesterson Reservoir, a US Department of the Interior (DOI) National Wildlife Refuge in western San Joaquin Valley, California. The cause of these adverse biological eVects was determined to be poisoning by selenium carried by irrigation drainage into areas used by wildlife (Ohlendorf et al., 1988). In the western United States, about 160,000 square miles of land, which includes about 4100 square miles of land under irrigation, has been identified as being susceptible to selenium leaching (Seiler et al., 1999).
E. WATER FOR ECOSYSTEMS A river needs to flow over its entire length to support the riparian ecosystems that depend on it. How much flow is required for ecosystem health is a matter of debate, and undoubtedly depends on the local conditions.
THE EMERGING GLOBAL WATER CRISIS
23
Falkenmark and Rockstrom (2004) indicate that roughly 30% of the base flow of a river should remain untouched. This amounts to 3780 km3 of the 12,500 km3 available supply. Postel et al. (1996) arrived at a number of 2350 km3 year1 required for instream uses by a diVerent method. Human appropriation and use of water has caused enormous damage to ecosystems during the last half century, through activities such as draining of wetlands, damming of rivers, and pollution of lakes and streams. Participants in the Millennium Ecosystem Assessment concluded that humans have changed terrestrial and aquatic ecosystems more rapidly and extensively over the past 50 years than in any comparable period of time in human history, largely to meet rapidly growing demands for food, freshwater, timber, fiber, and fuel. Their analysis showed that 15 of the 24 ecosystem services examined during the Millennium Ecosystem Assessment are being degraded or used unsustainably (Millennium Ecosystem Assessment, 2005). The degradation or loss of ecosystem function has huge economic implications, since freshwater ecosystems provide several trillion dollars in annual services (Postel, 1997). 1.
Stream Flow Modification
A major analysis of international water resources was made recently by the Global International Waters Assessment (GIWA) project of the UN Environmental Program. Nineteen GIWA regional teams identified stream flow modification as having severe impacts, particularly in sub‐Saharan Africa, North Africa, Northeast Asia, Central America and Europe, and Central Asia. On a global scale, the most widespread and adverse consequences result from the modification of stream flow by dams, reservoirs, and river diversions, as well as by land‐use changes in the catchment area. Downstream ecosystems and riparian communities are severely impacted by changes to the flow regime of international rivers (UNEP, 2006). Today, dams and reservoirs intercept about 35% of river flows as they head toward the sea—up from 5% in 1950 (Postel, 2005). Many rivers are so overused that they run dry before reaching the sea for extended periods, causing severe damage to fisheries and coastal zones. Rivers falling into this category include the Huang He (Yellow River) in China, the Indus and Ganges in South Asia, the Nile in Africa, the Syr Darya in Central Asia, the Chao Phraya in Thailand, and the Colorado in the western United States (Postel, 1999). 2.
Wetlands Loss
Wetlands provide a wealth of valuable ecosystem services and support diverse habitat. Some estimates show that half of the world’s wetlands have been destroyed by humans in the last 100 years. Much of this loss occurred in
24
W. A. JURY AND H. J. VAUX
northern countries during the first 50 years of the century, but since the 1950s increasing pressure for conversion to alternative land use has been put on tropical and subtropical wetlands. Examples of the impacts of the loss and degradation of wetlands include: impaired or reduced water supply, loss of water flow regulation and flood control, saline intrusion into groundwater and surface water, increased erosion, reduced sediment and nutrient retention, and loss of capacity for pollution removal (Davies and Claridge, 1993). Land conversion for agricultural production is the principal cause of wetlands destruction worldwide. Between 56% and 65% of the available wetland had been drained for intensive agriculture in Europe and North America by 1985. The figures for tropical and subtropical regions were 27% for Asia, 6% for South America, 2% for Africa, and a total of 26% worldwide. Future predictions show the pressure to drain land for agriculture intensifying in these regions (Moser et al., 1996).
F. GROUNDWATER OVERDRAFT Groundwater residing below the near subsurface was an unexploited commodity for most of human history, until technological development allowed it to be extracted from great depths. But with that innovation, it has become a reliable source of supply for a variety of municipal, industrial, and agricultural needs. Annual groundwater use for the world as a whole has been estimated at 750–800 km3 (Shah et al., 2000b), a relatively small fraction of the total use (Table II). However, most of the world’s cities and towns depend on groundwater to supply at least part of their needs. For example, approximately half the population of the United States relies on groundwater for drinking, and more than 90% of rural residents obtain their water from groundwater through wells or springs (US EPA, 2006). Groundwater also provides a significant part of the industrial water demand in most countries. In some of the poorest and most populous regions of the world, particularly in South Asia, groundwater has become critical for feeding the population. In India, for example, some 60% of the irrigated food grain production now depends on irrigation from groundwater wells. Groundwater overdrafting occurs when the rate at which water is extracted from an aquifer exceeds the rate at which the aquifer is replenished or recharged. Chronic overdraft causes persistent lowering of water tables, which leads ultimately to economic exhaustion of the aquifer. Some aquifers have no significant recharge at all and in these instances the water is available on a one‐time basis much like stock resources in a mine. The extent of persistent groundwater overdrafting on a global basis is diYcult to estimate because of limited data and extensive variability in groundwater levels over time and space. Postel (1999) calculated that as much as 163 km3 year1 of
THE EMERGING GLOBAL WATER CRISIS
25
persistent overdraft is occurring globally, about 80% of which was occurring in India and China and most of the rest in the Americas and Africa. The implications of overdraft are not always clear. Intermittent overdraft, where periods of overdraft alternate with periods of net recharge, is generally an acceptable practice. Intermittent overdrafting is a common way of coping with drought, for example. By contrast, persistent overdrafting is more problematic and has serious long‐term impacts. There is great concern over the fact that overdraft is not only unsustainable but tends to be self‐ terminating when water table depths fall below the level from which it is economical to pump. In such cases, accustomed levels of water supply will have to be reduced unless alternative sources of water can be found. In many such instances, alternative sources of supply are not available (Vaux, 2007). Since much of the water that is thought to be overdrafted is primarily used for irrigation, it can be argued that close to 500 million people are being fed with food grown by a water supply that could disappear in the future. One analysis concluded that as much as 25% of India’s grain harvest could be in jeopardy (Seckler et al., 1999). An analysis of persistent overdrafting cautions that the uncertainties in making estimates of the extent of this practice are too great to support a quantitative analysis from the existing database (Moench et al., 2003). These authors argue that the lack of comprehensive monitoring as well as the short time series available where adequate monitoring has occurred means that conclusions about the presence or absence of overdraft are not based on solid empirical evidence in many instances. In addition, these authors question the connection between groundwater overdraft and food security, noting that trade may produce alternative sources of food and citing a number of other factors that may tend to disconnect groundwater overdraft from the production of food. A final concern about groundwater and its management relates to the role of groundwater in providing environmental amenities and services. Glennon (2002) notes that estimates of safe yield (the yield that just equals recharge) fail to account for environmental uses of groundwater and the interconnectedness of groundwater and surface water in many instances. The fact that groundwater is treated as a common pool resource in many parts of the world means that environmental uses tend to be ignored and little attention is given to overdrafting. The lack of adequate governance of groundwater resources is frequently cited as a reason for suspecting that such resources are overexploited (Glennon, 2002). The brief overview provided in this section has identified a number of water‐related problems in various regions of the world. These problems threaten to diminish the available supply and pose severe threats to the integrity of aquatic ecosystems throughout the world. Even if water demands were to remain static these problems suggest that it may be diYcult to
26
W. A. JURY AND H. J. VAUX
continue meeting them as we have in the past. However, even more serious challenges lie in the years ahead as the world adds billions more in population which will augment water demand even as available supplies are shrinking. In Section III, projected changes in population and the impacts of these changes on water resources and their quality are discussed.
III.
POPULATION TRENDS AND WATER STRESSES
A single snapshot in time of water resource supply, quality, and consumption for a region does not provide suYcient information to develop a water management strategy. Many factors that influence water resources are time dependent, so it is necessary to characterize temporal trends as well as to evaluate the water balance at any given time. Water resource availability and quality are already serious problems in certain parts of the world today, and other regions that are currently not under stress could face shortages in the next few decades as demands increase. The first step in making a credible assessment of future water supplies and demands is to forecast population growth, which can be done relatively accurately in the short run, but which depends on assumptions that make estimates progressively more uncertain for later times. Lutz et al. (2001) estimated world population growth over the next century together with the 80% confidence limits, and found that the mean population would level oV by the second half of the century after reaching a maximum of a little more than 9 billion. However, by mid century, the 80% confidence limits of the mean projection ranged over nearly three billion. We will use the US Census international database (US Census, 2006) for population estimates in this chapter unless otherwise noted, and confine our discussion to the first half of the twenty‐first century. Over that period the US Census prediction agrees reasonably well with the mean curve estimated by Lutz et al. (2001). According to the US Census estimate, global population will rise from 6.1 billion to 7.9 billion between 2000 and 2025, and to 9.2 billion by 2050.
A. WATER‐SHORT AND WATER‐STRESSED COUNTRIES The FSI (Section II.B) classified 29 countries with a population of 460 million as being in either a water‐stressed or a water‐scarce situation in 1995 (Table III). By 2025, this number rises to 47 countries and 2.8 billion people (Table VI), including 19 countries with annual assets that fall below the water barrier of 500 m3 year1 per person (Table VII). An additional factor that needs to be taken into account in assessing the eVect of population growth on water supply is where the growth is occurring
THE EMERGING GLOBAL WATER CRISIS
27
Table VI Population and Numbers of Countries Predicted to Experience Water Stress or Scarcity by 2025 According to the FSI Category Water scarce Water stressed Water scarce or stressed
Countries
Population (millions)
29 (17) 19 47
802 (218) 2027 2829
Number in parentheses indicates countries below water barrier of PWR < 500 m3 year1 per person. Data adapted from Population Reports (1998).
Table VII Countries Predicted to Experience Water Stress or Scarcity by 2025 According to the FSI Below water barrier PWR < 500 m3 year1 Algeria Bahrain Barbados Burundi Cape Verde Israel Jordan Kuwait Libya Malta Oman Qatar Rwanda Saudi Arabia Singapore Tunisia Yemen
Water stressed PWR ¼ 500–1000 m3 year1
Water scarce PWR ¼ 1000–1700 m3 year1
Comoros Cyprus Egypt Ethiopia Haiti Iran Kenya Malawi Morocco Somalia South Africa UAE
Belgium Burkina Faso Eritrea Ghana India Lebanon Lesotho Mauritius Niger Nigeria Peru Poland South Korea Syria Tanzania Togo Uganda United Kingdom Zimbabwe
relative to available local water resources. Figure 4 shows the historical and projected population growth of the world and various subdivisions. Several features are worth noting. First, virtually all of the projected population growth between the present and 2050 is expected to occur in developing countries, a number of which are already experiencing water shortages. Second, the population of water‐short countries (PWR < 1700 m3 year1) is a small fraction (7.5%) of the world population in 1995, but a significant fraction (36%) by 2025, largely due to India falling below the threshold PWR.
28
W. A. JURY AND H. J. VAUX 9 Population in billions
8 7 6
World Developed countries Less developed countries China and India Water-stressed or Water-scarce countries
5 4 3 2 1 0 1950
1970
1990
2010
2030
2050
Year Figure 4 Historical and projected population growth of population subgroups. Water‐ stressed or ‐scarce countries are those whose per capita resources are below 1700 m3 year1. (Data from US Census, 2006).
China and India together constitute over 35% of the world’s population, so their water use and food requirements will have a dominant influence on global trends. Figure 5 shows the population and per capita water resources for these two countries between 1950 and 2050, based on the assumption that total water resources will remain constant in the future. By 2015, India will drop below the FSI threshold of 1700 m3 year1, reaching a low of 1300 m3 year1 by mid century. China approaches but remains above the threshold as its population levels oV. However, each country has wet and dry regions, so the national average can be misleading. China has approximately half of its population in the north, where only 20% of its water resources are found. Thus, if North China was viewed as a separate country, its per capita resources would be only 40% of those shown in Fig. 5, and would dip to a low value of 770 m3 year1, falling well below the FSI threshold. Although the various indices for expressing water scarcity have some common elements, they do not produce the same classification when applied to the countries of the world. Figure 6 shows a comparison of three indices on a group of 38 countries considered water short in 2025 by some criterion. All of the 38 countries are classified as water stressed by the FSI, but only 22 of them by the UN criterion. Even fewer countries (14) are classified in the most stressed water group 1 by the method used by Seckler et al. (1998). Reasons for the variances in classification are due to the diVerent criteria used. For example, Burundi has only 267 m3 year1 of per capita annual water resources, making it extremely water scarce by the FSI, but is using only 9% of it in annual withdrawals, which causes it to be rated as unstressed by the other criteria.
THE EMERGING GLOBAL WATER CRISIS 5000
China India
1.6
4000
1.2
3000
0.8
2000 Falkenmark threshold
0.4 0.0 1950
1970
1990
2010
1000
2030
Water resources (km3 year −1 person−1)
Population in billions
2.0
29
0 2050
Year Figure 5 Population growth and per capita water resources of China (solid curve) and India (dashed curve). Shaded region marks the zone with water resources below the Falkenmark stress indicator threshold of 1700 m3 year1.
Falkenmark threshold
Annual use as % of AWR
10,000
1000
Group 1 Group 2 Group 3 Group 4 Group 5
100
10
1 10
UN threshold
100 1000 Annual water resource (m3 year −1 person−1)
10,000
Figure 6 Classification of the degree of water scarcity of a country in 2025 according to various benchmarks of water use and water availability. The Falkenmark threshold is 1700 m3 year1 per person and the UN threshold is 40% annual use of the available water resource.
Table VIII summarizes 29 countries that consume more than 20% of their annual water resources in a given year. Many of the countries listed in Table VII as being water scarce or stressed according to the FSI do not appear in Table VIII, usually because irrigation is not required to grow crops and hence less water per capita is needed for food production. This shows that the indices by themselves cannot adequately classify whether a country is likely to experience hardship in the next few decades due to inadequate water
30
W. A. JURY AND H. J. VAUX
Table VIII Groupings of Countries According to the Fraction of Annual Water Resources Consumed in a Given Year W/AWR > 1.0 Bahrain Jordan Kuwait Libya Oman Qatar Saudi Arabia Tunisia UAE
0.5 < W/AWR < 1.0
0.3 < W/AWR < 0.5
0.2 < W/AWR < 0.3
Belgium Egypt Iraq Israel Pakistan Yemen
Afghanistan Algeria Cyprus Iran South Korea Morocco Singapore South Africa Syria
India Japan Lebanon Sri Lanka Swaziland
resources. For example, a country with a low PWR that can aVord to import food may have a more than adequate supply of water for all other uses. In contrast, poor countries with inadequate PWR will have diYculty finding the capital required to import food, and may face starvation in the future if they do not have the means to grow what they need. There is clearly no single strategy adequate for dealing with future water challenges on a global level. Optimization of the conflicting requirements for water by the agricultural, industrial, urban, and environmental sectors will require a systems perspective driven by both short‐term needs and a long‐term perspective.
B. URBANIZATION TRENDS According to the 2004 assessment of the United Nations, the projected change in world population between 1995 and 2030 will be 2.51 billion, while at the same time the increase in urban population will be 2.44 billion. While much of the change is due to urban migration, it is equivalent for planning purposes to assume that virtually all of the increased population in the next half century will live in cities. This trend will significantly raise the proportion of the population in urban areas of developing countries (Fig. 7). Cities are also becoming much larger, with a number reaching mega‐city size, denoting an urban area of more than 10 million (Table IX). Jenerette and Larson (2006) estimate that the number of cities with more than 5 million residents is expected to increase globally from 46 to 61 between 2015 and 2030, with disproportionate increases in Asia and Africa. Large cities place special demands on water resources because of the high population density, and the challenges presented in maintaining adequate sanitation. These authors analyzed the resource requirements of the 524 urban
THE EMERGING GLOBAL WATER CRISIS
31
100 90 Urban % of population
80 70 60 50 40 30
Africa Asia Europe Latin America North America World
20 10 0 1950
1970
1990
2010
2030
Year Figure 7 Percentage of the population of various population sectors residing in cities as a function of time.
Table IX Number of Mega Cities in the World at DiVerent Times Year Population > 10 million Population > 15 million Population > 20 million 1985 2000 2015
9 18 22
2 5 11
1 1 4
regions with populations greater than 750,000 as of 2000 using an ecological footprint (EF) analysis. The EF calculates the land area required to provide sustainable services to the urban unit. Table X shows how the EF required to provide water resources has grown over time, and that the largest of the mega cities in water‐short regions have enormous EF. Saudi Arabia, for example, had five cities with EF ranging from 1.4 to 2.4 million km2. The study also suggested that cities with a high EF are also especially sensitive to climate change. Providing adequate water for urban uses in metropolitan areas with huge populations in the future will be extraordinarily challenging. Many regions now provide services only by extracting groundwater at rates greatly in excess of recharge, which not only jeopardizes future availability but also causes ancillary problems such as land subsidence and increased vulnerability to aquifer contamination. Other metropolitan areas, while currently self‐suYcient, have no obvious sources of supplemental supply to support population growth.
32
W. A. JURY AND H. J. VAUX
Table X Mean Ecological Footprint for Water Resources Required to Provide the World’s Cities of Greater Than 750,000 Population (Jenerette and Larson, 2006) Baseline scenario (year) 1950 2000 2015
Mean footprint area (km2)
Total urban water footprint (km2)
29,937 (6.23) 35,397 (5.44) 38,400 (5.18)
15,686,988 18,548,028 20,121,600
CoeYcient of variation of estimate is given in parentheses of column 2.
C. INDUSTRIAL AND MUNICIPAL WATER DEMANDS Municipal water demands consist of the water withdrawals made by the populations of urban areas for domestic use plus withdrawals for industrial, public, and commercial uses. In many cities, a considerable volume of water is applied to vegetable gardens and residential landscapes. The volume of public water use depends on population, the level of services and utilities, the availability of conveyance and supply infrastructure, and climatic conditions (Shiklomanov, 2000). In industrially developed countries of Europe and North America, per capita domestic water withdrawal is of the order of 500–800 liter day1 (Shiklomanov, 2000). In contrast, it is only about 50–100 liter day1 in developing agricultural countries of Asia, Africa, and Latin America, and 10–40 liter day1 in regions with insuYcient water resources (Gleick, 1996). Although it is tempting to suggest that economic development is associated with increasing per capita water use, Gleick (2006) has shown that, at least in the United States, per capita water consumption has been falling. Water in industry is employed for cooling, transportation, as a solvent, and as an ingredient of finished products. The dominant user is electric power generation, which requires a great amount of cooling water. Other heavy industrial water users are the chemistry and petroleum chemistry industries, the wood pulp and paper industries, the metallurgy industry, and machine building. The water needed by a given industry depends mainly on whether the water withdrawn is passed once through the system or circulated internally. With a circulating system, the water is cooled, treated, and routed back to the water supply system after use, whereas eZuent from a once‐through system is returned to the source water. In addition, many industries in developed countries find it economical to recycle process water in order to meet prevailing pollution discharge regulations (Shiklomanov, 2000). Although the consumption of water for industrial and public use is considerably less than for agriculture (Table II), it is not insignificant and will become a challenge to manage in certain regions in the future. Figure 8
THE EMERGING GLOBAL WATER CRISIS 1000 Withdrawal (km3 year−1)
800 600
33
Europe Africa Asia South America North America
400 200
Consumption (km3 year−1)
0 100 75 50 25 0 1900
1925
1950 1975 Year
2000
2025
Figure 8 Withdrawal and consumption of water for industrial and domestic purposes as a function of time for various continents. Data after 1995 is projected (Data taken from Shiklomanov, 1997).
shows the withdrawal and consumption of water as a function of time by the municipal and industrial sectors in the five major continents (Shiklomanov, 1997). The projected increases for the early part of the twenty‐first century reflect both the relative population stability in the developed countries and the explosive growth in urbanization and industrialization projected for Asia. Also notable is the huge diVerence between consumption and withdrawal, which can be misleading if the water returned to the surface water or groundwater supply by the municipal and industrial sectors is in a degraded state from pollution. The projected increases in global withdrawal and consumption of water by various sectors between 1995 and 2025 are given in Table XI. Note again that the withdrawal of water by the municipal and industrial sectors during this time frame is comparable to the withdrawal for irrigation by agriculture, but actual consumption is only about 20%. Reservoir losses from evaporation are also substantial. When there is competition for water between sectors, agriculture could lose significant food production capability in a region. For example, assuming 1200 m3 year1 to feed one person, the additional 681 km3 year1 of water withdrawn for municipal and industrial use between 1995 and 2025 represents the water needed to grow food for nearly 570 million people.
34
W. A. JURY AND H. J. VAUX
Table XI Projected Increase in Water Withdrawal and Consumption in km3 year1 by Various Sectors Between 1995 and 2025 (after Shiklomanov, 2000)
Sector or use category Agriculture Municipal Industry Reservoir losses Total
Withdrawal increase (km3 year1)
Withdrawal increase (% of total)
Consumption increase (km3 year1)
Consumption increase (% of total)
685 263 418 81 1447
47.3 18.2 28.9 5.6 100.0
499 24.3 86.4 81 690
72.3 3.5 12.5 11.7 100.0
D. TRANSBOUNDARY ISSUES Water does not respect international boundaries. Today, some 146 countries of the world share a river with at least one other nation. There are 261 international river basins whose drainage areas span more than one country, covering in total some 45% of the land area of the planet (Wolf et al., 1999). Table XII shows the number of countries sharing various international river basins, led by the Danube which flows through 17 nations in Europe. A number of these shared basins operate without treaties governing their use. The absence of treaties or operating agreements frequently leads to overextraction, conflicting management plans, and border tensions. In the most extreme cases, these conflicts have reached the level of hostility though not violence. Table XIII lists five of the more serious river basin conflicts around the world. Each is briefly discussed in the following sections.
1.
Jordan River
The Jordan River drains part of Israel, Jordan, Lebanon, and Syria. It is a small water body, extending only 93 km from its source waters in Lebanon to its final discharge into the Dead Sea. Each of the three streams forming the river’s headwaters was originally in a diVerent country, but since the end of the 1969 war Israel has controlled all of the stream areas. The upper reach of the Jordan drains into Lake Tiberias (also called Sea of Galilee or Lake Kinneret), which at 21 km in length and 13 km in width is Israel’s largest freshwater body. The lake’s outflow moves to the Dead Sea along the Jordan River valley. The Jordan River has two principal tributaries, the Yarmouk originating in Syria and the Zarqa which flows out of Jordan. Two major diversion works extract water from the river, Israel’s National Water Carrier and
THE EMERGING GLOBAL WATER CRISIS
35
Table XII Number of Countries Sharing a River Basin (Wolf et al., 1999) Countries 17 11 10 9 8 6 5 4 3 2
Basins
Name of basins
1 2 1 2 2 8 3 17 49 176
Danube Congo, Niger Nile Rhine and Zambezi Amazon, Lake Chad – – – – –
Table XIII Examples of Serious International River Basin Disputes and the Countries Involved River Jordan Tigris‐Euphrates Nile Indus Ganges
Countries involved Israel, Jordan, Lebanon, Syria Iraq, Syria, Turkey Egypt, Ethiopia, Sudan India, Pakistan India, Bangladesh
Jordan’s East Ghor Canal. The National Water Carrier transports water from Lake Tiberias through a network of pipes to Tel Aviv and the Negev, while the East Ghor Canal diverts water from the Yarmouk River to agricultural areas in the Jordan Valley (McCaVerty, 1998). The basin has no treaty governing its use, although a US‐brokered agreement to allocate the Jordan and Yarmouk’s flows (known as the Johnston Plan) was reached in the 1950s by technical representatives from all countries but was never ratified by the governments concerned. That agreement would have given Jordan 19% of the flow of the upper Jordan River and 75% of the Yarmouk. Since the Six Days’ War, however, due to its downstream position on the Jordan River and its weak strategic standing on the Yarmouk, Jordan has been greatly disadvantaged in its water use opportunities. Israel has virtually monopolized the waters of the Upper Jordan since the late 1960s, with only a highly polluted residual of wastewater flowing to the Dead Sea. Jordan’s use of the Yarmouk has also been restricted, both by Israeli withdrawals to restock Lake Tiberias and by Syria’s increasing upstream use. As a consequence, Jordan is only using about 25% of the Yarmouk’s flow, which is only
36
W. A. JURY AND H. J. VAUX
one third of the allocation it was granted in the Johnston plan (Libiszewsk, 1995). Despite adopting a number of conservation and water eYciency practices, Jordanian farmers in the lower valley are struggling to survive on the supply of water available (Ayadi, 2006). The Jordan River Basin is complicated further by the fact that both the West Bank and Gaza Strip of Palestine lie within the Basin and its service area. Per capita water availability in Palestine is a fraction of what is available to Jordan and an even smaller fraction of what is available to Israel. This makes the problem of allocating the very scarce waters of the basin among diVerent users and claimants even more diYcult (National Research Council, 1999a). One recent study shows that water could be allocated according to its economic value, thereby minimizing the costs and disruptions now attributable to a lack of water. This study also suggests that desalination is not necessarily needed on a broad scale to address the region’s water problems (Fisher et al., 2005).
2.
Tigris‐Euphrates Basin
The Tigris and Euphrates rivers are often treated as one basin because they merge before reaching their final destination. Both rivers originate in the mountains of Turkey and flow through or at the boundary of Syria before entering Iraq. Iraq is heavily dependent on the two rivers for its water supply, which provides the only source for much of its population. Turkey is currently constructing a large water project on the Euphrates in southeastern Anatolia, known as the Greater Anatolian Project (GAP). The GAP will eventually consist of 21 dams to be used for hydroelectric power production and the irrigation of over a million hectares of agricultural land. When complete, the dam system could cause Syria to lose up to 40% and Iraq up to 90% of their water from the Euphrates (McCaVerty, 1998). The three riparian nations have had some success in addressing their diVerences over the project and other water issues peacefully. Bilateral agreements exist between Turkey and Iraq and between Syria and Iraq on certain issues in their water relations. However, the GAP poses a significant environmental threat to Turkey’s downstream neighbors in the future.
3.
Nile River
The Nile is the longest river in the world, draining an area over 3 million km2 in 10 countries. About 85% of the Nile’s flow originates in Ethiopia as the Blue Nile, with the remainder coming from the White Nile which begins in Tanzania. Until recently, use of the Nile was dominated by Egypt, which is
THE EMERGING GLOBAL WATER CRISIS
37
dependent on its flow for virtually all of its water. Egypt has a bilateral treaty with Sudan, in which Egypt is entitled to 55.5 km3 of annual flow, and Sudan an additional 18.5 km3. The other countries with access to the Nile do not have an agreement governing its water use. Egypt currently is using all of its allocation, and its population is rising rapidly. It has plans to expand its irrigation by at least 1 million ha over the next 20 years, which at current use rates would require an additional 8 km3 of water (Postel, 1999). Ethiopia would like to build dams for hydroelectric power and to provide water to irrigate substantial land in its country, activities which have the potential to divert as much as 7.2 km3 year1 of flow from the Nile. The Nile is virtually fully appropriated and very little flow reaches the Mediterranean Sea, which has drastically altered the aquatic habitat of the delta. The situation is further complicated by the fact that the other nations along the Nile have never recognized the Egypt–Sudan water agreement because they were not involved in its negotiation. Although tensions over water have nearly led to armed intervention in the past, more recently the countries involved have developed a mechanism for regional cooperation. The Nile Basin Initiative, launched in February 1999, is a regional partnership within which countries of the Nile Basin have united in common pursuit of the long‐term development and management of Nile waters. The Initiative partnership is developing consensus on a basin‐wide framework and is guided by the countries’ shared vision to achieve sustainable socioeconomic development through the equitable utilization of, and benefit from, the common Nile Basin water resources (Foulds, 2002). The early results of these eVorts seem promising but many diYcult issues remain to be addressed.
4.
Indus River
The Indus River, which originates in Tibet and flows 2900 km through India and Pakistan, has been the subject of controversy since the India– Pakistan division in 1947. The partition left part of the basin in each country, with the majority of the canal system and irrigated lands residing in Pakistan. Early conflicts over water were frequent, and even led to India temporarily stopping the supply of water to the canals in Pakistan in 1948. The World Bank was successful in getting both countries to adopt a comprehensive water agreement known as the Indus Waters Treaty in 1960, in part because it sponsored projects that would increase the water allocation to both countries. With minor exceptions, the treaty gives India exclusive use of all of the waters of the Eastern Rivers and their tributaries before the point where the rivers enter Pakistan. Similarly, Pakistan has exclusive use of the
38
W. A. JURY AND H. J. VAUX
Western Rivers. Pakistan also received one‐time financial compensation for the loss of water from the Eastern Rivers. Although the treaty is not legally binding, it has had the eVect of quelling water disputes in this region for over 40 years (McCaVerty, 1998). However, recent declines in the flow of the Indus have increased stresses, particularly in water‐short Pakistan. 5.
Ganges River
The Ganges originates in the Himalayas and flows through India to Bangladesh, where it joins the Brahmaputra to form the Padma, which empties into the Bay of Bengal. Between 1961 and 1975 India constructed a dam just upstream from the Bangladesh border, in order to divert water to Calcutta. This action left Bangladesh short of irrigation water needed in the dry months, of water needed to prevent siltation and subsequent flooding of the river, and of water needed to prevent seawater intrusion from the Bay of Bengal. Bangladesh subsequently appealed to the general assembly of the United Nations, and the countries eventually were persuaded into adopting a plan known as the 1977 Agreement on Sharing of the Ganges Waters, which allocated flow during the dry season. The vast majority of water disputes involving international basins have been resolved without armed conflict. Researchers at Oregon State University have compiled a dataset of every reported interaction (conflictive or cooperative) between two or more nations that was driven by water in the last half century. The findings show that cooperation—not conflict—is the norm. In the last 50 years, only 37 international water disputes have involved violence, and 30 of those occurred between Israel and one of its neighbors. Outside of the Middle East, researchers found only 5 violent events while 157 treaties were negotiated and signed. They also found that over 70% of the 1735 water‐related events recorded between nations were devoid of any conflict (Wolf, 1998).
E. PROJECTED WATER DEFICIT UNDER BUSINESS AS USUAL PRACTICES There are already serious water deficits in certain parts of the planet today, and continuation of current policies and trends will create many more in the future. Since the increase in population between 1995 and 2025 is expected to be 2.2 billion, per capita consumption patterns may be extrapolated to project the water that would be used in that year if current patterns of use and consumption are maintained in the face of significant population and economic growth. Assuming an average global water use for food production of 1200 m3 year1 per person (Rockstrom et al., 1999), the population increase implies that an additional 2740 km3 year1 of water would be required to grow the food
THE EMERGING GLOBAL WATER CRISIS
39
needed. Adding the 762 km3 year1 increase projected for domestic use, industrial use, and reservoir losses from Table XI, we obtain about 3500 km3 year1 of new water required to provide for the population of 2025. Beyond merely feeding the increased population, Falkenmark and Rockstrom (2004) calculate that an additional 2200 km3 year1 of freshwater will be needed to eradicate malnutrition in the 2050 population. If we assume that half of this can be obtained by 2025 and the remainder by 2050, then an additional 4600 km3 year1 of freshwater will be required by 2025 to accommodate human needs. Table XIV summarizes these estimates. If the additional 1.2 billion population increase between 2025 and 2050 proves correct, another 1560 km3 year1 would be utilized by 2050 for food, an additional 430 km3 year1 needed for cities and industry, and the remaining 1100 km3 year1 required to alleviate malnutrition for a total increase of 3090 km3 year1 between 2025 and 2050. This leads to the staggering conclusion that nearly 7700 km3 of additional water would have to be found by 2050 to supplement global supplies at the 1995 level. Not much of this needed water can come from expanded irrigation operations. Falkenmark and Rockstrom (2004) estimate that irrigation water use can be increased by not more than 800 km3 year1 through expansion of agricultural land and improvements in production eYciency. The remaining water must come from other sources, additional rainfed agriculture, or through increased eYciency and conservation eVorts. The numbers cited above merely represent an extrapolation of the business as usual policies and employ average estimates for all segments of the population. This extrapolative ‘‘requirements’’ approach to water planning and forecasting has proved to be notoriously unreliable. To some extent, the quantities of water used in diVerent sectors are a matter of choice. And, within boundaries, capital and labor can be substituted for water. Additionally, there are almost always opportunities to improve the productivity of water, many of which result in water savings (National Research Council, 1999a). For example, a reanalysis of the water needed to feed the future population has been made by Rockstrom et al. (2007), and demonstrates that Table XIV Water Requirements in km3 year in the Future Relative to 1995 Under Business as Usual Assumptions with No Changes in Consumption Patterns Time period 1995–2025 2025–2050 1995–2050 a b
Food production
Municipal and industriala
Hunger eradicationb
Total
2740 1560 4300
760 430 1190
1100 1100 2200
4600 3090 7690
Includes reservoir losses. Assumes half alleviated by 2025 and the rest by 2050.
40
W. A. JURY AND H. J. VAUX
considerably less water may be required if additional factors are taken into account. In their new study, they assume that countries producing food on the low end of the yield spectrum will be able to increase yields substantially in the future, and that they will experience proportionately higher benefits from these yield increases because the additional biomass will lower evaporative losses. They also analyze separately the vegetative and animal portions of the diet and distinguish between irrigated feedland and grazing contributions. Their analysis concluded that 1910 km3 year1 additional water equivalent of needed food would have to be provided by cultivated rainfed land by 2025. Although this number is far below the figure obtained from simple extrapolation of current consumption rates and eYciencies, it still represents a huge gap that is likely to be filled only by impinging on natural ecosystems and their needed resources.
F.
THREATS TO ECOSYSTEM HEALTH
Perhaps the greatest threat posed by projected freshwater scarcity in the future is to terrestrial ecosystems. If business as usual policies are continued into the future, more water will be diverted from rivers, more wetlands will be drained, more forests will be felled for additional cropland, more agricultural pollution will stress aquatic organisms, and additional dams will be constructed. As of 1995, humans appropriated 54% of the freshwater in lakes, lagoons, rivers, and streams (Postel et al., 1996). By 2025, that value could reach 70% (Postel, 1998), which would require utilizing the entire flow of rivers in many regions. Habitat destruction, water diversions, and pollution are contributing to sharp declines in freshwater biodiversity. Globally, the world has lost half of its wetlands, with most of the destruction occurring in the past half century. Destruction of habitat is the largest cause of biodiversity loss in almost every ecosystem, but biologists have found that most of the plant and animal extinctions have been those species dependent on freshwater and related habitats. One‐fifth of all freshwater fish are threatened or have recently become extinct. On continents where studies have been done, more than half of amphibians are in decline, and more than 1000 bird species are threatened (Hinrichson, 2003). The competition between people and wildlife for water is intensifying in many of the most biodiverse regions of the world. Of the 35 biodiversity hot spots designated by Conservation International (C2006), 10 are located in water‐short regions. These regions—including Mexico, Central America, the Caribbean, the western United States, the Mediterranean Basin, southern Africa, and southwestern China—house an extremely large number of threatened species.
THE EMERGING GLOBAL WATER CRISIS
41
The services that freshwater ecosystems provide to humans such as fisheries, flood protection, recreation, and wildlife are estimated to be worth trillions of US dollars annually (Constanza et al., 1997; Postel and Carpenter, 1997). A global assessment of the status of freshwater ecosystems (Revenga et al., 2000) showed that their capacity to provide the full range of such goods and services appears to be drastically degraded. Many freshwater species are facing rapid population decline or extinction, and yields from many wild fisheries have dwindled as a result of flow regulation, habitat degradation, and pollution. Much of the damage is due to inadequate flow in rivers as a result of human diversion and consumption. The environmental water requirement (EWR) required to maintain riparian ecosystem health has been estimated to range from 20% to 50% of the mean annual river flow in a basin, depending on local climate and conditions. Even at estimated modest levels of EWR, parts of the world are already or soon will be classified as environmentally water scarce or stressed. The total population living in basins, where modest EWR levels are already in conflict with current water use, is over 1.4 billion and this number is growing (Smakhtin et al., 2004). Over the past 50 years, humans have changed ecosystems more rapidly and extensively than in any comparable period in human history, largely to meet rapidly growing demands for food, freshwater, timber, fiber, and fuel. The changes that have been made to ecosystems have contributed to substantial net gains in human well‐being and economic development, but these gains have been oVset by degradation of many ecosystem services, increased risks of nonlinear changes, and tragic exploitation of some of the world’s poorest peoples. These problems, unless addressed, will substantially diminish the benefits that future generations obtain from ecosystems. Approximately 60% (15 of 24) of the ecosystem services examined during the Millennium Ecosystem Assessment are being degraded or used unsustainably, including freshwater, capture fisheries, air and water purification, and the regulation of regional and local climate, natural hazards, and pests (Millennium Ecosystem Assessment, 2005). The full costs of the degradation of these ecosystem services are diYcult to measure, but available evidence demonstrates that they are substantial and growing. Many ecosystem services have been degraded as a consequence of actions taken to increase the supply of other services, such as food, which shift the costs of degradation from one group of people to another or defer costs to future generations. The most important drivers of ecosystem change are habitat alteration, overexploitation, invasive alien species, pollution, and climate change. Evidence is growing that stresses to ecosystems are increasing the likelihood of nonlinear changes that have important consequences for human well‐being. Examples of such changes include disease emergence, abrupt alterations in water quality, the creation of ‘‘dead zones’’ in coastal waters, the collapse of fisheries, explosions in the populations of pest organisms and other
42
W. A. JURY AND H. J. VAUX
organisms, and shifts in regional climate (Millennium Ecosystem Assessment, 2005). Historically, water for the environment has been thought of as the ‘‘supplier of last resort.’’ In developed countries, water to service municipal, industrial, and agricultural uses has been diverted from environmental uses for the most part. The specter of significant and costly environmental change serves as a warning that continued diversions from and degradations of aquatic environments will be far more costly in the future than it has been in the past.
G. THE WILD CARD OF CLIMATE CHANGE Although debate continues about the extent of human influence on climate change, there is no disagreement that the world is getting warmer and will continue to do so for at least the immediate future. Within the context of this chapter, the most relevant question to be addressed is what the eVect of this change is likely to be on global and regional water resources. The only tool available for making projections into the future is climate modeling, which is an advancing but still evolving science. Climate predictions of changes in the global water regime must therefore be regarded as uncertain. Nonetheless, these models are now able to match observations of past climate behavior, and diVerent models involving alternate hypotheses agree on a number of projections relevant to the water regime. The most significant of these are (Frederick, 1997): Climate change simulations predict that globally averaged surface temper-
ature will increase from 1.4 to 5.8 C relative to 1990 by the end of the twenty‐first century. The timing and regional patterns of precipitation will change, and more intense precipitation days are likely. Models used to predict climate change suggest that a 1.5–4.5 C rise in global mean temperature would increase global mean precipitation about 3–15%. Although the regional distribution is uncertain, precipitation is expected to increase in higher latitudes, particularly in winter. Because potential evapotranspiration (PET) increases at higher air temperature, larger PET rates may lead to reduced runoV, even in areas with increased precipitation, implying a possible reduction in renewable water supplies. Annual runoV is likely to increase at high latitudes, while some lower latitude basins may experience large reductions in runoV and increased water shortages. Flooding is likely to occur more frequently in many areas, although the amount of increase for any given climate scenario is uncertain and impacts will vary among basins. Floods may become less frequent in some areas.
THE EMERGING GLOBAL WATER CRISIS
43
Droughts could become more frequent and severe in some areas as a result
of a decrease in total rainfall, more frequent dry spells, and higher evapotranspiration. Seasonal disruptions might occur in the water supplies of mountainous areas if more precipitation arrives as rain rather than snow, and if the length of the snow storage season is reduced. Water quality problems may increase where there is less flow to dilute contaminants introduced from natural and human sources. Agriculture and forestry are likely initially to benefit from carbon dioxide fertilization and increased water use eYciency of some plants at higher atmospheric CO2 concentrations. The optimal climate for crops may change as temperature increases, requiring extensive regional adaptations. Hydrologic impacts could be significant in regions where much of the water supply is dependent on the amount of snow pack and the timing of the spring runoV, such as in the western United States. Increased rainfall rates could impact pollution runoV and flood control. Coastal regions could be subject to increased wind and flood damage if sea levels rise, even if tropical storms do not change in intensity. Significant warming also could have far‐reaching implications for ecosystems. Observed recent changes in climate have already had significant impacts on biodiversity and ecosystems, including causing changes in species distributions, population sizes, the timing of reproduction or migration events, and an increase in the frequency of pest and disease outbreaks. By the end of the century, climate change and its impacts may be the dominant direct driver of biodiversity loss and changes in ecosystem services globally (Millennium Ecosystem Assessment, 2005). Global warming could well have serious adverse societal and ecological impacts by the end of this century, especially if globally averaged temperature increases approach the upper end of the modeling projections. Even in the more conservative scenarios, the models predict temperatures and sea levels that continue to increase well beyond the end of this century, suggesting that assessments that examine only the next 100 years may well underestimate the magnitude of the eventual impacts (National Research Council, 2001).
IV.
DIMENSIONS OF WATER SCARCITY
The issues described in previous sections should make it abundantly clear that water scarcity will intensify in the future, and that current water consumption practices cannot be maintained without causing enormous problems. Every sector of society will have to become more eYcient, and proactive measures will have to be taken to prevent further degradation of
44
W. A. JURY AND H. J. VAUX
remaining supplies. In this section, the possibilities for meeting the growth in water use through conservation, improvements in productivity, economic methods, and technological developments are examined.
A. WATER SAVINGS THROUGH CONSERVATION 1.
Domestic Water Savings
Total global domestic withdrawals are projected to be about 600 km3 year1 by 2025, up from 344 km3 year1 in 1995 (Tables II and XIII). Thus, while conservation improvements may be critically important to specific metropolitan areas and particularly those that currently rely on groundwater overdrafting, the totality of domestic conservation cannot be of major significance on a global scale. Thus, for example, the world’s water reuse capacity is expected to rise by 12.6 km3 year1 between 2005 and 2015 (GWI, 2005), which is insignificant compared to the projected global water need for all purposes. However, reuse will have a significant impact locally. According to Rosegrant et al. (2002), urban households connected to water sources used an average of 43.4 m3 year1 per person, compared to 24.8 m3 year1 for unconnected urban dwellers. Thus, household water demand for a city of 10 million would be 0.25–0.43 km3 year1, and the projected 12.6 km3 year1 increase in water reuse could meet the needs of about 300 million urban dwellers. It should be noted that reuse is currently quite expensive and widespread adoption of reuse technology would result in increases in water prices. Such increases could lead to further reductions in use as consumers seek to economize in the face of higher prices. Educational programs, strengthened water codes, retrofitted plumbing, and installation of dual water supply systems could all have a significant influence in reducing the per capita levels of domestic consumption. Similarly, changes in home landscaping approaches in many developed countries might save up to 50% of annual household use. In short, there are significant opportunities to conserve on domestic water in urban areas. The totality of such conservation in the future may make a significant diVerence in local and regional water supply conditions, but is unlikely to be significant in terms of overall global water use. 2. Industrial Water Savings Global industrial withdrawals are projected to be over 1000 km3 year1 by 2025, up from 752 km3 year1 in 1995 (Tables II and XIII). Of the total for 2025, only 170 km3 year1 will be consumed in industrial processes.
THE EMERGING GLOBAL WATER CRISIS
45
The diVerence between these two numbers represents industrial waste water that is returned to the input stream. Should this water be in a polluted state, it will not only be lost, but will further degrade the source water as well. As developing countries industrialize, they face substantial water losses, should they not require industrial reclamation prior to discharge. At the present time, only developed Western countries have regulations governing industrial water use and disposal. In many instances, strict discharge regulations have provided incentives for industries to recycle. The possibilities for recycling together with the relatively high value of water in industrial uses suggest that world water supplies would be fully adequate to meet the growth in industrial demands over the coming decades.
3.
Reducing Storage Losses
The volume of water lost in reservoir storage is substantial, totaling some 188 km3 year1 in 1995. Improvements in the eYciency of water storage could reduce this number in the future either by using underground storage or by utilizing surface storage in areas with less evaporation. For example, Lake Nasser loses about 16% of its volume to evaporation each year (FAO, 1997), resulting in a reduction of some 10 km3 year1 in annual flow, or about 20% of Egypt’s annual use (Shaltout and El Housry, 1996). Storing an equivalent amount of water in the Ethiopian Highlands rather than in the lower desert portion of the Nile would reduce this loss to about 3% of the storage volume, liberating substantial quantities of additional water (FAO, 1997). Such a strategy would only be possible through basin‐wide agreements for water and hydropower sharing. As in the case of domestic and industrial conservation, water savings would be local or regional and of insuYcient volume to substantially reduce the global deficit created by population increase.
B. EXPANSION AND IMPROVEMENT OF IRRIGATION Irrigated agriculture is the dominant consumptive user of water. Thus, increases in the productivity of irrigation water through changes in management and improvements in eYciency oVer the greatest potential for global water savings. Regardless of how much more eYcient current use becomes, however, it seems unlikely that the increased demand for food resulting from population growth can be met without some expansion in irrigated acreage. There is both potential for future expansion of irrigated agriculture and opportunities for improvement in existing agricultural water use practices. To some extent, the expansion of irrigated acreage may depend on savings
46
W. A. JURY AND H. J. VAUX
which can be achieved in current and forecasted agricultural water use. It is important to recognize, however, that improvements in agricultural water use eYciency may not yield water that would otherwise be lost because drainage water or conveyance losses by upstream users may contribute to the water budget of downstream users. Thus, it is important for the implications of changes in irrigation eYciency to be analyzed and addressed locally and regionally. 1.
Potential for Expansion
Constraints to the expansion of irrigation are of three types: insuYcient land, insuYcient water, or excessive cost. The combination of these factors appears to explain the slower rates of expansion of irrigated lands that have prevailed throughout the last half century. Figure 9 shows the average growth rate of irrigated land globally since 1800. The expansion between 2000 and 2003 has been less than 0.3%, in contrast to the rapid growth from 1950 to 1990 following the Green Revolution. This trend, if it continues, will limit the contribution that expanded irrigated land can make to supplying the needs of the additional population in the next 50 years. There is a great deal of land on the planet not under cultivation that is potentially suitable for irrigated agriculture. However, conversion of much of this additional land may be seriously constrained by both environmental and financial costs. Extensive development of additional irrigated land worldwide would entail the destruction of many valuable natural ecosystems. Balancing the benefits of irrigation development against the losses of ecosystem services will have to be arrived at locally or regionally.
Growth in irrigated area (%)
4
3
2
1
0 1800
Figure 9
1850
1900 Year
1950
2000
Rate of growth of irrigated area over the last 200 years (data from FAOSTAT).
THE EMERGING GLOBAL WATER CRISIS
47
Increasing financial costs are another constraint on expansion of irrigated acreage. Most of the desirable locations for irrigation have already been developed. The remaining sites are more remote from markets and water supplies, may not be as fertile, and may be significantly more costly to develop. Financial costs may be a critical factor limiting expansion of irrigation in the developing world where financial resources themselves are sharply constrained. For these reasons, estimates of how much new land might be brought under irrigation vary considerably. Shiklomanov (2000) estimates that irrigation water withdrawals will expand by 685 km3 year1 from 1995 to 2025 (Table XI), to a total of 3189 km3 year1. This 685 km3 year1 reduces the water requirement to produce food for the new population from 2740 to 2055 km3 year1. Similarly, Rockstrom et al. (2007) assume that irrigation can increase by 790 km3 year1 between 2002 and 2030. Given the constraints to expansion of irrigation, it is unlikely that new development will exceed these estimates.
2.
Efficiency Improvements
Care must be taken in assessing the extent to which improvements in irrigation eYciency will result in true water savings. The overall eYciency of irrigation water use is often defined as the amount of useful crop transpiration relative to the amount of water withdrawn from the source point (i.e., the stream or aquifer). By this criterion, losses during conveyance as well as the extent of subsurface drainage after application count as ineYciencies that could be reduced through technology or better management. By all accounts, current irrigation use is very ineYcient by this definition. However, the extent to which the improvement of irrigation eYciency leads to water saving is complicated by the fact that drainage waters and deep percolation are often available for subsequent use. Savings in water that is available for reuse are not true savings. Thus, it will be important to assess regionally the extent to which improvements in water use eYciency locally lead to true water savings in a basin‐wide context. In short, the aim of eVorts to improve water use eYciency should be to reduce the irrecoverable losses of water. Seckler et al. (1998) estimated the average irrigation eYciency (water required for 100% yield divided by irrigation withdrawals) for 118 countries of the world in 1990 as 43%, and showed that increasing irrigation eVectiveness to 70% reduces the need for development of additional water supplies for all the sectors in 2025 by roughly 50% with a total water savings of 944 km3 year1. Table XV, which is adapted from Wood et al. (2001) using data from Seckler et al. (1998), gives the estimated irrigation withdrawals and eYciencies in 1990 by region. The true savings that are likely to be achieved are probably less than what is reported in this table simply because some of
48
W. A. JURY AND H. J. VAUX Table XV Irrigation Use and EYciency by Region in 1990 (Wood et al., 2001)
Region North America Latin America Europe Middle East/North Africa Rest of Africa India China Rest of Asia World
Irrigated area (Mha)
Irrigation withdrawals (km3 year1)
Irrigation eYciency (%)
21.6 16.2 16.7 22.6 6.1 45.1 48.0 61.3 243.0
202 163 103 219 68 484 463 377 2086
53 45 56 60 48 40 39 32 43
the water ‘‘saved’’ is water that is currently being used by others. Nevertheless, globally eYciency improvements will result in additional water supplies and there are a wide variety of ways in which they can be made. Postel (1998) divides the ways in which eYciency improvements can be made into four categories: technical, agronomic, managerial, and institutional. Technical improvements consist of methods for applying water more uniformly and reducing evaporation or runoV losses. Precision land leveling by laser improves uniformity of water application and allows a smaller volume of water to reach all areas of the field in suYcient quantity to ensure high yield. Sprinklers can be improved in several ways, including lowering the spray to reduce air losses and reducing the kinetic energy of impact. Surge irrigation is the intermittent application of water to a furrow achieved by alternating the flow between two irrigation sets through the use of an automated valve. This allows a more uniform application of water between the upstream and downstream ends of a furrow. Drip or subsurface irrigation minimizes water loss from evaporation and can achieve high levels of uniformity. Their cost limits the types of crop they may be used on. Poor management is a leading cause of irrigation ineYciency, particularly in developing countries (Jensen et al., 1990). Improvements in irrigation scheduling and water delivery timing will reduce water losses, as well as recognizing crop sensitivity to water stress at diVerent stages of development. Switching to demand‐based irrigation, either by soil monitoring or PET estimates, helps ensure that the right amount of water is added at the proper time. Proper tillage and field preparation can help promote infiltration and reduce evaporation (Wallace and Batchelor, 1997), and on‐farm recycling of drainage and tail water can produce significant savings. The eYciency of storage and water delivery from the source to the field averages about 70% globally (Bos, 1985), and can be improved by canal lining or other repair measures.
THE EMERGING GLOBAL WATER CRISIS
49
Proper crop selection can greatly improve irrigation water productivity and eYciency (Postel, 1999). Matching crops to climatic and soil conditions and the quality of water available can ensure optimum yields for a given irrigation volume. Crop sequencing can increase productivity in saline soils, and intercropping can increase transpiration relative to evaporation. Breeding new crop varieties for tolerance to drought, salinity, and water use eYciency can potentially have a huge eVect on food production eYciency. Postel (1999) lists five institutional measures for increasing irrigation water eYciency: development of water user organizations, reducing irrigation water subsidies, establishing conservation incentives, enhancing the legal framework for water marketing, fostering infrastructure for private sector dissemination of eYcient technologies, and better training end extension eVorts. These are discussed in a subsequent section.
3.
Deficit Irrigation
As long as water is readily available and inexpensive, irrigation practice calls for applying water so as to ensure maximum yield. However, as water scarcity intensifies, it may be more economical to under‐apply irrigation at various stages of crop growth, provided these stages are not critically sensitive to water stress. Such a strategy, sometimes called ‘‘regulated deficit irrigation’’ (RDI), can greatly increase the productivity of water (yield per water applied) provided that yields are not substantially suppressed. Appropriate use of RDI requires knowledge of the stages of crop sensitivity to water stress so that stress is applied at times when the impact on yield and crop quality is minimized. Table XVI, taken from Zhang (2003), shows yield and water productivity values for wheat and maize grown under diVerent water regimes in Texas and Syria. In all cases shown, reducing applications from regular irrigation levels by one‐third results in small yield reductions and significant water savings. Fereres and Soriano (2006) reviewed the literature on deficit water use and concluded that there was potential for improving the water productivity of a number of field crops provided that the level of supply of water is relatively high (i.e., >60% of PET). Fereres and Soriano (2007) showed that strategic application of water to permanent crops at stages in the life cycle where water stress was well tolerated could lead to minor reductions in yield while crop quality was protected or even augemented. RDI in permanent crops appears to be a highly eVective way to manage limited water supplies during periods of drought. It also has important implications for economizing on irrigation water in both annual and permanent crops in nondrought periods.
50
W. A. JURY AND H. J. VAUX
Table XVI Comparison of Water Productivity (PAW) of Irrigation Levels for Wheat and Maize (reproduced from Zhang, 2003) Wheat, Texas, United Statesa
Irrigation level Full 67% of full 33% of full Rainfed a b
Wheat, Syria
Maize, Texas, United Statesb
Yield (t ha1)
PAW (kg m3)
Yield (t ha1)
PAW (kg m3)
Yield (t ha1)
PAW (kg m3)
4.76 4.74 3.88 2.19
0.64 0.76 0.80 0.61
5.79 5.24 5.15 3.27
0.93 1.19 0.99 0.93
13.95 11.36 6.62 1.36
1.42 1.53 1.21 0.43
From Schneider and Howell (1996). From Howell et al. (1997).
C. PRODUCTIVITY IMPROVEMENTS IN RAINFED AGRICULTURE Since irrigated agriculture produces about 40% of the world’s food on only 17% of the total land under production (Fereres and Soriano, 2007), it is about 325% as productive as rainfed agriculture. This suggests that there may be considerable opportunity to increase the productivity of the latter. Indeed, many believe that the key to averting food shortages in the coming century lies in increasing the eYciency of rainfed agriculture. This makes sense for an additional reason. Much of the future growth in population will occur in poor countries relying on rainfed agriculture for part of their needs, especially in Africa where irrigation is not widespread. There are three main ways in which rainfed agriculture may be enhanced economically. First, there are numerous water‐harvesting schemes that may be employed to increase available water. Rainwater harvesting can focus on: (1) capturing water for domestic use, for example, by collection of rain falling on rooftops in cisterns; (2) replenishing green water, for example, through stone bunds on the contour line; or (3) increasing blue water available locally, for example, through small check dams that either increase recharge to the groundwater or store water in small reservoirs. Rainwater harvesting has been used successfully to increase water for domestic, agriculture, and ecosystem uses by hundreds of thousands of communities, particularly in India. It has even brought rivers back to life. However, when practiced on a large scale in upper watersheds, rainwater harvesting will reduce water available further downstream (IWMI, 2006). The second way in which rainfed agriculture can be enhanced is by strategic supplements of irrigation water. Supplemental irrigation with about 100 mm of water, provided during crucial drought spells, can double
THE EMERGING GLOBAL WATER CRISIS
51
rainfed yields of cereals from about 1 to 2 t ha1, increasing water productivity to 0.5 kg m3 of water consumed. There are many technologies for supplemental irrigation that range from farm ponds to microirrigation with shallow groundwater pumped with treadle pumps (IWMI, 2006). The third method for increasing rainfed agricultural productivity is through improved land management. Typically, a significant amount of the rainwater striking the land surface is lost through evaporation or runoV. By enhancing infiltration and water storage capacity, more of the rainwater can be converted into transpiration and hence into enhancement of crop yields. Use of terracing, contouring, and microbasins are important measures for maximizing rainfall infiltration into the soil to increase yields, especially for farmers in sub‐Saharan Africa, Latin America, and South Asia. Conservation or zero tillage—where crop residue is used as mulch—is a promising technology (IWMI, 2006). Modest amounts of supplemental fertilization in concert with strategic water additions and improved soil management can have a dramatic eVect on crop yields in rainfed systems facing periodic droughts. Rockstrom et al. (2002) concluded that there were no agronomic or hydrologic barriers to doubling crop yields in the semiarid tropics, and called for a new strategy of integrated rainfed management that focused on alleviating water stress and maximizing transpiration through optimized use of water, fertilization, and land management improvements. Figure 10 shows crop yields of maize and sorghum in Africa under standard and supplemented conditions. The enhancements in yield through supplemental irrigation alone were comparable to those achieved solely by fertilization, but the combination of the two resulted in yields that were up to twice as large as the controls.
D. ECONOMIC METHODS FOR WATER SUPPLEMENTATION IN DEVELOPING COUNTRIES The costs of large‐scale irrigation projects or sophisticated technologies such as conventional drip irrigation are too high for many small‐scale poor farmers in developing countries. Yet it is at this scale where the greatest gains in yield and water productivity may be gained through supplemental water additions at strategic times to avoid damaging stresses to the plant. A variety of inexpensive methods have emerged recently that are helping to raise yields in developing countries where water is scarce. These are of four types: inexpensive pumps, microirrigation devices, on‐farm water harvesting, and flood recession farming (Postel, 1999). Human‐ or animal‐powered pumps for lifting shallow groundwater to the surface have become extremely popular on small farms in developing countries. The most promising of these is the treadle pump, a low‐lift, high‐capacity,
52
W. A. JURY AND H. J. VAUX 4
A
3
Water use efficiency (kg mm−1 ha−1)
2
1
0 10
500
0
1000
1500
2000
B
8 6 4 2 0 0
1000 2000 3000 Grain yield (kg ha−1)
4000
Figure 10 Water use eYciency (kg grain per unit rainfall þ supplemental irrigation) for sorghum in Burkina Faso (A) and maize in Kenya (B). Control ¼ traditional farmers’ practice with no fertilizer application (circles), WH ¼ supplemental irrigation using water harvesting (squares), FERT ¼ fertilizer application (30 kg N ha1) (triangles), WH þ FERT ¼ supplemental irrigation combined with fertilizer application (diamonds) (after Rockstrom et al., 2002).
human‐powered pump designed for farms of 0.5 ha or less. It operates like a Stairmaster exercise machine, using a walking motion to provide the lift. It can fetch 5–7 m3 of water per hour from wells and boreholes up to 7‐m deep as well as from surface water sources such as lakes and rivers. Under typical conditions, the treadle pump costs only about 25% of the retail price of motorized pumps of comparable flow rate capacity. It also costs much less to operate, having no fuel and only limited repair and maintenance costs (Perry and Dotson, 1996). As a result, farmers can recoup their investment several times over in less than 1 year. The treadle pump was introduced in Bangladesh in the 1980s, and over 1.2 million units have been sold there alone. Sales in India started later and had reached about 200,000 units by 2000, but the market potential is as high as 10 million (Shah et al., 2000a).
THE EMERGING GLOBAL WATER CRISIS
53
Drip irrigation has been demonstrated to reduce water use by 30–70% and to raise crop yields by 20–90% in countries as diverse as India, Israel, Jordan, Spain, and the United States (Postel, 1999). However, conventional drip and sprinkler irrigation have capital investments that place them outside the reach of the small farmer in a developing country. Low‐cost alternatives suitable for small plots or home gardens are proliferating, involving gravity feed in lieu of mechanized pressure‐driven technology. The simplest of these is the bucket kit, consisting of a water‐filled bucket or tank placed at shoulder height and connected to microtubes that are placed at strategic locations on the plot. Systems costing as little as US$5–10 can irrigate 100 plants (Postel, 1999). A somewhat larger version of the bucket kit is the drum kit, which uses a larger water source and can irrigate a larger area. At a larger scale are shiftable drip and stationary microtube systems, which operate like drip systems but use gravity feed and passive emission through holes or microtubes (Postel et al., 2001). On‐farm water harvesting is an ancient practice now being revived to augment water supply to rainfed fields. It may consist of building embankments around the field to capture and infiltrate water during the rainy season, or more elaborate channeling to divert runoV from adjacent areas to the field. Storage tanks to collect water during the rainy season are also being revived (Postel, 1999). Flood recession farming is the practice of growing crops on land that is flooded annually during the recession period. It has the advantage of fertility replenishment through sediment deposition, and brings more land into production. It is another ancient practice that is being revived as an alternative to dams (Postel, 1999).
E.
DESALINATION
For many decades desalination was thought to hold promise for substantially alleviating the global problems of water scarcity by drawing on the nearly unlimited reservoirs of the world’s oceans to make water readily available. Historically, the promise of seawater conversion has remained unrealized because of the relatively high costs of converting seawater to freshwater. The diYculties were compounded by the fact that desalting technologies tend to be energy intensive and there was little willingness to link the cost of water supplies to prices as volatile as those in the energy sector. There have been important advances in desalination technology in the last few decades that have brought costs down to the point where desalinated seawater may be a feasible supplement to conventional water supplies where the value of water is high. Previously, seawater conversion was utilized only in very wealthy countries that had virtually no alternative sources of supply.
54
W. A. JURY AND H. J. VAUX
The costs of seawater conversion are still relatively high so that virtually all applications are for high‐valued urban uses. Thus, by the late 1990s, there were more than 12,500 desalination plants in operation in the world, generating more than 6 billion gallons of freshwater per day and accounting for about 1% of the world’s daily production of drinking water (Martin‐ Lagardette, 2001). Inasmuch as the costs of desalination depend on a variety of factors including the degree of salinity and composition of the source water, disposal costs, and the cost of energy, not all of these systems entail the conversion of seawater. Table XVII shows comparative costs of water for diVerent source waters. Brackish water desalination costs are low enough that it may be economically feasible to use the process to treat saline groundwater in certain areas. There is a huge global supply of subsurface water that is currently too saline for practical use. In New Mexico, for example, 75% of the groundwater is too saline for most uses without treatment (Reynolds, 1962). It is projected that more than $70 billion will be spent worldwide over the next 20 years to design and build new desalination plants and facilities (Sandia National Laboratories, 2002). It remains to be seen, however, whether the costs can ever be brought low enough to make desalinated water attractive for agricultural uses. High capital costs will tend to mitigate against extensive use in developing countries while high and uncertain environmental costs and the volatility of energy costs will tend to reduce its attractiveness in developed countries. Ultimately, desalination technologies may have important applications in the treatment of wastewater, though these will tend to be relatively expensive. Research on the desalinating technologies continues apace and significant research breakthroughs in the future could make desalination a more attractive source of supply than it has been in the past (National Research Council, 2004).
Table XVII Water Costs to Consumer, Including Treatment and Delivery, for Existing Traditional Supplies and Desalinated Water (AMTA, 2001)
Supply type Existing traditional supply New desalted water: Brackish Seawater 50% traditional supply and 50% brackish water 90% traditional supply and 10% seawater
Unit cost ($ per 1000 m3) 240–660 400–800 800–2100 320–730 290–800
THE EMERGING GLOBAL WATER CRISIS
55
F. IMPROVEMENTS THROUGH INSTITUTIONAL CHANGES Throughout much of the world the historical means of developing water supply entailed the construction of large‐scale infrastructure like dams and canals. For a number of reasons, the era dominated by the construction of large‐scale water supply infrastructure is over. It is now widely understood that such facilities cause significant environmental damages and the dollar costs of those damages appears to be high (Constanza et al., 1997). The relative costs of constructing and operating such infrastructure have also grown, making this approach less attractive in developed countries, and generally unaVordable in developing countries. Moreover, such projects frequently do not serve the poor, who are now a major component of the world’s unserved population. The passing of the era of large infrastructure means that the greatest potential for improving global water security lies with better water management. Unfortunately, the current institutional arrangements for managing water suVer from a host of deficiencies. Water institutions are defined as all of the collective arrangements people have made and make to facilitate the use and management of water resources. They include laws, codes, public organizations, boards, and water districts. Existing arrangements are not well adapted to modern circumstances because many institutions were created when the problems of developing and managing water resources were very diVerent from what they are today. Water institutions tend to embody a focus on narrow interests as opposed to being holistic. They create and maintain artificial distinctions between water quantity and quality. They embody multiple and fragmented jurisdictions across river basins and watersheds, thereby making integrated resource management more diYcult. Above all, there is a general absence of institutions equipped to deal with the fundamental water problem of the twenty‐first century, which is scarcity (Jury and Vaux, 2005). In some instances needed institutions are entirely absent or incompletely developed. For example, 60% of all river flows are found in transnational river basins. Yet, institutions for managing international rivers are either nonexistent or incompletely developed. There are examples of eVective institutions such as the International Joint Commission which governs boundary waters between the United States and Canada and the Commission for the Protection of the Danube (International Joint Commission, 2006). And, there are a number of eVective treaties. But these arrangements are the exception rather than the rule. EYcient use of water and eVective management require certainty with respect to rights and ownership, and waters in international basins that are not managed to achieve these outcomes will typically not be eYciently used. Modern water management institutions will need to incorporate two important attributes which have not received much emphasis historically. The first is stakeholder involvement. There is a growing body of evidence
56
W. A. JURY AND H. J. VAUX
showing that the engagement of stakeholders specifically and the public generally is essential to the development of eVective water management plans and institutions as well as the ongoing implementation of plans and policies. It is also important to recognize that stakeholder involvement should not be restricted to water plans in the developed world. On the contrary, evidence suggests that stakeholder involvement is every bit as important in the developing world as it is in the Western world (Benabdallah, 2006). The other crucial attribute needed by institutions is adaptive management, which is a systematic response to the uncertainty inherent in hydrologic and water management systems. Adaptive management embodies flexible rules and policies which permit management routines and prescriptions to be changed as more learning and experience is gained with the hydrologic system in question (National Research Council, 1999d). Increasingly, institutions which are adaptive and permit water resources to be managed adaptively are needed to accommodate hydrologic and other types of uncertainty. The fact that specific local and regional impacts of future global climate change are largely uncertain provides another compelling justification of the need for adaptive institutions. There are several other important institutional characteristics that flow from long‐standing prescriptions. For example, the European Water Framework Directive creates uniform standards for water policy within the European Union, but places the focus of management strategies on the regional or river basin level. The notion of creating uniform standards but allowing them to be applied and enforced in a decentralized way allows variations from place to place to be accounted for but ensures that the standards are of high level and that competitive environments in which they might be diminished are absent (Young and Haveman, 1985). Inasmuch as economics is the science of managing scarcity, economic instruments such as prices and markets need to be incorporated into modern water management institutions. Water prices typically do not reflect the scarcity value of water. That is, prices generally reflect the cost of capturing and delivering water but assign a scarcity value of zero to water. This is inappropriate in an era of scarcity because it signals consumers that the water is freely available which it is not. Care must be taken, however, in utilizing prices to provide protection for poverty‐stricken people who may be able to pay nothing for the water needed to support basic needs. There are numerous schemes which permit pricing to reflect the scarcity value but allow a basic or lifeline quantity of water to be available at little or no cost. Water markets are also an important institutional arrangement for managing scarcity, and market‐like institutions for allocating water have begun to appear with increasing frequency in the developed world. Research on water management institutions has lagged in recent years, which is unfortunate since much remains to be learned about human
THE EMERGING GLOBAL WATER CRISIS
57
behavior in water use. For example, more needs to be known about structures of incentives, about how stakeholders should be organized, about the influence of culture on water‐using behavior, and on how to devise eVective programs of public education and social learning. Innovative institutions will surely be an important part of an eVective solution to emerging global water problems. It will be important to update and modernize many existing institutions as well as to rely on new innovations as part of all eVorts to improve water management.
V. PATHS TO SUSTAINABILITY There is no quick fix for solving the world’s water problems. Rather, improvements in water management practices are needed in virtually all of the arenas where water is used. This includes the consumptive uses that are dominant in the agricultural, industrial, and municipal sectors. It also includes practices for managing critical instream or nonconsumptive uses, particularly those related to the maintenance of aquatic and associated ecosystems. Given the expected growth in population and the large number of places where water will continue to be scarce or become so, following the path to sustainability will require a global eVort at improving water use practices. All regions and locales must be part of the global eVort, simply because the fundamental drivers of water supply and demand are inherently regional or local. And, this global eVort must begin immediately to avoid reaching a state where the crisis will be unmanageable and the cost to the planet and its inhabitants unimaginable. Over time, it is reasonable to assume that new scientific advances will oVer potential help in adapting to the water realities of the future. In this connection, it is important to recognize that some of our existing scientific knowledge about water and its management remains underutilized or ignored. For example, it is well known that it is almost always cheaper to prevent the pollution of water courses and groundwater at the outset than it is to clean up and remediate pollution events after they have occurred. This suggests that a premium should be placed on developing and enforcing regulations that tend to make episodes of pollution the exception rather than the rule. One important principle guiding future management paradigms should be to avoid actions that make the situation worse than it needs to be. Examples of these undesirable actions are practices that lead to water pollution, lead to environmental impairment or destruction, or encourage low‐valued uses of water while higher valued uses go unserved. A corollary principle is that sustainable management practices and norms of use should be developed wherever they can. As used here, the principal of sustainability requires that
58
W. A. JURY AND H. J. VAUX
the quality and quantity of water available to future generations should not be significantly impaired by the actions of current generations (Rawls, 1971; Weiss, 1989). As a general proposition, sustainable practices will usually be preferable to those that are inherently unsustainable. Some unsustainable practices will be inevitable, however. The extraction and use of fossil groundwater or other groundwaters that do not recharge is inherently unsustainable because any use hastens the day of exhaustion. In these instances, it will be important to ensure that the resource in question is used in ways that build the physical and social capital available to the society of the future. Of necessity, the concept of sustainability will be highly nuanced. Nevertheless, available science as well as common sense points to numbers of unsustainable water management and water use practices that need to be ended as a matter of priority.
A. ENDING UNSUSTAINABLE PRACTICES The first step on the path to sustainability is to implement a strategy to reverse unsustainable practices that are depleting or damaging water resources excessively. Despite the a priori urgency of ending unsustainable practices, it will not always be cheap or even practical to reverse them. This is particularly true in developing countries that do not have the resources and, in some instances, the know‐how to implement known management practices that would be more sustainable. These diYculties will be identified in the discussion that follows. The three most significant unsustainable practices that exist worldwide are persistent groundwater overdraft, the continuing contamination and pollution of groundwater and surface waters, and inappropriate management of watersheds. Persistent groundwater overdraft is always self‐terminating, and potentially disastrous if left unmanaged. The transition to sustainability therefore requires that termination of overdraft be managed in terms that are most economically and hydrologically advantageous. It also requires that such changes in management regimes be timely to allow for the development of alternative supplies, where possible and justified, or for scaling back water use in an orderly manner where that is called for. The economic and hydrologic principles for managing groundwater are well established. What is needed is the institutional and political will to apply them in regions that are persistently overdrafting groundwater. Although opinions diVer about the severity of consequences that may occur in regions where critically needed food is produced through excessive pumping, an immediate plan is needed for reducing water use to levels that can be supported by available supplies. While there may be doubts about the seriousness of overdraft or the availability of substitute sources of supply, the loss of ability to feed
THE EMERGING GLOBAL WATER CRISIS
59
hundreds of millions of persons would be such a serious prospect that careful contingency plans should be devised and implemented to avoid it (Moench et al., 2003). The world is already feeding some 500 million people with food grown in regions suVering chronic overdraft (Postel, 1999), and a significant part of the projected population growth in the future will be in these same regions. Strategies for reducing persistent groundwater overdraft by agriculture may involve changes in cropping patterns, investment in new technology, repair, and upgrading of infrastructure, and even deficit irrigation. To be successful, all of these strategies must result in reduction of extractions. Where extractions are not reduced suYciently, reductions in crop acreage may be the only alternative. In some cases, water transfers or the development of supplemental supplies may help to oVset the loss of accustomed levels of groundwater use that are required to terminate persistent overdraft. In almost all instances, the development of new supplies is likely to be enormously expensive and must be considered to be a last resort. It is important to recognize that there are circumstances in which coordinated groundwater management schemes may not be possible. In India, for example, there are some 22 million farmers independently extracting groundwater from common and interrelated aquifers. It probably would not be feasible, much less practical, to develop the institutional arrangements needed to coordinate their water‐extracting behavior to achieve the desired reductions. For this reason, the Indian government is developing a massive surface water importation scheme that will cost many billions of dollars. Whether the Indian government or any other government can aVord such schemes remains an open question (Shah, 2000, 2003). Persistent groundwater overdraft by large metropolitan areas can be even more serious because reduction of water supply could have immediate life‐ threatening consequences for large numbers of people. In addition, groundwater overdraft can cause damaging land subsidence within a metropolitan region. In Mexico City, for example, land subsidence of 7.5 m has occurred in the downtown area since pumping began (National Research Council, 1995). Since urban demands for water will always take precedence over other priorities, water‐short cities may end up drawing water from adjacent agricultural operations to satisfy their needs, or alternatively developing large water transfers. The Mexico City case is particularly alarming because there are apparently no locally available substitute supplies and the costs of lifting remote supplies to the Valley of Mexico at an elevation of 2500 m would make such water unaVordable to most of the residents of Mexico City. By contrast, the chronic water supply shortfall in Beijing, China has forced the government to divert substantial water from surrounding farmland and to initiate a water diversion from the Yangtse River in the south at enormous expense (Postel, 1999). There are many less expensive opportunities to achieve
60
W. A. JURY AND H. J. VAUX
water savings in metropolitan areas. These include the repair and maintenance of infrastructure, metering water use, pricing water realistically, education, water reuse, and other conservation programs (National Research council, 1995). The assertion that inadequate measuring and monitoring of groundwater casts doubts about the severity of overdraft does not justify inaction (Moench et al., 2003). There is certainly a crucial need to develop real‐time monitoring and measuring systems for aquifers around the world. However, the urgency of emerging water problems will not allow the luxury of waiting for a period suYcient to document beyond all doubt that an aquifer is being persistently overdrafted and economic exhaustion is in sight. This is particularly true where alternative sources of supply are either unavailable or largely uneconomical to develop. Pollution of surface water and groundwater poses a threat that must be sharply reduced and eliminated where possible. Moreover, pollution problems must be solved in both developed and developing countries. In developed countries, the continuing creation of new chemicals that can ultimately be dispersed in the environment, the incompleteness of strategies for controlling nonpoint source contaminants, the emergence of new constituents of concern such as endocrine disruptors, and the continuing threat to groundwater quality from past chemical use and disposal practices will all require renewed and more intense attention. Failing that, available high‐ quality supplies in advanced countries will continue to diminish even as demands for those waters grow. The water quality problem is even more serious in developing countries, where excessive pollution poses significant health threats as well as a loss of needed water supply. The absence of adequate sanitation services for over 2 billion people—most of whom live in developing countries—by itself poses an enormous threat to water quality. That threat is likely to intensify as population grows. It has the potential to condemn the populations of developing countries to an ever‐tightening spiral of population growth followed by increased pollution leading to a further loss of badly needed high‐quality water supplies. Examples abound. China expert Vaclav Smil has estimated that as much as 20% of China’s river water is too polluted for even irrigation purposes (Postel, 1999). The aquifer on which Mexico City depends for 75% of its water supply is potentially vulnerable to pollution from a variety of pathogens and toxic chemicals (National Research Council, 1995). Economic development in many of these countries will be diYcult to initiate and sustain. As industrialization and more intense agriculture develop, there will be an understandable reluctance to restrain such development through the imposition of eVective anti‐pollution measures. Additionally, the expense of undertaking more centralized government‐sponsored pollution clean‐up programs may be greater than can be supported in a developing
THE EMERGING GLOBAL WATER CRISIS
61
economy. It is also true that in the absence of eVective regulation, industry and agriculture in developing economies will be likely to emit contaminants that have long been controlled or banned in more developed nations. Many developing countries are likely to have to rely on external technical and financial aid if they are to address successfully the pollution problems that are likely to be associated with economic growth. The enormous costs of remediation and clean up of groundwater and surface water supplies suggest that, from a broad perspective, it would likely be cheaper to take whatever actions are needed to avoid severe pollution episodes. But from the internal perspective of a developing country, it may not look this way at all. This suggests, then, that global and regional programs of financial aid to protect and enhance water quality may need to be underwritten by the developed world. The potential for salinization of lands where irrigated agriculture is practiced is another particularly insidious threat to water quality. Virtually all irrigation water contains salts which are left behind in the root zone of the soil profile as the water is evaporated from the soil surface and transpired by plants. As salts build up they restrict plants from extracting water from the soil and ultimately the land ceases to be productive. The threat of this process is present wherever irrigated agriculture is practiced and is more serious the larger the salinity concentration of the water. The scientific principles of managing salinity are well known and are competently practiced in some areas of the world (Knapp, 1991). Ironically, the best way of managing salinity requires more water to leach salts from the soil and drainage facilities to carry the leachate oV into a suitable disposal site. Agricultural salinity is insidious both because it can destroy the productivity of agricultural land and because controlling it requires additional supplies of water that are already scarce in most regions of the world. Up to 47.7 million ha of irrigated land worldwide (about 21% of the total) has been degraded by salinity, with many millions more likely to be degraded in the next decades in the absence of management changes (Postel, 1999). In the coming decades when there will be a need to increase global agricultural productivity substantially, every eVort will have to be made to attenuate or stop the destruction of agricultural lands through salinity. Failure to deal with this problem eVectively will greatly increase the diYculties of meeting future food demands and managing scarce water resources. Salinization is not the only water‐driven process aVecting land that must be addressed to achieve sustainability. Inappropriate land management practices on upland watersheds can lead to the degradation of water quality and increase runoV volume and variability over time (National Research Council, 1999b). Integrated water management strategies have worked well in some instances in addressing the problems of protection and appropriate land use in watersheds. Much remains to be done in the developed world,
62
W. A. JURY AND H. J. VAUX
however, where the problems of watershed management are formidable. Frequently, the residents of upland watersheds are poverty stricken and must use all of their meager resources to wring a living from the land. The pressures of mere survival often preclude any eVorts to manage and husband watershed lands in a sustainable way (Benabdallah, 2006; Sullivan, 2006). There are many other nonsustainable practices beyond those discussed which are site specific and restricted to particular regions and locales. It will nonetheless be important to modify such practices where feasible so that they are sustainable.
B. MANAGEMENT STRATEGIES There is much that can be done by way of improvement of water management strategies around the world. Management strategies include those that are technically based such as integrated resource management, conjunctive management of groundwater and surface waters, underground storage, and irrigation scheduling, and institutional strategies which include a panoply of economic, political, and other instruments designed to change behavioral approaches to water management. While it is unlikely that all water management strategies can be used everywhere, it is clearly possible to devise a mix of strategies that will be appropriate for each region and locale. Again, the involvement of stakeholders in the development and implementation of management strategies will be crucial to achieving success. Technically based management strategies rely for their eVectiveness on scientific knowledge about the behavior of water and the eVectiveness of various ways of managing it. Integrated resource management refers to a strategy that manages water and associated land and air resources as an integrated whole. The concept acknowledges that management actions focused on one resource have implications which are frequently very significant for other resources. In addition, integrated management embodies the notion that the watershed should be the fundamental unit of management because it is the fundamental hydrologic unit. Typically, the management of watersheds is confounded by the fact that diVerent political jurisdictions overlay a single watershed. These jurisdictions can be anything from nations to provinces or states, to overlapping local units of government. The job of managing hydrologic resources on a unified basis as well as the practice of integrated resource management is made enormously diYcult by the absence of a single jurisdictional entity. The usual result is that watersheds are not managed in a unified way and integrated resource management is rarely, if ever, practiced (Naiman, 1992; National Research Council, 1999b). Experience indicates that eVorts to move toward strategies of integrated resource management, including watershed management, must proceed in a
THE EMERGING GLOBAL WATER CRISIS
63
stepwise fashion. In general, experience shows that such eVorts are more likely to be successful on smaller watersheds. The larger the watershed, the more complications, so that experience with smaller units can be very important when devising management strategies for large basins. Additionally, there is no one prescription for integrated resource management. Each watershed is diVerent both physically and sociologically, and management eVorts must be adaptive. That is, the strategy should be flexible to begin with and should be adjusted or adapted periodically to reflect the results of experience with time. Again, eVorts at integrated resource management are more likely to be successful where stakeholders are involved and the public is engaged (Doppelt et al., 1993; National Research Council, 1999b). Although resource management is typically more diYcult to practice in an integrated fashion in very large river basins, there is still a critical need for formal basic allocation mechanisms (McCaVerty, 1998). Most of the surface waters and nearly all of the groundwaters of the planet that transcend international boundaries are not subject to treaties or formal decrees or doctrines specifying the rules of allocation. This means that in most cases, entitlements are clouded or uncertain. A lack of certainty about water entitlements constrains the development and sometimes the use of water resources. People are often understandably reluctant to rely on water supplies whose legal allocation is clouded. A high priority in every transnational watershed on the globe should be to establish clear treaties or allocations which firm up legally the respective rights to use water and apportion various quantities of groundwater and surface water flows. This will not be an easy undertaking, yet it will be essential. The longer the wait the higher the stakes in any eVort to allocate and the higher the stakes the more diYcult it will be to forge multilateral agreements. Conjunctive use of groundwater and surface water acknowledges that there are inherent hydrological interconnections between these apparently diVerent sources of supply. At its simplest, conjunctive use entails the reliance on surface waters during times of average or above average precipitation and runoV. During drought periods or other times when surface water availability is constrained, use shifts to groundwater which tends to be buVered to some extent from the variabilities that surface water is subject to. Looked at diVerently, groundwater can be managed as a reservoir for use during periods of surface water shortfall and recharged during periods of normal or above normal availability of surface water. Sophisticated schemes of conjunctive use employ managed recharge whereby excess surface waters are captured and transformed into groundwater. Managed recharge can be accomplished simply through the use of surface spreading or through direct injection, which generally requires significant investment in facilities. There are several preconditions for eVective conjunctive use. First, the management strategy must be structured to acknowledge the holistic nature
64
W. A. JURY AND H. J. VAUX
of groundwater and surface waters. Second, there must be clear systems of water rights for both groundwater and surface water. Too often, groundwater rights are poorly defined or absent. This latter circumstance tends to lead to underinvestment in conjunctive use or no investment at all. The scientific principles of conjunctive use have been well understood for decades, but there has been a lack of will and resources to apply them widely (Morel‐ Seytoux, 1985). Conjunctive use is one of the strategies that will need to be employed on a widespread basis as part of the response to emerging water problems. Conjunctive use and managed recharge are one means of addressing persistent groundwater overdraft. Strategic utilization of underground storage is related to conjunctive use. The construction of surface water storage facilities to capture water in wet times and places and hold it for use in dry times or convey it to dry places has been a time‐honored method of dealing with hydrologic variability and related water scarcity. The easily accessible and economically attractive surface water storage sites have already been developed. With a few exceptions, those that remain are either very expensive to develop or in remote locations. In addition, surface water storage facilities are now known to cause significant damages to riparian ecosystems. In contrast, underground storage oVers the potential for sequestering large quantities of water while avoiding environmental and economic problems that tend to be associated with surface storage facilities. Underground storage is likely to be more costly than simple conjunctive use schemes, however. Investment is required in recharge facilities, even where simple land spreading is used. Care must be taken to ensure that recharge waters are of appropriate quality (National Research Council, 1994). There are a number of technical problems such as clogging that must be managed. Underground storage oVers significant opportunities worldwide to alleviate water scarcity, but the costs may be beyond the capacity of many developing nations to finance. External financial aid and technical assistance will likely be required if the full potential of underground storage is to be realized, particularly in the developing world. There are, of course, many ways in which water‐scarce countries may adapt to an associated scarcity of food some of which are explored by Moench et al. (2003). One important example concerns the concept of virtual water. Currently, there are countries that do not have suYcient indigenous water supplies to feed existing populations. In the next 15–20 years many more countries will join this list, including India and possibly China (Yang et al., 2003). There is evidence to suggest that when countries become unable to grow suYcient food to feed their populations, they respond by importing cereal grains and other agricultural commodities. One way of viewing these imports of agricultural commodities is based on the proposition that the importation has the same impact as developing locally the water needed to
THE EMERGING GLOBAL WATER CRISIS
65
grow the commodities. In another words, agricultural commodities carry with them embodied water or the water that is needed to grow them. Importing countries then, in eVect, import water by importing crops. Such water is sometimes called ‘‘virtual water’’ (Yang and Zehnder, 2002). This route oVers one potentially significant adaptation via international trade through which water‐poor countries can import water‐intensive agricultural commodities from countries that are more generously endowed with water resources. There are several constraints and implications of this method of adaptation which need to be noted. The extent to which countries can adapt to water scarcity by importing foodstuVs will, of course, be partly a function of their ability to generate foreign exchange. This may be problematic for the poorest of countries, particularly those in Africa. In the absence of adequate foreign exchange, world food relief organizations may be able to assist but there has been no systematic assessment of the potential for such organizations to respond to a world food crisis. Thus, it is unclear to what extent water‐scarce countries can oVset that scarcity by importing food and to what extent international trade can ameliorate water scarcity globally. There is in addition another crucial fact that emerges from analyses of the prospects of virtual water. If water scarcity manifests itself rather directly as food scarcity, the demand for food exports from countries that have relatively generous endowments of water should expand. (These countries are found in Europe and the Americas.) To the extent that food for export is grown in irrigated agriculture the derived demand for water in those countries will also expand. Through this mechanism the water scarcity of water‐short countries worldwide will have direct impacts on the demand and availability of water in the water‐rich countries. For this reason alone, it will be important for water‐rich countries to stop unsustainable water management practices, adopt improved management strategies, and, in general, practice water stewardship more carefully than has been the case in the past (Vaux, 2004). All of these general management strategies as well as those that are specially adapted to particular regional and local circumstances will have to be devised and implemented in an environment that is cloaked with uncertainty because of global environmental change. While knowledge of the likely impacts of global warming grows significantly with time, it is still not possible to predict the specific eVects that will occur on a regional basis. What is known is that change is likely, and weather extremes will become more common (Section III.G). The fact that the world will have to adapt to water scarcity in the face of this added source of uncertainty makes the task of global water management more diYcult. It will place a premium on the capacity to devise strategies that are adaptable and can be adjusted as climate changes. For that reason, the focus should be on the delivery of services and not on the development of infrastructure. Adaptive management will be
66
W. A. JURY AND H. J. VAUX
critical and large‐scale infrastructure is typically diYcult to manage adaptively. Moreover, it is likely to be easier to respond to climate change if water resources are well managed to begin with. This means that unsustainable practices need to be corrected and management strategies should be as adaptive and flexible as possible (Vaux, 1991).
C. AGRICULTURE AND WATER MANAGEMENT IN THE DEVELOPING WORLD Agriculture will continue to be the dominant consumptive user of water globally. As population and the demand for food grow, it will be important to find ways to increase the productivity of agriculture everywhere. Irrigated agriculture is far more productive than rainfed agriculture, and there are many ways in which the productivity of irrigated agriculture can be increased in developed countries (Section IV.B). The poorest regions of the planet are likely to be the ones hurt worst by a water crisis in the future. But many of these same regions have the greatest potential for increasing crop yields from currently low levels. Modest investments or subsidies in the soft‐path technologies can produce dramatic increases in productivity at the local level not just with irrigated agriculture but with rainfed agriculture as well (Section IV.C). Improvements in the water use eYciency of rainfed agriculture need to be developed and disseminated through training programs. Thus, for example, simple water harvesting techniques, improved crop rotations, and other relatively inexpensive and decentralized techniques and technologies can have a disproportionate impact on the productivity of rainfed agriculture worldwide. While past water management practices have focused on the development and utilization of surface water and groundwater, future practices will have to focus relatively more on the utilization of rainfall through improvements in rainfed agriculture. Just as soft‐path, decentralized, and inexpensive technology will be the key to improving agricultural productivity in the developing world, the same sorts of technology will have to be created to provide drinking water and sanitation services to the underserved poor in developing countries. The evidence suggests that it will be critically important to have low‐cost, community‐managed water supply and sanitation services. Infrastructure alone often does not lead to an increase in access to water and sanitation services because top‐down technology‐driven projects frequently do not involve the users directly, tend to be poorly maintained, are subject to breakdown, and have short usage times. In addition, low levels of financial recovery in poor countries mean that operating and maintenance costs are not covered, so the systems do not function as intended and are badly managed (Rijsberman, 2004).
THE EMERGING GLOBAL WATER CRISIS
67
The Millenium Development Goals, established in 2000, call for a halving of the number of people who do not have adequate access to water and sanitation services by the year 2015. Although few expect those goals to be achieved, progress is being made. The UN Development Programme made an early analysis and noted that a group of countries containing 40% of the world’s population, primarily in Asia, have either achieved the goals or are on track to achieving them. China and India containing roughly half the world’s unserved population are among this group and their high rates of economic growth suggest that the prospects for achieving the goals are good (Rijsberman, 2004; United Nations Development Programme, 2003). A second group of countries containing 30% of the world’s population, primarily in Africa, are not making progress (United Nations Task Force on Water and Sanitation, 2003). The results of this analysis suggest that external funding should be focused on Africa and a few other countries that appear not to be making progress. The important point here is that the Millenium Development Goals have focused attention on the problem of drinking water and access to sanitation services and significant progress is being made toward worldwide provision of these services although the Millennium Goals are unlikely to be met in the proposed time frame. There are no similar programs focusing on the use of water in agriculture and the production of food to feed a more populous world, however. A clear conclusion is that although provision of adequate water supply and sanitation services is deserving of the highest priority, finding and implementing ways to improve the productivity of water in agriculture, particularly rainfed agriculture in developing countries, is not far behind.
D. SOCIETAL CHANGES There are a number of collective actions that can be taken by the global population that can also make a diVerence. Such actions will require individuals to make choices that will benefit the larger population as a whole. Such choices will, at times, run counter to personal preferences. Thus, such collective action will probably require a new global water ethic or some other system of incentives if collective action is to be eVective. There are several examples of important collective actions that could be taken and they include changes in dietary patterns, a conservation ethic, and cooperative management of shared resources. Selection of appropriate dietary patterns can have an enormous positive eVect on the global water balance. Evidence shows that as countries develop economically diets change and become much richer in meats. Yet, meat consumption entails approximately eight times the water input per food calorie produced compared to a vegetarian diet (Falkenmark and Rockstrom, 2004).
68
W. A. JURY AND H. J. VAUX
A simple example illustrates the impact. A person for whom meat accounts for 10% of the daily caloric intake will require 1.7 times more water input to produce that food than a person whose diet contains no meat. Of course, to the extent that the water input comes from rainfed circumstances the impact may not be as great. Nevertheless, dietary shifts on a global basis can be quite important. Even shifts from beef to pork or chicken will result in substantial water savings (Smil, 2000). It seems unlikely that populations in developing countries would be willing to forego meat in their diets without some concomitant changes in developed countries. Another example of collective action would be a global water conservation ethic. In developed countries, levels of personal and household water use are usually much higher than those in developing countries (National Research Council, 1999a). Some of that use is either low valued or outright waste. Thus, one strategy for stretching water supplies during drought periods entails appeals and other incentives to use water more carefully in everyday uses such as landscape irrigation and interior household uses. The typical pattern is that when such appeals are eVective, water use returns to pre‐drought levels once the drought is over. There is little question but that much water could be saved with a resulting amelioration of world water scarcity if the kind of water conservation practices that are typical of drought circumstances in developed countries were practiced on a consistent basis. This would might, in turn, provide an example for developing countries. A final example of collective action would be the widespread adoption of collaborative arrangements for governing and allocating shared water resources. There is a strong tendency to treat such resources in a competitive fashion. When property rights to water are treated in this way, it is frequently true that low‐valued uses in one sector are served while higher valued uses in other sectors go begging. This means that the aggregate productivity of water is less than it might otherwise be. Collaborative arrangements in which there is a commitment to allocating water to its most productive and highest valued uses and a parallel commitment to flexibility and adaptability in allocation would help to ensure that water is used as productively as possible. Use opportunities change over time and in a world of water scarcity it makes little sense to continue to allocate water to existing, relatively low‐valued uses, as new higher valued uses are emerging. We are not optimistic about the prospects for development of worldwide collective action to conserve and economize on water. There are few, if any, examples of such collective action, and the worldwide approach to global warming illustrates how diYcult the development of such collective action can be. Nevertheless, such collective action could make a big diVerence in addressing the emerging world water scarcity and we would be remiss in failing to mention it.
THE EMERGING GLOBAL WATER CRISIS
VI.
69
CONCLUSIONS
In this chapter we have sought to portray an emerging world water crisis and to identify the global responses that will likely be eVective in addressing it. It is clear that the need to provide water and sanitation services to a large portion of the world’s population and the need to feed a sharply larger population, most of it in the developing world, will place unprecedented strains on the world’s water resources. There is much that can be done in response to manage water in a sustainable fashion. There are many modern management strategies that are not yet employed on an extensive basis. New and emerging technologies and scientific findings will help as well improvements in governing and managing institutions. The problems are daunting but we are not without means to address them. It is important to recognize, however, that water is just one of the challenges that will have to be addressed as the twenty‐first century progresses. As population grows and economic development proceeds there will be threats and crises in many sectors that sustain life and are important to the quality of life. Demands for energy and materials will grow, as will the problems of disposing of their residuals. There will be enormous pressures for provision of housing and education, for food and nutrition, and in protecting and preserving planetary life support systems (National Research Council, 1999c). All these loom as compelling problems that will have to be solved. Thus, water is but one resource and poses but one of many sets of problems that will have to be addressed and managed if a sustainable world with a much larger population is to be created. Just as water is woven through many of the other challenges such as the provision of adequate food and nutrition and the preservation of life support systems, those challenges and how we address them will also have important implications for our success in addressing the emerging water scarcity of the twenty‐first century.
REFERENCES ADB (1997). ‘‘Emerging Asia: Changes and Challenges.’’ Asian Development Bank, Manila, the Philippines. AMTA (American Membrane Technology Association). (2001). ‘‘Desalting Facts: How Much Does Desalted Water Cost?’’ Online at http://www.membranes-amta.org/media/pdf/ desaltingcost.pdf Andreoli, C. V. (1993). The influence of agriculture on water quality. In ‘‘Prevention of Water Pollution by Agriculture and Related Activities.’’ Proceedings of the FAO Expert Consultation, Santiago, Chile, 20–23 Oct. 1992. pp. 53–65. Water Report 1. FAO, Rome.
70
W. A. JURY AND H. J. VAUX
Ayadi, Y. (2006). Policy and adaptation in the Jordan Valley. In ‘‘Proceedings of Fifth International Rosenberg Forum on Water Policy, BanV, Canada.’’ Online at http:// rosenberg.ucanr.org/ Benabdallah, S. (2006). Upland watershed management: Perspectives from the developing world. In ‘‘Proceedings of Fifth Rosenberg International Forum on Water Policy, BanV, Canada.’’ Online at http://rosenberg.ucanr.org/ Borlaug, N. (2002). The Green Revolution revisited and the road ahead. Nobel Peace Prize anniversary lecture. Online at http://nobelprize.org/nobel_prizes/peace/articles/borlaug/index. html Bos, M. G. (1985). Summary of International Commission on Irrigation and Drainage definitions on irrigation eYciency. ICID Bull. 34, 28–31. Brown, L. (2004). ‘‘Outgrowing the Earth.’’ W.W. Norton & Co., New York. Capel, P., Giger, W., Reichert, P., and Wanner, O. (1988). Accidental input of pesticides into the Rhine River. Environ. Sci. Technol. 22, 992–997. Centers for Disease Control. (1996). Spontaneous abortions possibly related to ingestion of nitrate‐contaminated well water—LaGrange County, Indiana, 1991–1999. Morb. Mort. Weekly Rep. 45, 569–572. Clarke, R. (1993). ‘‘Water: The International Crisis.’’ MIT Press, Cambridge, MA. Conservation International. (2006). ‘‘Conservation International Biodiversity Hotspots.’’ Online at www.biodiversityhotspots.org. Constanza, R., D’Arge, R., De Groot, R., Farber, S., Grasso, M., Hannon, B., Limburg, K., Naeem, S., O’Neill, R., Paruelo, J., Raskin, R., Sutton, P., et al. (1997). The value of the world’s ecosystem services and natural capital. Nature 387, 253–260. Davies, J., and Claridge, G. F. (1993). ‘‘Wetland Benefits. The Potential for Wetlands to Support and Maintain Development.’’ International Waterfowl and Wetlands Research Bureau Special Publication No. 27. Doppelt, R., Scurlock, M., Frissell, C., and Karr, J. (1993). ‘‘Entering the Watershed: A New Approach to Save America’s River Ecosystems.’’ Island Press, Washington, DC. Eswaran, H., Lal, R., and Reich, P. F. (2001). Land degradation: An overview. In ‘‘Responses to Land Degradation’’ (E. M. Bridges, I. D. Hannam, L. R. Oldeman, F. W. T. Penning de Vries, S. J. Scherr, and S. Sombatpanit, Eds.). Proceedings of 2nd International Conference on Land Degradation and Desertification, Khon Kaen, Thailand, Oxford Press, New Delhi, India. European Environment Agency (1997). ‘‘Air Pollution in Europe in 1997.’’ OYce for OYcial Publications of the European Communities, Luxembourg, and Elsevier Science, Oxford, United Kingdom. European Environment Agency (1998). ‘‘Europe’s Environment: The Second Assessment.’’ OYce for OYcial Publications of the European Communities, Luxembourg, and Elsevier Science, Oxford, United Kingdom. Falkenmark, M., and Rockstrom, J. (2004). ‘‘Balancing Water for Humans and Nature.’’ Earthscan, London. FAO (1997). ‘‘Irrigation Potential in Africa. A Basin Approach.’’ FAO Land and Water Bulletin 4, Food and Agriculture Organization of the United Nations, Rome. FAO (2003). ‘‘World Agriculture Towards 2015/2030.’’ Summary Report. Food and Agriculture Organization of the United Nations, Rome. FAS (2006). ‘‘Grain: World Markets and Trade.’’ Foreign Agricultural Service Circular Series FG 05–06 US. Department of Agriculture, Washington, DC. Fereres, E., and Soriano, M. A. (2007). Deficit irrigation for reducing agricultural water use. J. Exp. Bot. 58, 147–159. Fisher, F., Huber‐Lee, A., Amir, I., ArlosoroV, S., Eckstein, Z., Haddadin, M., Hamati, G., Jarrar, H., Jayyousi, A., Shamir, U., and Wesseling, H. (2005). ‘‘Liquid Assets: An
THE EMERGING GLOBAL WATER CRISIS
71
Economic Approach for Water Management and Conflict Resolution in the Middle East and Beyond.’’ Resources for the Future Press, Washington, DC. Foulds, K. (2002). The Nile Basin initiative: Challenges to implementation. Paper presented at Managing Shared Waters Conference, Hamilton, Ontario, Canada. Online at http:// nilebasin.com/documents/kim1.htm Frederick, K. (1997). ‘‘Water Resources and Climate Change.’’ Climate Issues Brief No. 3, Resources for the Future, Washington, DC. Freshfields (2003). ‘‘Environmental Liability in Germany.’’ Freshfields Bruchhaus Derringer Environment, Planning and Regulatory publications Online at www.freshfields.com/ practice/environment/publications. Gilliom, R., Barbash, J., Crawford, C., Hamilton, P., Martin, J., Nakagaki, N., Nowell, L., Scott, J., Stackelberg, P., Thelin, G., and Wolock, D. (2006). ‘‘The Quality of Our Nation’s Waters: Pesticides in the Nation’s Streams and Ground Water, 1992–2001.’’ US Geological Survey Circular 1291, US Geological Survey, Reston, VA. Gleick, P. (1996). Basic water requirements for human activities: Meeting basic needs. Water Int. 21, 83–92. Gleick, P. (1998). An introduction to global fresh water issues. In ‘‘Water in Crisis: A Guide to the World’s Fresh Water Resources’’ (P. Gleick, Ed.). Oxford University Press, New York. Gleick, P. (2002). ‘‘Dirty Water: Estimated Deaths from Water‐Related Disease 2000–2020.’’ Pacific Institute Research Report. Online at www.pacinst.org/publications/. Gleick, P. (2006). ‘‘The World’s Water 2006–2007: The Biennial Report on Freshwater Resources’’ (P. Gleick, Ed.). Island Press, Washington, DC. Glennon, R. (2002). ‘‘Water Follies: Ground Water Pumping and the Fate of America’s Fresh Waters.’’ Island Press, Washington, DC. Glover‐Kerkvliet, J. (1995). Environmental assault on immunity. Environ. Health Perspect. 103, 236–239. Goyer, R. A. (1996). Results of lead research: Prenatal exposure and neurological consequences. Environ. Health Perspect. 104, 1050–1055. Greenland, D., Gregory, P., and Nye, P. (1998). Land resources and constraints to crop production. In ‘‘Feeding a World Population of More Than Eight Billion People’’ (J. C. Waterlow, D. G. Armstrong, L. Fowden, and R. Riley, Eds.). Oxford University Press, New York. GWI (2005). ‘‘Water Reuse Markets 2005–2015: A Global Assessment & Forecast.’’ Global Water Intelligence Editorial Team Report, Media Analytics Ltd., Oxford, United Kingdom. Hinrichson, D. (2003). A human thirst. Worldwatch Mag. 16, 12–18. Hoisington, D., Khairallah, M., Reeves, T., Ribaut, J., Skovmand, B., Taba, S., and Warburton, M. (1999). Plant genetic resources: What can they contribute toward increased crop productivity? Proc. Natl. Acad. Sci. USA 96, 5937–5943. Howell, T. A., Schneider, A. D., and Evett, S. R. (1997). Subsurface and surface microirrigation of corn in the Southern High Plains. Trans. ASAE 40, 635–641. International Joint Commission (2006). ‘‘Canada and the United States: Annual Report 2005.’’ Online at www.ijc.gov. IWMI (2006). Beyond more crop per drop. Background paper prepared for 4th World Wide Water Forum, Mexico 2006. International Water Management Institute. Online at www. iwmi.cgiar.org/WWF4/html/action_2.htm Jenerette, G., and Larson, L. (2006). A global perspective on changing sustainable urban water supplies. Glob. Planet. Change 50, 202–211. Jensen, M. E., Rangeley, W. R., and Dieleman, P. J. (1990). Irrigation trends in world agriculture. In ‘‘Irrigation of Agricultural Crops’’ (B. A. Stewartand and D. R. Nielsen, Eds.). American Society of Agronomy, Madison, WI. Jury, W. A., and Vaux, H., Jr. (2005). The role of science in solving the world’s emerging water problems. Proc. Natl. Acad. Sci. USA 102, 15715–15720.
72
W. A. JURY AND H. J. VAUX
Knapp, K. C. (1991). Optimal intertemporal irrigation management under saline, limited drainage conditions. In ‘‘The Economics and Management of Water and Drainage in Agriculture’’ (A. Dinar and D. Zilberman, Eds.). Kluwer Academic Publishers, Norwell, MA. Lampman, W. (1995). Susceptibility of ground water to pesticide and nitrate contamination in predisposed areas of southwestern Ontario. Water Qual. Res. J. Can. 30, 443–468. Law, G., Parslow, R., McKinney, G., and Cartwright, R. (1999). Non‐Hodgkin’s lymphoma and nitrate in drinking water: A study in Yorkshire, United Kingdom. J. Epidemiol. Community Health 53, 383–384. Li, J., and Cao, S. (1994). Recent studies on endemic fluorosis in China. Fluoride 27, 125–128. Libiszewsk, S. (1995). Water disputes in the Jordan Basin region and their role in the resolution of the Arab‐Israeli coflict. Environment and Conflicts Project Occasional Paper No. 13, Center for Security Studies, Zurich. Lutz, W., Sanderson, W., and Scherbov, S. (2001). The end of world population growth. Nature 412, 543–545. Martin‐Lagardette, J. L. (2001). Desalination of sea water. Water Eng. Mgmt. 148, 18–20. McCaVerty, S. (1998). Water, politics, and international law. In ‘‘Water in Crisis: A Guide to the World’s Fresh Water Resources’’ (P. Gleick, Ed.). Oxford University Press, New York. Millennium Ecosystem Assessment (2005). ‘‘Ecosystems and Human Well‐being: Synthesis.’’ Island Press, Washington, DC. Moench, M., Burke, J., and Moench, Y. (2003). ‘‘Rethinking the Approaches to Ground Water and Food Security.’’ Water Reports 24, Food and Agricultural Organization, Rome. Morel‐Seytoux, H. J. (1985). Conjunctive use of surface and ground waters. In ‘‘Artificial Recharge of Groundwater’’ (T. Asano, Ed.). Butterworth Publishers, Stoneham, MA. Moser, M., Prentice, C., and Frazier, S. (1996). A global overview of wetland loss and degradation. In ‘‘Proceedings of the 6th Meeting of the Conference of Contracting Parties of the Ramsar Convention,’’ Vol. 10, March 1996. Naiman, R. J. (1992). ‘‘Watershed Management: Balancing Sustainability and Environmental Change.’’ Springer‐Verlag, New York, NY. National Research Council (2001). ‘‘Climate Change Science: An Analysis of Some Key Questions.’’ National Academy Press, Washington, DC. National Research Council (1994). ‘‘Groundwater Recharge Using Waters of Impaired Quality.’’ National Academy Press, Washington, DC. National Research Council (1995). ‘‘Mexico City’s Water Supply: Improving the Outlook for Sustainability.’’ National Academy Press, Washington, DC. National Research Council (1999a). ‘‘Water for the Future: The West Bank and Gaza Strip, Israel and Jordan.’’ National Academy Press, Washington, DC. National Research Council (1999b). ‘‘New Strategies for America’s Watersheds.’’ National Academy Press, Washington, DC. National Research Council (1999c). ‘‘Our Common Journey: A Transition Toward Sustainability.’’ National Academy Press, Washington, DC. National Research Council (1999d). ‘‘Downstream Adaptive Management of Glen Canyon Dam and the Colorado River.’’ National Academy Press, Washington, DC. National Research Council (2004). ‘‘Review of the Desalination and Water Purification Technology Roadmap.’’ National Academy Press, Washington, DC. Nolan, B., Hitt, K., and Ruddy, B. (2002). Probability of nitrate contamination of recently recharged ground waters in the conterminous United States. Environ. Sci. Technol. 36, 2138–2145. Ohlendorf, H. M., Kilness, A. W., Simmons, J. L., Stroud, R. K., HoVman, D. J., and Moore, J. F. (1988). Selenium toxicosis in wild aquatic birds. J. Toxicol. Environ. Health 24, 67–92. Ongley, E. (1996). Control of water pollution from agriculture—FAO Irrigation and Drainage Paper 55, Food and Agricultural Organization, Rome, Italy.
THE EMERGING GLOBAL WATER CRISIS
73
Pacific Institute (2003). Water Data from the World’s Water, Table V. Online at www. worldwater.org/data.html Perry, E., and Dotson, B. (1996). The treadle pump: An irrigation technology adapted to the needs of small farmers. IPTRID GRID Issue 8. Online at www.hrwallingford.co.uk/ projects/IPTRID/grid/iptgrid.htm. Peterson, B., Holmes, R., McClelland, J., Vorosmarty, C., Lammers, R., Shiklomanov, A. I., Shiklomanov, I. A., and Rahmstorf, S. (2002). Increasing river discharge to the Arctic Ocean. Science 298, 2171–2173. Population Reports (1998). Solutions for a water‐short world. Volume XXVI/1, Johns Hopkins University. INFO Project. Online at www.infoforhealth.org/pr/online.shtml#m Postel, S. (1997). ‘‘Lost Oasis.’’ W.W. Norton & Company, New York. Postel, S. (1998). Water for food production: Will there be enough in 2025? Bioscience 48, 629–635. Postel, S. (1999). ‘‘Pillar of Sand.’’ W.W. Norton & Company, New York. Postel, S., Polak, P., Gonzales, F., and Keller, J. (2001). Drip irrigation for small farmers: A new initiative to alleviate hunger and poverty. Water Int. 26, 3–13. Postel, S. (2005). Liquid assets: The critical need to safeguard freshwater ecosystems.Worldwatch Paper, 170, Worldwatch Institute, Washington, DC. Postel, S., and Carpenter, S. (1997). Freshwater ecosystem services. In ‘‘Nature’s Services: Societal Dependence on Natural Ecosystems’’ (G. C. Daily, Ed.). Island Press, Washington, DC. Postel, S., Daily, G. C., and Ehrlich, P. R. (1996). Human appropriation of renewable fresh water. Science 271, 785–788. Raskin, P., Gleick, P., Kirshen, P., Pontius, R., and Strzepek, K. (1997). ‘‘Comprehensive Assessment of the Freshwater Resources of the World.’’ Document prepared for UN Commission for Sustainable Development, 5th Session, Stockholm Environmental Institute, Sweden. Rawls, J. (1971). ‘‘A Theory of Justice.’’ Harvard University Press, Cambridge, MA. Revenga, C., Brunner, J., Henninger, N., Kassem, K., and Payne, R. (2000). ‘‘Pilot Analysis of Freshwater Ecosystems: Freshwater Systems.’’ World Resources Institute, Washington, DC. Reynolds, S. E. (1962). ‘‘Twenty‐Fifth Biennial Report of the State Engineer of New Mexico.’’ The Valliant Company, Albuquerque, New Mexico. Rijsberman, F. (2004). Sanitation and access to clean water. In ‘‘Global Crises, Global Solutions’’ (B. Lomborg, Ed.). Cambridge University Press, Cambridge, UK. RIVM (1992). ‘‘The Environment in Europe: A Global Perspective.’’ National Institute of Public Health and Environmental Protection (RIVM), Bilthoven, The Netherlands. Rockstrom, J., Gordon, L., Falkenmark, M., Folke, C., and Engvall, M. (1999). Linkages among water vapor flows, food production and terrestrial ecosystems services. Conserv. Ecol. 3, 1–28. Rockstrom, J., Barron, J., and Fox, P. (2002). Rainwater management for increased productivity among small‐holder farmers in drought prone environments. Phys. Chem. Earth 27, 949–959. Rockstrom, J., Lannerstad, M., and Falkenmark, M. (2007). Assessing the water challenge of a new green revolution in developing countries. Proc. Natl. Acad. Sci. USA 104, 6253–6260. Rosegrant, M., Cai, X., and Cline, S. (2002). ‘‘Global Water Outlook to 2025: Averting an Impending Crisis.’’ International Food Policy Research Institute Food Policy Report, International Water Management Institute, Colombo, Sri Lanka. Sandia National Laboratories (2002). U.S. Bureau of Reclamation Report to Congress: Tularosa Basin National Desalination Research Facility Study online at http://www.sandia.gov/ water/docs/TBrpt0203ev1.pdf. Schneider, A. D., and Howell, T. A. (1996). Methods, amounts, and timing of sprinkler irrigation for winter wheat. Trans. ASAE 40, 137–142.
74
W. A. JURY AND H. J. VAUX
Seckler, D., Amarasinghe, U., Molden, D., de Silva, R., and Barker, R. (1998). ‘‘World Water Demand and Supply 1990 to 2025: Scenarios and Issues.’’ International Water Management Institute Research Report 19, International Water Management Institute, Colombo, Sri Lanka. Seckler, D., Barker, R., and Amarasinghe, U. (1999). Water scarcity in the twenty‐first century. Int. J. Wat. Res. Dev. 15, 29–42. Seiler, R. L., Skorupa, J., and Peltz, L. (1999). ‘‘Areas Susceptible to Irrigation‐Induced Selenium Contamination of Water and Biota in the Western United States.’’ US Geological Survey Circular 1180, US Geological Survey, Reston, VA. Shah, T. (2000). ‘‘Wells and Welfare in the Ganga Basin: Public Policy and Private Initiative in Eastern Uttar Pradesh India.’’ IWMI Research Report No. 54. International Water Management Institute, Colombo, Sri Lanka. Shah, T. (2003). Decentralized water harvest and groundwater recharge: Can these save Saurashtra and Kutch from dessication? Anand, India. IWMI‐Tata Water Policy Programs, Colombo, Sri Lanka. Online at www.iwmi.cgiar.org/iwmi-tata_html/PartnersMeet/papers/ Tushaar.htm Shah, T., Alam, M., Kumar, M. D., Nagar, R. K., and Singh, M. (2000a). ‘‘Pedaling Out of Poverty: Socio‐economic Impact of a Manual Irrigation Technology in South Asia.’’ IWMI Research Report 45, International Water Management Institute, Colombo, Sri Lanka. Shah, T., Molden, D., Sakthivadivel, R., and Seckler, D. (2000b). ‘‘The Global Ground Water Situation: Overview of Opportunities and Challenges.’’ International Water Management Institute, Colombo, Sri Lanka. Shaltout, M., and El Housry, T. (1996). Estimating the evaporation over Nasser Lake from METEOSAT observations. In ‘‘ESA SP‐394: Space Mission Operations and Ground Data Systems’’ (T. D. Guyenne, Ed.). European Space Agency, Paris. Shiklomanov, I. A. (1997). ‘‘Comprehensive Assessment of the Fresh water Resources of the World.’’ World Meteorological Organization and Stockholm Environment Institute, Stockholm, Sweden. Shiklomanov, I. A. (2000). Appraisal and assessment of world water resources. Water Int. 25, 11–32. Smakhtin, V., Revenga, C., and Doll, P. (2004). ‘‘Taking into Account Environmental Water Requirements in Global‐Scale Water Resources Assessments.’’ Comprehensive Assessment Research Report 2, Comprehensive Assessment Secretariat, Colombo, Sri Lanka. Smedley, P. L., and Kinniburgh, D. G. (2005). Arsenic in ground water and the environment. In ‘‘The Essentials of Medical Geology: Impacts of the Natural Environment on Public Health’’ (O. Selinus, Ed.). Elsevier, Boston. Smil, V. (1999). China’s great famine: 40 years later. Br. Med. J. 319, 1619–1621. Smil, V. (2000). ‘‘Feeding the World: A Challenge for the 21st Century.’’ MIT Press, Boston. Smith, A. H., Lingas, E. O., and Rahman, M. (2000). Contamination of drinking‐water by arsenic in Bangladesh: A public health emergency. Bull. World Health Organ. 78, 1093–1103. Sullivan, C. A. (2006). Do investments and policy interventions reach in the poorest of the poor. In ‘‘Water Crisis: Myth or Reality’’ (P. Rogers, M. R. Llamas, and L. Martinez‐Cortina, Eds.). Taylor and Francis, London, UK. US Census (2006). IDB Data Access Spreadsheet. Online at www.census.gov/ipc/www/idbsprd. html US EPA (1992). ‘‘EPA National Pesticide Survey: Update and Summary of Phase II Results.’’ EPA 570/9–91–021, Washington, DC. US EPA (2006). ‘‘Ground Water: Basic Information‐Major Laws.’’, Online at www.epa.gov/ compliance/basics/majorlaws/groundwater.html UN (2006). ‘‘The Millennium Development Goals Report 2006.’’ United Nations, New York.
THE EMERGING GLOBAL WATER CRISIS
75
UNEP (1996). ‘‘Groundwater: A Threatened Resource.’’ Environment Library No. 15, United Nations Environment Program, Nairobi, Kenya. UNEP (2006). ‘‘Challenges to International Waters—Regional Assessments in a Global Perspective.’’ United Nations Environment Program, Nairobi, Kenya. UNICEF (1999). ‘‘Fluoride in Water: An Overview.’’ UNICEF, New York.? Online at www. unicef.org/programme/wes/info/fluor.htm UNIDO (1996). ‘‘Global Assessment of the Use of Freshwater Resources for Industrial and Commercial Purposes.’’ Industry, Sustainable Development, and Water Program Formulation, United Nations Industrial Development Organization Technical Report UNIDO/ ISED R53, United Nations, New York. United Nations Development Programme (2003). ‘‘Human Development Report 2003.’’ United Nations Development Programme, New York, NY. United Nations Task Force on Water and Sanitation (2003). ‘‘Achieving the Millennium Development Goals for Water and Sanitation: What Will It Take?’’ Interim Full Report of Task Force 7. Online at www.unngls.org/MDG/Goal_7.htm Vaux, H. J., Jr. (1991). Global climate change and California’s water resources. In ‘‘Global Climate Change and California: Potential Impacts and Resources’’ (J. Knox and A. F. Scheuring, Eds.). University of California Press, Berkeley, CA. Vaux, H. J., Jr. (2004). Managing water scarcity in Western North America. In ‘‘Proceedings of Lethbridge Conference on Water Scarcity.’’ University of Lethbridge, Lethbridge, Canada. Vaux, H. J., Jr. (2007). The Economics of ground water resources and the American experience. In ‘‘The Global Importance of Groundwater in the 21st Century’’ (S. Ragone, N. Herna´ndez‐Mora, A. de la Hera, G. Bergkamp, and J. McKay, Eds.). Proceedings of the International Symposium on Groundwater Sustainability, National Groundwater Association Press, Columbus, OH. Wallace, J. S., and Batchelor, C. H. (1997). Managing water resources for crop production. Phil. Trans. R. Soc. Lond. B 352, 937–947. Weiss, E. B. (1989). ‘‘In Fairness to Future Generations: International Law, Common Patrimony and Intergenerational Equity.’’ United Nations University Press, Tokyo, Japan and Dobbs Ferry, New York. Weyer, P., Cerhan, J., Kross, B., Hallberg, G., Kantamneni, J., Breuer, G., Jones, M., Zheng, W., and Lynch, C. (2001). Municipal drinking water nitrate level and cancer risk in older women: The Iowa Women’s Health Study. Epidemiology 12, 327–338. WHO (1990). ‘‘Public Health Impact of Pesticides Used in Agriculture.’’ World Health Organization, Geneva. WHO (1993).‘‘Guidelines for Drinking‐Water Quality,’’ Vol. 1, 2nd ed. World Health Organization, Geneva. WHO (2004). ‘‘Water, Sanitation and Hygiene Links to Health: Facts and Figures.’’ World Health Organization, Geneva. Online at www.who.int/water_sanitation_health/ publications/facts2004/en/ WHO/UNICEF (2000). ‘‘Global Water Supply and Sanitation Assessment 2000 Report.’’ World Health Organization, Geneva. Online at www.who.int/water_sanitation_health/monitoring/ globalassess/en/ Wolf, A. T. (1998). Conflict and cooperation along international waterways. Water Policy 1, 251–265. Wolf, A., Natharius, J., Danielson, J., Ward, B., and Pender, J. (1999). International river basins of the world. Int. J. Wat. Res. Dev. 15, 387–427. Wood, S., Sebastian, K., and Scherr, S. J. (2001). ‘‘Pilot Analysis of Global Ecosystems: Agroecosystems.’’ World Resources Institute, Washington, DC. World Development Report (1993). ‘‘Investing in Health.’’ Oxford University Press, Oxford, UK.
76
W. A. JURY AND H. J. VAUX
Yang, H., and Zehnder, A. (2002). Water scarcity and food import: A case study for Southern Mediterranean countries. World Dev. 30(8), 1413–1430. Yang, H., Reichert, P., Abbaspour, K., and Zehnder, A. (2003). A water resources threshold and its implications for food security. Environ. Sci. Technol. 37(14), 3048–3054. Young, R. A., and Haveman, R. H. (1985). Economics of water resources: A survey. In ‘‘Handbook of Natural Resource and Energy Economics’’ (A. V. Kneese and J. V. Sweeney, Eds.). Vol. II, Elsevier Science Publishers, New York, NY. Zhang, H. (2003). Improving water productivity through deficit irrigation: Examples from Syria, the North China Plain and Oregon, USA. In ‘‘Water Productivity in Agriculture: Limits and Opportunities for Improvement’’ (J. W. Kijne, R. Barker, and D. Molden, Eds.). CABI Publishing, Wallingford, UK.
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE Richard A. Dixon,1 Joseph H. Bouton,2 Brindha Narasimhamoorthy,2 Malay Saha,2 Zeng‐Yu Wang2 and Gregory D. May3 1
Plant Biology Division, Samuel Roberts Noble Foundation, 2510 Sam Noble Parkway, Ardmore, Oklahoma 73401 2 Forage Improvement Division, Samuel Roberts Noble Foundation, 2510 Sam Noble Parkway, Ardmore, Oklahoma 73401 3 National Center for Genome Resources, 2935 Rodeo Park Drive East, Santa Fe, New Mexico 87505
I. Introduction II. Sequenced Genomes, Model Systems, and Comparative Genomics A. Introduction B. A. thaliana C. Rice D. Poplar E. Medicago truncatula and Lotus japonicus F. Genetic Resources for Functional Genomics III. Transcriptomics, Proteomics, and Metabolomics A. Introduction B. Approaches for Transcript Profiling C. Proteomics D. Metabolomic Analysis IV. Molecular Markers A. Marker Types B. Molecular Genetic Maps C. Genomics for Generation of Molecular Markers D. Metabolomic‐Based ‘‘Markers’’ E. Advantages of Marker‐Assisted Breeding V. Transgenesis A. Transgenesis as a Tool for Functional Genomics B. Current Approaches to the Generation of Transgenic Plants C. Strategies for Overcoming Recalcitrance of Crop Species to Genetic Transformation D. Transgenesis for Trait Integration and Commercialization E. Virus‐Induced Gene Silencing as an Alternative to Stable Transformation for Functional Genomics F. TILLING as an Alternative to Transgenesis for Gene Knockdowns VI. Case Studies for Alfalfa Improvement A. Introduction 77 Advances in Agronomy, Volume 95 Copyright 2007, Elsevier Inc. All rights reserved. 0065-2113/07 $35.00 DOI: 10.1016/S0065-2113(07)95002-6
78
R. A. DIXON ET AL. B. Improvement of Aluminum Tolerance C. Gene Discovery and Metabolic Engineering for Forage Quality Enhancement D. Issues for Molecular Development of Alfalfa VII. The Future: Bridging the Gap from Models to Crops VIII. The Future Technologies, Opportunities, and Challenges Acknowledgments References
The past decade has seen unparalleled advances in our understanding of plant genomes, and genomic (primarily DNA sequence) information now underpins many aspects of plant trait improvement, through gene discovery to transgenesis and use of molecular markers in breeding. This chapter provides an overview of the genomic and postgenomic technologies that are likely to have the greatest impacts on agronomy over the next 10–20 years and describes a number of case studies of their application. Although the impacts of these technologies are already apparent, the amazing and still accelerating pace of technology development promises much, maybe more than can easily be assimilated into traditional plant improvement programs at present. A new breed of plant scientist with skills in understanding and integrating multiple disciplines, and making use of increasingly sophisticated computational approaches, is needed to take full advantage of even the present knowledge. # 2007, Elsevier Inc.
I. INTRODUCTION The past several years have seen major advances in our ability to gather whole‐genome‐scale information from plants. Central to these developments, several projects have assembled working models of the complete or near complete genome sequences of the model crucifer Arabidopsis thaliana (Arabidopsis Genome Initiative, 2000; Bevan et al., 2001), rice (GoV et al., 2002; Yu et al., 2002), poplar (Tuskan et al., 2006), and two model legumes (VandenBosch and Stacey, 2003). Several other projects have targeted a range of species for the sequencing of expressed sequence tags (ESTs) representing genes expressed in particular tissues or under particular developmental or environmental conditions. For species with either sequenced genomes or extensive EST resources, commercial DNA microarrays are now available for global transcript profiling (Rensink and Buell, 2005). Technologies such as serial analysis of gene expression (SAGE; Matsumura et al., 1999; Velculescu, 1999), massively parallel signature sequencing (MPSS; Brenner et al., 2000), and cDNA‐AFLP (Goossens et al., 2003) provide tools for analysis of genome‐wide changes in
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
79
transcriptional activity for plant species that lack sequenced genomes or even extensive EST resources. However, it has been increasingly realized that profiling changes in gene transcripts only provides part of the picture of the impacts of diVerential gene expression in plants (Hall et al., 2002), and considerable eVorts have therefore been put into developing robust and comprehensive methods for profiling the metabolome (the metabolite complement of a tissue, organ, or whole organism; Hall et al., 2002) and the proteome (the complete set of proteins; Watson et al., 2003). These latter technologies rely heavily on mass spectrometry (Roessner et al., 2001; Yates, 1998). The integration of data spanning the genome, proteome, and metabolome is a major goal of ‘‘postgenomics’’ biology. A whole new field of ‘‘Systems Biology’’ has been defined, encompassing the collection and interpretation of holistic data for biological systems. The goal is to understand the organizational principles that operate at the cellular and organismal levels and that relate individual components to the whole system. This new way of thinking about biological systems poses some major challenges, none more so than in the area of gene annotation. As discussed by Huang (2000), most gene products function as part of one or more complex regulatory systems, and exactly how they do this is often not apparent from the types of in vitro (enzymatic, interaction mapping) or simple genetic analysis currently utilized. Although postgenomics biology faces major challenges in taking the next step to a full understanding of gene function at the organismal level, the techniques associated with this branch of biology are already finding application in agronomy. For example, holistic analysis of genome content and gene expression provides novel approaches to the design of markers for breeding and speeds the identification and isolation of genes associated with important traits. Metabolome analysis also holds promise as a tool for trait mapping (Dixon et al., 2006). In contrast to the exponential increase in genomic information, improvements in the eYciency of plant genetic transformation have occurred linearly over the past 10 years. Nevertheless, public perception and regulatory issues aside, transgenesis provides the most rapid means of introducing truly novel traits to crop plants, and is also a major technology for functional genomics in plants. This chapter provides only a brief overview of the technical bases of the important new genomic and postgenomic technologies. Our major aim is to present the reader with a feeling for how the genomics revolution is set to impact plant science in general, and agronomy in particular, over the next 10–20 years. To provide case studies of the uses of several of the outlined technologies, we describe four projects in Sections VI and VII, three ongoing at the Noble Foundation, in which genomic approaches have been utilized for introducing important agronomic traits into legumes. Much of the work applying postgenomic technologies to agriculture requires collaboration and understanding between scientists with quite
80
R. A. DIXON ET AL.
diVerent academic backgrounds. Multidisciplinary approaches of this type will play an increasing role in basic and applied agronomy in the future.
II. SEQUENCED GENOMES, MODEL SYSTEMS, AND COMPARATIVE GENOMICS A. INTRODUCTION By definition, postgenomics technologies take as their starting point the availability of genome‐wide information for any particular target species. The partially annotated sequence of the model crucifer A. thaliana first appeared in 2000 (Arabidopsis Genome Initiative, 2000) as the result of a major international eVort that took nearly 10 years to bring to fruition. Arabidopsis was chosen primarily for its small genome size, self‐fertility, short time to flowering and seed set, and the availability of many naturally occurring ecotypes (geographical variants of the species; KoornneV et al., 2004). While the sequencing program was ongoing, dramatic improvements in the genetic transformation eYciency of Arabidopsis (Clough and Bent, 1998) accelerated the rate of development of genetic resources through T‐DNA tagging based on random insertion of the Agrobacterium tumefaciens transfer DNA following transformation (Alonso et al., 2003; Azpiroz‐Leehan and Feldmann, 1997). The combination of a sequenced genome and availability of a range of genetic resources such as defined ecotypes or mutant populations derived from chemical or DNA‐based (deletion or insertion) mutagenesis greatly facilitates the use of the model system for gene discovery and annotation. However, this does not mean that postgenomics technologies can only be applied to species with sequenced genomes. A number of approaches can be used for global transcript profiling, mining mutations, and developing molecular markers in species that do not have extensive genome, or even EST, sequence information. Furthermore, comparative genomic approaches that link genomic data from less well‐defined systems to the well‐defined model species are becoming increasingly useful for gene discovery.
B. A. THALIANA Several articles have reviewed the strategies used for sequencing the Arabidopsis genome (Arabidopsis Genome Initiative, 2000; Bevan et al., 1999, 2001). Essentially, the approach relied on a robust physical map of yeast artificial chromosome clones of genomic DNA fragments to ‘‘place’’ the emerging sequence in its genomic context, since this was primarily
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
81
obtained by sequencing bacterial artificial chromosomes harboring Arabidopsis genomic DNA that had been anchored to the physical map. Whole‐ genome ‘‘shotgun’’ sequencing was not attempted for Arabidopsis, since at the time this approach had several drawbacks, particularly as regards final sequence assembly, especially of the highly repetitive heterochromatic regions around the centromeres. Advances that greatly increase the power and throughput of both sequencing technology and computational analysis of genome sequence information have since made the shotgun approach more feasible, and this approach was therefore taken as the principal technique for obtaining the genome sequences of rice and poplar (GoV et al., 2002; Tuskan et al., 2006; Yu et al., 2002). In the Arabidopsis project, as in all subsequent large‐scale genome projects, the availability of a large set of EST sequences was invaluable for genome assembly and annotation. Table I summarizes the current status of plant genome sequencing projects.
Table I A Summary of Plant Species Genome Projects Chromosome number (n)
Genome size (Mbp)
Number of TIGR ESTs Project status
5
120
616,064 Completea
CoVee Soybean
11 20
640 1115
Lotus Cassava
6 18
470 765
148,617 Near Complete 17,910 Initiated
8
500
217,148 Near Complete
Rice
12
430
1,169,591 Completeb
Black cottonwood
19
550
Completec
Tomato
12
950
200,248 Initiated
Potato
12
840
219,485 Initiated
Sorghum Maize
10 10
760 2300
203,575 Initiated 1,014,701 Initiated
Species
Common name
Arabidopsis thaliana CoVee arabica Glycine max cv Williams 82 Lotus japonicus Manihot esculenta Medicago truncatula cv Jemalong A17 Oryza sativa cv Nipponbare Populus trichocarpa cv Nisqually‐1 Lycoperiscon esculentum Solanum tuberosum Sorghum bicolor Zea mays cv B73
Thale cress
a
Barrel medic
Arabidopsis Genome Initiative, 2000. Yu et al., 2002. c Tuskan et al., 2006. b
1064 Initiated 351,935 Initiated
82
R. A. DIXON ET AL.
Of the more than 25,000 genes predicted to be present in Arabidopsis, only 9% had been characterized experimentally by 2001 (Bevan et al., 2001), and 30% could not be assigned to any putative functional category based on sequence identity alone. These 25,000 genes (more recently updated to closer to 29,000) represented over 11,000 diVerent protein types, and 35% of the predicted proteins occurred only once in the genome. The US National Science Foundation initiated a program, the ‘‘2010 Program,’’ with the aim of understanding the functions of all the genes in the Arabidopsis genome by the year 2010 (Chory et al., 2000). Awardees study specific gene families through combined approaches such as expression and analysis of recombinant proteins and characterization of the phenotypes of knockout mutations in the target genes. There have been regular reports of overall progress, which has been significant (Ausubel, 2002; Chory et al., 2000), but it appears unlikely that the functions of all the genes will be understood within the next 4 years. The initial Arabidopsis genome sequence yielded several surprises. For example, at least 47 expressed genes that encoded a wide variety of diVerent protein types were found within the highly repetitive centromeric regions, gene families containing two or more members arranged in tandem arrays were common, and 60% of the genome was present in 24 duplicated segments, each of more than 100 kb, suggesting that Arabidopsis may have had a tetraploid ancestor. Generation of full‐length cDNA resources and application of DNA tiling array technology (Section III.B.1) has revealed a significant number of previously unsuspected genes in the Arabidopsis genome, many of which are transcribed but do not appear to code for proteins (Yamada et al., 2003).
C. RICE After the success in sequencing Arabidopsis, it was fitting that the next plant genome to be sequenced should be a monocot, and furthermore the world’s major staple crop. The draft sequence of the rice genome was reported in 2002, for both the indica and japonica subspecies (GoV et al., 2002; Yu et al., 2002). The euchromatic portion of the rice genome was estimated to be 430 Mb, some 3.7 times larger than that of Arabidopsis. Similar to Arabidopsis, an apparent whole‐genome duplication has occurred in rice, in this case 40–50 million years ago. The high degree of synteny among grass genomes (Freeling, 2001; Gale and Devos, 1998; GoV et al., 2002), coupled with the ease of rice transformation (Tyagi et al., 1999), excellent physical and genetic maps, and availability of mutant resources (Hirochika et al., 2004), make rice an excellent model for other monocot crops. Indica rice is predicted to contain 54,000 genes, of which only about
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
83
20% could be given a functional classification based on sequence alone (Yu et al., 2002). Approximately 85% of the predicted Arabidopsis proteins have significant homologues in rice, with overall mean identity of about 50%. Nevertheless, a significant number of Arabidopsis genes, most without functional annotation, are not present in rice and may represent dicot‐specific genes. In contrast, most cereal genes discovered to date have very close homologues in rice; homologues of 98% of the maize, wheat and barley protein‐coding genes known in 2002 were found in the rice genome (GoV et al., 2002). This observation, coupled with the close synteny among cereal genomes, makes rice a valuable ‘‘scaVold’’ or nodal species for assembly of other economically important cereal genomes such as wheat (Triticum aestivum), barley (Hordeum vulgare), corn (Zea mays), and sorghum (Sorghum bicolor) (Devos and Gale, 2000; GoV et al., 2002).
D. POPLAR The publication of a draft sequence of the poplar (Populus trichocarpa) genome in 2006 provided the first insights into the genomic organization of a tree species (Tuskan et al., 2006). As with rice, the approach taken, by an international consortium, involved whole‐genome shotgun sequencing and assembly, integrated with detailed genetic mapping. The Populus genome size is estimated to be 485 Mbp, of which about 70% appears to be euchromatic. The 75 Mbp of heterochromatic DNA remained unassembled. A significant proportion of the Populus genome appears to have arisen from a major genome duplication event. Poplar contains more than 45,000 putative coding genes, with similar frequencies of protein domains to those found in Arabidopsis, but a higher number of Populus homologues for each Arabidopsis gene. This is particularly apparent for genes involved in cell wall (lignocellulose) biosynthesis and defense (Tuskan et al., 2006).
E. MEDICAGO TRUNCATULA AND LOTUS JAPONICUS As a family, legumes are unique in their ability to fix atmospheric nitrogen through their association, in root nodules, with nitrogen‐fixing bacteria (Rhizobia; Downie, 1997; Shanmugam et al., 1978; Stacey et al., 2006). From an agronomic perspective, legumes crops can be divided into the grain legumes, such as soybean, bean, and pea, and the forage legumes, such as alfalfa and clover. There was considerable debate in the late 1990s as to the best model species for legume genomics. Although considerable genetic resources were available in the above‐mentioned grain legumes, the
R. A. DIXON ET AL.
84
sizes of their genomes, coupled with recalcitrance to genetic transformation, argued against their adoption. Eventually, one forage legume, Medicago truncatula (Cook, 1999; May and Dixon, 2004; Oldroyd and Geurts, 2001), and one leguminous weed, Lotus japonicus (Kawasaki and Murakami, 2000; Udvardi et al., 2005), were selected as model species. As with Arabidopsis, a small diploid genome (Table I), self‐fertility, rapid generation time, and availability of genetic transformation (although at nothing like the frequency achievable using the floral dip method with Arabidopsis) were the factors driving these choices. M. truncatula is very closely related to alfalfa (Medicago sativa), the world’s major forage legume, which is, however, not itself useful as a model species, being an outcrossing autotetraploid. Initially, progress with legume genomics was primarily in the area of EST sequencing (Asamizu et al., 2000), although whole‐genome projects are now well underway in both M. truncatula (in the United States and Europe) and L. japonicus (in Japan) (VandenBosch and Stacey, 2003), and reports of the full draft sequences are expected within the next 12 months. Soybean (Glycine max) has also joined the list of legume species for which genome projects are underway (Jackson et al., 2006). Extensive EST resources are also available for soybean, and for other legume species that are subjects of more modest genomics projects (VandenBosch and Stacey, 2003; Table II). In some cases, more limited EST projects have targeted specific metabolic processes, such as the biosynthesis of storage polysaccharides in guar (Cyamopsis tetragonoloba; Dhugga et al., 2004; Naoumkina et al., 2007). Surprisingly, many of the World’s most important crop legume species lack substantial levels of EST resources (Table II). Syntenic relationships exist among legume genomes. For example, linkage group V of M. truncatula exhibits macrosynteny with linkage groups V and I
Table II Crop Legume EST Totals in GenBank, as of January 2006 Common name
Species
EST totals
Soybean Common bean Alfalfa Pea Peanut White lupin Chickpea Pigeon pea White clover Lentil Broad bean
Glycine max Phaseolus vulgaris Medicago sativa Pisum sativum Arachis hypogaea Lupinus albus Cicer arietinum Cajanus cajan Trifolium repens Lens culinaris Vicia faba
356,808 21,377 6613 3035 2171 2128 724 55 31 8 1
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
85
of alfalfa and pea, respectively (Cook, 1999). However, this does not extend to comparisons between Medicago and Arabidopsis, which are nevertheless quite closely related within the dicot subclass Rosidae. In spite of the lack of macrosynteny between Medicago and Arabidopsis, marker colinearity is often observed over small genetic distances (Zhu et al., 2003). Medicago, Lotus, and soybean share a genome duplication event that occurred 54 million years ago (Mudge et al., 2005; Pfeil et al., 2005). This genome duplication occurred after the divergence of the Populus lineage from legumes, but before the divergence of Medicago/Lotus and soybean (Cannon et al., 2006).
F. GENETIC RESOURCES FOR FUNCTIONAL GENOMICS The value of Arabidopsis as a model system comes from the ability to combine genomic sequence with genetic resources, and in this respect Arabidopsis is probably the best model plant system. Most importantly, the gene space has been nearly saturated with over 225,000 random Agrobacterium transferred DNA (T‐DNA) insertion events, and the precise locations of the insertions in more than 20,000 of the Arabidopsis genes determined (Alonso et al., 2003). Thus, loss‐of‐function mutants can be readily found for most of the Arabidopsis genes. Furthermore, if a gene from another species (e.g., an important crop) has a close orthologue in Arabidopsis, its function can initially be deduced by study of the phenotype of the corresponding Arabidopsis knockout, or by complementation of the Arabidopsis mutant with the gene from the crop species. Among the many examples of this approach are confirmation of function of a rice ethylene‐signaling component (Mao et al., 2006), a soybean jasmonate signaling component (Wang et al., 2005b), a maize cell division regulator (Lim et al., 2005), and a maize ABA signaling gene (Suzuki et al., 2001). Gain‐of‐function mutants are also available in Arabidopsis, at a lower frequency than the knockouts, from activation tagging projects in which the T‐DNA insert contains multiple 35S enhancer sequences at the right border. Integration of the T‐DNA construct within a gene can lead to a knockout, but integration near to a gene can result in the overexpression of that gene, irrespective of the orientation of the enhancer sequences relative to the transcription start site of the proximal gene (Weigel et al., 2000). A good example of this approach is the discovery of the producer of anthocyanin pigment (PAP1) mutant in which a MYB transcription factor (TF) controlling anthocyanin pigment formation is ectopically expressed as a result of the integration of proximal enhancer sequences (Borevitz et al., 2000). The value of this discovery for the development of a ‘‘bloat‐safe’’ alfalfa is described in Section VI.C.
86
R. A. DIXON ET AL.
The Medicago research community has developed a number of genetic resources to assist in gene discovery and functional annotation of legume genes. These include various populations of mutants. The first were produced by classical chemical mutagenesis of a polymorphic ecotype (A17; Penmetsa and Cook, 2000). Use has been made of fast neutrons to generate chromosomal deletions, and large populations of fast neutron deletion lines are now being generated (Wang et al., 2006). In addition, following the demonstration that the transgenically inserted tobacco retrotransposon Tnt1 could be activated in M. truncatula following tissue culture, and therefore be used for insertional mutagenesis (d’Erfurth et al., 2003), eVorts have been put in place to develop transposon‐tagged populations of Medicago (Tadege et al., 2005). With both approaches, the goal is to facilitate forward and reverse genetic screens. Although fast neutron deletions are very easy to generate and access through forward genetic screens for altered phenotype, cloning the deleted genes is less straightforward than in transposon‐ or T‐DNA‐tagged lines, and until recently has required map‐based cloning. Success has now been reported for cloning genes based on comparisons of transcript abundance between wild‐type and mutant lines using microarrays (Mitra et al., 2004), which opens up possibilities for eYcient gene identification through readily generated genetic resources. Targeting induced local lesions in genomes (TILLING) is a new genetic tool for identifying genetic variation at the single base pair level (HenikoV et al., 2004). It is a nontransgenic reverse genetics approach for identifying novel genetic variations. Reverse genetic screens using mutant populations have utilized TILLING in L. japonicus (Perry et al., 2003), and a similar approach is being taken in M. truncatula (VandenBosch and Stacey, 2003). Without any prior knowledge of gene products, TILLING can investigate functions of a gene of interest in potentially any crop, and thus it is a useful tool for functional genomics. TILLING uses DNA pools from chemically mutagenized plants, and relies on the ability of an endonuclease (CEL1 from celery) to detect mismatches in heteroduplexes formed between wild‐type and mutant PCR products of a specific sequence (McCallum et al., 2000). Note that TILLING requires the generation of a mutant population, but does not require additional resources such as a sequenced genome or DNA arrays. PCR‐based approaches may also be adaptable for the rapid reverse genetic screening of pooled fast neutron deletion populations to provide a rapid route to identification of individual plants harboring deletions in specific genes (Wang et al., 2006). Similarly, a database of transposon‐flanking sequences should be developed to provide a reverse genetic resource based on transposon insertion lines (Tadege et al., 2005). Hopefully, all these resources will be in place for Medicago by the time the genome sequence is completed, thereby facilitating functional annotation of a legume genome.
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
III.
87
TRANSCRIPTOMICS, PROTEOMICS, AND METABOLOMICS A. INTRODUCTION
Global transcript profiling has become one of the most popular tools for analysis of plant gene expression, and this revolution has been driven primarily through the development of DNA microarray technology. The transcriptomes of plant species without genomics resources can also be interrogated, on a hitherto unprecedented scale, through the use of diVerential display or serial sequencing procedures. Being able to determine how an external stimulus, or endogenous developmental factor, regulates gene expression at the scale of the whole‐genome provides a powerful tool for gene discovery and for understanding transcriptional networks. Technically speaking, transcript‐profiling approaches are easier and higher throughput than massively parallel analysis of proteins (proteomics) or metabolites (metabolomics), and this explains their popularity and preponderance as analytical tools. This does not, however, mean that analyzing the transcriptome is more informative than proteomics or metabolomics. Indeed, changes in transcript levels are often quite transient, whereas the longer half‐lives of proteins and metabolites give a more balanced and integrated ‘‘readout’’ of the biochemical phenotype of an organism.
B.
APPROACHES FOR TRANSCRIPT PROFILING 1. For Sequenced Genomes
ESTs are rapidly generated, single pass sequences of cDNAs. Many sources of EST sequence information for plants are available online. Table III summarizes available resources for legumes, and also includes information on transcriptome and proteome information. Table II provides an indication of the numbers of ESTs sequenced for various legume species as of January 2006. The various ‘‘Gene Indices’’ available through the Institute for Genome Research (TIGR; http://www.tigr.org/tdb/tdb.html) [now the Dana‐Farber Cancer Institute (DFCI)] are among the most extensive and user‐friendly sources of EST information (Quackenbush et al., 2000). Simply mining these data online can provide a rapid, first‐pass analysis of the expression profile of a particular gene of interest. This is because ESTs are derived from transcripts sequenced from cDNA libraries that represent a particular tissue type, or tissue subjected to a specific biotic or abiotic stress. As an example, the current M. truncatula Gene Index (MtGI) contains over
R. A. DIXON ET AL.
88
Table III Legume Genomic Resources Database (URL)
Data types
Represented organism(s)
The Legume Information System (comparative‐legumes.org/) The Institute for Genomics Research (www.tigr.org) NCBI (www.ncbi.nlm.nih.gov)
EST, genome, QTL, and comparative maps EST, genome, repeat sequence, and pathways EST, genome, and expression
MtDB–CCGB (www.medicago. org/MtDB/) Medicago EST Navigation System (MENS) (http://medicago. toulouse.inra.fr/Mt/EST/) OpenSputnik Comparative genomics platform (http://sputnik.btk.fi/ests) PlantGDB (www.plantgdb.org)
EST and genome
Glycine, Medicago, Lotus, and Phaseolus Glycine, Medicago, and Lotus Glycine, Medicago, M. sativa, Lotus, and Phaseolus Medicago
EST and pathways
Medicago
EST, BLAST, and SNP
Glycine, Medicago, Lotus, and Phaseolus
EST and BLAST
Glycine, Medicago, M. sativa, Pisum, Arachis, and Phaseolus Glycine
SoyBase (http://soybase.ncgr.org)
Sequencing M. truncatula, University of Oklahoma (http://www.genome.ou.edu/ medicago.html) Medicago Genome Database (http://mips.gsf.de/projects/ medicago) Kazusa Lotus japonicus (www. kazusa.or.jp/lotus/) M. truncatula Consortium (www. medicago.org/genome/) Soybean Functional Genomics (Vodkin) (http://soybeangenomics.cropsci. uiuc.edu/) Soybean Genomics and Microarray Database (http://psi081.ba.ars. usda.gov/SGMD/Default.htm) Noble Foundation (Sumner) (www. noble.org/2DPage/Search.asp) M. truncatula Functional Genomics and Bioinformatics (http:// medicago.vbi.vt.edu/) Mt Proteomics (http://www. mtproteomics.fr.st/)
EST, genome, QTL and genetic maps, pathways, germplasm Genome and BLAST
MIPS genome
M. truncatula
M. truncatula
EST, genome, and genetic Lotus map Linkage maps, BAC overlap M. truncatula and clone/marker data Transcriptomics Glycine
Transcriptomics
Glycine
Proteomics
Medicago
Medicago Transcriptomics, proteomics, metabolomics, pathways, and literature Proteomics Medicago and Sinorhizobium meliloti (continued)
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
89
Table III (continued) Database (URL) Australian National University 2D‐PAGE Database (http://semele.anu.edu.au/) AlfaGenes (http://ukcrop.net/perl/ ace/search/AlfaGenes) BeanGenes (http://beangenes.cws. ndsu.nodak.edu/) CoolGenes (http://ukcrop.net/perl/ ace/search/CoolGenes)
Data types
Represented organism(s)
Proteomics
Medicago
EST, genetic map, and pathways Genetic map, gene classification, pathology, and cultivar data Genetic map
M. sativa Phaseolus and Vigna
Cicer and Lens
36,878 tentative consensus sequences (TCs) or unigenes representing transcripts from over 61 diVerent cDNA libraries. Figure 1 and the inset in Fig. 6 provide examples of how such data can be mined to provide a first indication of expression pattern to assist in identification of gene function (Section VI.B.1 below). Specific examples relating to the functional annotation of genes involved in the synthesis of phenylpropanoid‐derived natural products have been reviewed (Costa et al., 2003; Dixon et al., 2002). Although simple, analysis of EST frequency in sequenced libraries requires some caution. First, there are problems associated with potential sequencing errors aVecting the assembly of ESTs into the ‘‘contigs’’ known as TCs in the case of MtGI (Fig. 1; Rudd, 2003). The word ‘‘tentative’’ is important, as the assemblies can change when additional EST sequence information becomes available. In view of this, the TIGR EST databases keep track of all previous TC numbers for each contig, and these will sometimes split or coalesce until final confirmation is obtained from whole‐genome sequence data. Second, because the selection phase of EST sequencing simply involves random picking of colonies, statistical issues aVect the reliability of EST frequency within a library, particularly when considering libraries with low numbers of sequenced ESTs. With these limitations in mind, the increasing number of EST resources for important crop plants (Kuenne et al., 2005) nevertheless provides an excellent starting point for selection of target genes, preliminary expression profiling, and development of molecular markers (Section IV.C below). DNA microarrays provide an adaptable and rapid approach to transcript profiling. However, because they rely on previously determined gene or EST sequences, they represent a targeted profiling technology unless the arrays contain a complete unigene set for a particular organism. Several articles summarize the most important factors associated with the production, hybridization, and analysis of microarrays, and their applications for plant science (Baldwin et al., 1999; Brazma et al., 2001; Kehoe et al., 1999; Robinson et al., 2004;
R. A. DIXON ET AL.
90 5⬘-EST
AAAAAA 3⬘-EST cDNA end sequencing
EST collection
Tentative consensus with nine ESTs (color shows tissue library of origin)
EST number
Clustering and assembly
“In silico northern”
Tissue (library)
Figure 1 Generation and analysis of ESTs. A population of transcripts is converted to cDNA, which is then cloned and randomly sequenced. Each sequence run provides an EST. ESTs are clustered and aligned into TCs; each TC theoretically represents the transcript from one gene, complete only as far as the random sequence information allows. The number of occurrences of a particular EST in a particular cDNA library represents the relative transcript level of the corresponding TC in the biological material from which the library was constructed.
van de Peppel et al., 2003; Wu et al., 2001). There are two major types of microarrays, representing diVerent technology platforms for generation and analysis of the arrays. Spotted arrays consist of a large number of DNA species arrayed as a grid on a glass slide. The DNA may be from a cDNA clone, particularly in the case of ‘‘custom arrays’’ made by one laboratory for analyzing a specific set of transcripts, although better results are often obtained if all the spots contain DNA fragments of the same size, optimized for hybridization characteristics, as in the case of oligonucleotide arrays. However, this requires significant informatics input, which can be provided by commercial providers, such as Qiagen Operon, Alameda, CA, who will then make the required oligonucleotide set for in‐house spotting, or Agilent Technologies, Palo Alto, CA, who will provide DNA arrays to you. AVymetrix DNA arrays utilize photolithographic masking methods and combinatorial chemistry to synthesize large numbers of unique probes on each array. Each annotated open reading frame is represented by around 11–13 pairs of oligonucleotides. Each pair is composed of a perfect match and one‐base pair mismatched oligonucleotide (Barnett et al., 2004).
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
91
As an example of the development of plant microarray resources, the early arrays generated for M. truncatula first consisted of spotted EST clones [2K, 6K, and 8K (the latter represented 6300 nonredundant genes); Firnhaber et al., 2005], followed by a commercial 16K oligonucleotide array (Aziz et al., 2005; Suzuki et al., 2005), a custom AVymetrix array containing 10,000 M. truncatula probe sets and the complete genome of Medicago’s rhizobial symbiont Sinorhizobium meliloti (Barnett et al., 2004), and finally a commercial AVymetrix array containing 32,167 M. truncatula EST/mRNA‐based and chloroplast gene‐based probe sets, 18,733 M. truncatula IMGAG and phase 2/3 BAC prediction‐based probe sets from the whole‐genome project, 1896 alfalfa EST/mRNA‐based probe sets (primarily from trichome ESTs), and 8305 S. meliloti gene prediction‐based probe sets. Further iterations of this array will occur as the Medicago genome attains completion. Examples of gene expression profiling using legume gene chips cover many aspects of legume biology, from determining those transcripts that are specifically associated with the nodulation process (Barnett et al., 2004; Colebatch et al., 2004) and flower and pod development (Firnhaber et al., 2005), to identifying genes of secondary metabolite biosynthesis activated in response to microbial elicitors or wound signals (Suzuki et al., 2005) or expressed in glandular trichomes (Aziz et al., 2005). The 16K Medicago oligonucleotide arrays have also been used to determine the ‘‘substantial equivalence’’ of transgenic plants expressing an engineered natural product pathway for isoflavone formation compared to plants not expressing the new pathway (Deavours and Dixon, 2005). An equally wide number of applications of microarray technology have been reported in Arabidopsis, ranging from studies on ethylene signaling (De Paepe et al., 2004) and response to UV light (Casati and Walbot, 2003) to methyl jasmonate (MeJA) signaling (Sasaki‐Sekimoto et al., 2005). Work is currently in progress to generate a publicly available ‘‘gene expression atlas’’ for Medicago through microarray analysis (using the commercial AVymetrix arrays) of RNA samples from multiple tissues and diVerent physiological treatments (M. Udvardi, personal communciation). These will supplement the large sets of Arabidopsis microarray data that are already publicly available (http://aVymetrix.Arabidopsis.info/narrays/experimentbrowse.pl; http://www. genevestigator.ethz.ch/), and the emerging whole‐genome expression profiles for important crops such as rice, soybean, barley, and tomato (Rensink and Buell, 2005). Some databases contain combinations of microarray and EST abundance data (Fei et al., 2006). The value of an EST collection or a microarray experiment for giving a readily accessible picture of the transcriptome of a particular tissue type is further enhanced if the degree of resolution can be increased from the organ to the cellular level. Microarray analysis was, in the past, limited by the relatively large amount of RNA required for hybridization (generally in the region of 50–200 mg of total RNA per hybridization, equating to 50–100 mg
92
R. A. DIXON ET AL.
of plant tissue). Such a requirement precluded the technology from taking advantage of the increasing refinement of tissue isolation procedures, such as laser capture microdissection (Kerek et al., 2003), or methods for physical isolation of appendages such as trichomes (Lange et al., 2000; Wagner, 1991). Methods have been developed for amplification of target RNA samples so that microarray analysis is now possible with as little as 100 ng of total RNA (Hertzberg et al., 2001), and PCR‐amplification methods have allowed for the generation of EST libraries from very small tissue samples such as isolated glandular trichomes (Aziz et al., 2005). Trichomes are a particularly attractive target tissue for EST (in the absence of a sequenced genome) or microarray analysis, since they often show high specialization for the synthesis and secretion of species‐specific bioactive secondary metabolites that confer insect and pest resistance (Georgieva, 1998; Lovinger et al., 2000; Maluf et al., 2001). Several examples illustrate the insights that can be gained into this specialized biochemistry by randomly sequencing, annotating, and functionally characterizing the biosynthetic enzyme gene transcripts that are often abundantly expressed in trichomes (Aziz et al., 2005; Fridman et al., 2005; Gang et al., 2001; Wagner et al., 2004). Availability of suYcient RNA for hybridization is not the only factor that can limit the scope and reproducibility of a microarray experiment. Some genes, particularly regulatory genes, such as TFs, are expressed at very low levels, and the absolute signal strength of their hybridization approaches the noise level on the array (Czechowski et al., 2004). In such cases, alternative profiling methods may be necessary. Sets of oligonucleotide probes have been developed for profiling the complete TF complement of Arabidopsis by quantitative real‐time polymerase chain reaction (qRT‐PCR; Czechowski et al., 2004), and the technique has been used to study both developmental and wound/pathogen defensive TF gene expression (Czechowski et al., 2004; McGrath et al., 2005). An application of this method to profile glycosyltransferase gene expression in Medicago has shown it to be highly sensitive, reproducible, and to correlate well with parallel analysis by AVymetrix microarray analysis (Modolo et al., 2007). It may therefore prove a popular technology for profiling transcript levels if highly accurate quantification is necessary. Classical DNA microarrays are assembled from a set of unigenes that generally represent protein‐coding transcripts. However, several types of noncoding RNAs have been discovered. The small noncoding RNAs play important roles in gene regulation (Bartel and Bartel, 2003), are generally in the region of 21–30 nucleotides in length, and fall into at least three distinct classes; microRNAs (miRNAs), small interfering RNAs (siRNAs), and repeat‐associated small interfering RNAs (rasiRNAs) (Zamore and Haley, 2005). It is becoming increasingly clear that a significant proportion of the RNA transcripts in human do not encode proteins (Claverie, 2005), and the
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
93
search is now on for their function using targeted genetic screens (Mattick, 2005). Global analysis of plant genomes, in addition to studies specifically targeting miRNAs and siRNAs (Bartel and Bartel, 2003), also supports the importance of noncoding RNAs in plants. Initially, EST sequences that were either short or did not appear to encode an open reading frame were ignored as artifacts, and for this reason such sequences did not generally appear on microarrays. With the advent of whole‐genome tiling arrays, it has become possible to verify, and determine the expression pattern, of noncoding RNAs on a global scale. An excellent example of this approach in plants used a combination of full‐length cDNA discovery and hybridization of RNA populations to whole‐genome arrays to define the transcriptional units of all Arabidopsis genes (Yamada et al., 2003). The tiling array consisted of 12 individual slides, each containing around 834,000 ordered 25‐mer oligonucleotides that together represented about 94% of the Arabidopsis genome. This seminal work resulted in a full transcriptional annotation of the Arabidopsis genome in terms of genes that were (1) annotated and expressed, (2) annotated but not expressed, and (3) not annotated but expressed. The latter class were discovered in what had been thought to be intergenic regions. Surprising findings were the relatively large number of antisense transcripts, the high transcriptional activity of the centromeric regions, and the transcription of many genes previously classed as ‘‘pseudogenes,’’ suggesting that these might serve a regulatory function.
2.
For Species Lacking Genomics Resources
It is possible to carry out global scale transcript profiling in crop plants for which neither genomic nor extensive EST sequence information is available. Where a crop plant is closely related to a model species, it may be possible to utilize microarray resources from the model species. Examples include the use of M. truncatula microarrays to profile transcripts in alfalfa (Aziz et al., 2005; Deavours and Dixon, 2005), and tomato arrays for profiling transcripts from other Solanaceous species such as pepper and eggplant (Moore et al., 2005). Among genes represented in the alfalfa glandular trichome ESTs as TCs that have orthologues in M. truncatula, 66.5% had 100% sequence identity to the corresponding M. truncatula orthologue (Aziz et al., 2005) and, of the total 5647 alfalfa trichome ESTs sequenced, 4804 had M. truncatula orthologues with E values of 20 and below. Because of this very high degree of sequence similarity between alfalfa genes and their M. truncatula orthologues, significant signal was observed for most of the 16,086 genes represented on the oligonucleotide arrays when hybridized with alfalfa total RNA (Aziz et al., 2005). The overall number of features with signal more
R. A. DIXON ET AL.
94
than 300 pixels above background was the same as observed when the same arrays were hybridized with labeled RNA from M. truncatula stems. cDNA amplified fragment length polymorphism (cDNA‐AFLP) and SAGE are nontargeted transcript‐profiling techniques that can theoretically be applied to any living organism. The basic principles behind the two methods are outlined in Fig. 2. cDNA‐AFLP is an RNA fingerprinting approach which involves cDNA synthesis from the RNA transcripts to be analyzed, restriction digestion of the primary templates and ligation of anchors to their termini, preamplification with anchor‐specific primers, and selective amplification of the restriction fragments with primers extended with one or more specific bases (Bachem et al., 1998). The method allows for the simultaneous analysis of multiple samples. Examples of its use in
A
RNA population
B
Synthesize double strand cDNA TTTTTTTTTTTTT@
Synthesize cDNA
AAAAAAA TTTTTTT AAAAAAA TTTTTTT AAAAAAA TTTTTTT AAAAAAA TTTTTTT
TTTTTTT AAAAAA@
Synthesize double strand cDNA
Cleave with anchoring enzyme (NIaIII) @ Digest with one frequent-and one rare-cutting restriction enzyme
@ Ligate anchors @
AAAAAAA TTTTTTT AAAAAAA TTTTTTT AAAAAAA TTTTTTT AAAAAAA TTTTTTT
Release SAGE tags
@ Preamplify @
Primary template
NN Selectively amplify fragments NN NN Secondary template NN Fingerprint-amplified fragments on a polyacryamide gel Elute and sequence differentially expressed bands
Form ditags and concatenate
Sequence and analyze data 30 50 20 10 Number of tags per 100,000
Figure 2 cDNA‐AFLP and SAGE, techniques for nonbiased transcript profiling. (A) Procedure for cDNA‐AFLP analysis. @ ¼ biotin group, black circle ¼ streptavidin bead. The rare‐ cutting site anchors and primers are shown in black, and the red circle symbol represents a 32P label. See Bachem et al. (1998) for further details. (B) Procedure for SAGE analysis. After cutting the cDNAs with a frequent cutting enzyme (usually NlaIII), linkers are ligated to the 50 ‐ends; these linkers contain a site for a type IIs restriction enzyme (BsmFI) which cuts a 15‐bp fragment (SAGE tag) of the cDNA (joined to the linker). These fragments are ligated tail to tail to form ditags, which are then amplified, concatenated, and sequenced. Special software determines the frequency of the SAGE tags among the sequenced DNA.
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
95
plant biology include the profiling of transcripts responding to ethylene in Arabidopsis (De Paepe et al., 2004) and to MeJA in tobacco BY2 cell suspension cultures (Goossens et al., 2003). In the latter example, the authors pointed out the importance of a nontargeted approach for discovering natural product biosynthetic genes from the many unrelated medicinal plants, none of which currently has genome resources. In SAGE analysis, a method that takes advantage of the target sequence recognition properties of class‐II DNA restriction enzymes that cut a short distance away from the enzyme’s DNA recognition site, transcripts are reduced to short tags which are then concatenated and sequenced (Velculescu, 1999) (Fig. 2). SAGE is therefore a nontargeted or ‘‘open’’ system. In contrast to DNA arrays which are considered a ‘‘closed system,’’ nontargeted transcript‐profiling methods such as SAGE allow for the identification and analyses of previously undescribed RNAs (e.g., antisense RNAs). Comparisons of SAGE and microarray analysis using the same RNA samples show quite good correlations (Ishii et al., 2000), and SAGE has become a popular transcript‐profiling technique for plants, from loblolly pine (Lorenz and Dean, 2002) to rice (Matsumura et al., 1999). Adaptations have been made to the technique to make it applicable to the analysis of transcripts from microdissected cells and other small samples (Velculescu et al., 2000). MPSS (Brenner et al., 2000) identifies short sequence signatures (20 bp) generated from a position immediately adjacent to the DpnII restriction enzyme site nearest to the poly‐A tail of an mRNA transcript. The relative abundance of these signatures in a given mRNA sample (library) represents a quantitative estimate of expression of that gene. MPSS and now clonal single molecule arrays (CSMATM) technologies were developed by Solexa, Inc. (www.solexa.com). Solexa has discontinued providing MPSS as a service and now exclusively oVers CSMATM. Although the data output from MPSS and CSMATM are essentially the same, CSMATM is based on a high‐density, eight distinct channel flow‐cell array format whereas MPSS is a bead‐based technology. Sequence data generated by the CSMATM platform is based on sequencing‐by‐synthesis (SBS) and reversible terminator chemistry, and leverages massively parallel sequencing of cDNA fragments to generate data from millions of fragments simultaneously. Solexa’s SBS approach is anticipated to generate up to one billion bases of data per run at costs more economical than MPSS.
C. PROTEOMICS Chemically, DNA is a relatively simple polymer with only four building blocks, in contrast to the 20 diVerent protein amino acids and the many thousands of primary and secondary metabolites found in plants. Profiling
96
R. A. DIXON ET AL.
the proteome and metabolome therefore poses significant technical challenges compared to transcript profiling. In fact, few if any studies have been able to profile the complete proteome or metabolome of a complex organism; this is as much a problem of initial separation of complex mixtures as it is one of final detection. Because both proteomics and metabolomics require specific chemical determination of molecules with quite diVerent structures, mass spectrometry has become the detection/analytical method of preference, and is capable of both extreme sensitivity and mass discrimination. Classically, proteomics has relied on two‐dimensional isoelectric focusing SDS‐polyacrylamide gel electrophoresis for initial protein separation, and this approach can routinely resolve around 1000 diVerent plant proteins (Lei et al., 2005; Watson et al., 2003; Yan et al., 2006). This is nevertheless, only a small fraction of the predicted proteins in a tissue based on the numbers of genes known to be expressed. This lack of penetration represents less of a problem if the technique is applied to specific subcellular fractions (Majeran et al., 2005; Nelson et al., 2006; Taylor et al., 2005; Ytterberg et al., 2006), tissues with a preponderance of a specific protein type under study (e.g., seed storage proteins; Thiellement et al., 1999), or specialized structures such as trichomes (Wienkoop et al., 2004). Approaches to profiling proteins have used shotgun methods without gel fractionation, relying instead on more rapid separation methods, protein tagging, and the versatility of modern mass spectrometers (Aebersold and Mann, 2003; Chen et al., 2006a; Hass et al., 2006; Shen et al., 2005; WolV et al., 2006). Proteomic approaches are being applied to address many of the same types of questions currently investigated by transcript profiling; these include genetic diversity, phylogenetic relationships, characterization of mutants, studying responses to abiotic stresses such as UV light and cold, and understanding seed development (Agrawal and Thelen, 2006; Casati et al., 2005; Thiellement et al., 1999; Yan et al., 2006). However, because of the high cost of the required mass spectrometers and current limitations to the depth of profiling, proteomics will likely remain, at least for the time being, less utilized than transcriptomics as regards applications to agronomy and plant breeding.
D. METABOLOMIC ANALYSIS 1.
Introduction
Levels of plant metabolites are controlled by both genetic and environmental factors, and the metabolome is often referred to as the functional manifestation of gene expression. Metabolite profiling can be classified into three approaches, targeted profiling, fingerprinting, or true metabolomics
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
97
(in depth and unbiased). Several articles provide reviews of the technology and its challenges for these various types of approach (Dixon et al., 2006; Fiehn, 2002; Fuell et al., 2004; Hall et al., 2002; Sumner et al., 2002, 2003). Early studies in plant metabolomics utilized gas chromatography‐mass spectrometry (GC‐MS) to profile mainly hydrophilic primary metabolites (Roessner et al., 2000, 2001). Importantly, this work demonstrated that metabolite profiling was of value for providing genetic, as well as chemical, understanding of plant systems. Thus, analysis of GC‐MS profiles of extracts from diVerent potato genotypes, when compared to profiles from transgenic potato lines modified in sucrose catabolism, revealed how metabolite profile analysis clearly shows the way in which environmental factors can lead to metabolic phenotypes linked to specific genetic changes (Roessner et al., 2001). This work defined both the strengths and potential diYculties of the approach. One major strength is that data mining tools such as hierarchical cluster analysis and principle component analysis (PCA) allow for clear visualization of factors that relate or distinguish diVerent metabolite profiles, thus making the profile a very rich source of information for comparative genetic analysis (Fiehn, 2002; Roessner et al., 2001). These, and other informatic approaches, have been reviewed (Sumner et al., 2003). As with proteomics, one weakness is that most metabolic profiling technologies only sample a proportion of the total metabolome, Thus, the early GC‐MS analysis of potato tissues only resolved about 80 diVerent compounds, whereas it is estimated that a simple plant such as Arabidopsis contains in excess of 5000 diVerent metabolites. Improvements in technology, for example, by the use of raid scanning time of flight mass spectrometry, have increased the number of metabolites detectable in crude plant extracts to around 1000 (Hall et al., 2002).
2.
Targeted Profiling
The large numbers of secondary metabolites produced by plants, perhaps in excess of 200,000 throughout the plant kingdom, present the biggest problem for nontargeted metabolomics. These compounds are chemically very diverse, often species specific, and the physical details of most are not present in chemical databases. Of course, such metabolites do not have to be actually identified in initial metabolomics experiments; an ‘‘unknown’’ can be treated exactly the same way as a ‘‘known’’ during clustering and statistical analysis, and can be treated as a genetic marker in the absence of its chemical identity. However, a greater problem for inclusion of secondary metabolites in the high throughput profiling necessary for the technology to be used in genetic mapping and breeding is the chemical diversity of these compounds. This necessitates specific extraction and sometimes separation protocols for
R. A. DIXON ET AL.
98
specific classes of secondary metabolites. Simple GC‐MS is seldom used because of problems with derivatization and subsequent identification of the derivatives. Rather, most studies on secondary metabolite profiling have used a targeted approach designed to address a single class of compound, and one of the most commonly used analytical approaches has been high performance liquid chromatography (HPLC) coupled with mass spectrometry and/or UV/visible spectrophotometric detection (Sumner et al., 2003). Targeted metabolite profiling is an old technology, and good methods for many diVerent classes of compounds have been available for some time. Examples include methods for profiling flavonoids and isoflavonoids (Graham, 1991; Liu and Dixon, 2001), phenylpropanoids (Chen et al., 2003), triterpenes (Huhman and Sumner, 2002), carotenoids (Fraser et al., 2000), various classes of alkaloids (He, 2000; Kale´n et al., 1992; Stashenko et al., 2000), and acyl CoAs (Larson and Graham, 2001). These and related targeted profiling approaches have been applied to understanding the genetic basis of metabolite abundance via quantitative trait locus (QTL) analysis (Morrell et al., 2006) (Section IV.D below), and determining phenotypic eVects of transgenic modification of plants for improved quality traits (Deavours and Dixon, 2005; Morreel et al., 2004; Xie et al., 2006). Figure 3 provides an example of targeted profiling of flavonoid compounds in alfalfa. The extraction and HPLC method used favored the extraction and separation of (iso)flavonoids and their glycosides (Deavours and Dixon, 2005). By this approach, it was possible to show that constitutive expression of an isoflavone synthase transgene in alfalfa led to accumulation of isoflavone glucosides in the leaves, whereas the endogenous flavonoids found in the leaves were glucuronic acid conjugates. Plants constitutively expressing the isoflavone synthase produced higher levels of potentially defensive isoflavonoid metabolites following exposure to biotic or abiotic stress (Deavours and Dixon, 2005).
3.
Metabolic Fingerprinting
As suggested above, it is not necessary to know the exact chemical nature of the components of a metabolic profile to be able to use the profile as a genetic and phenotypic tool. Because of this, more rapid analytical methods that provide a ‘‘metabolic fingerprint’’ rather than a detailed profile of individually separated molecules are being applied in the field of molecular agriculture. These include nuclear magnetic resonance (NMR) and near infrared (NIR) spectroscopy. NMR and NIR profiles can be subjected to the same types of statistical analysis as GC or HPLC elution profiles, and regions of the profiles that exhibit the greatest variation between samples/ treatments can be correlated with genotype, environment, or expression of
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
99
mAU A 200 150 100 50
Absorbance at 254 nm
0 20 mAU
30
40
50
60
min
60
min
60
min
B 200 150
1 3
100
2
50
5
4
0 20 mAU
30
40
50
C 200
1
3
T
150 100 2
50 0 20
30
4
40 50 Retention time (min)
5
A
Figure 3 Use of targeted metabolite profiling to show production of isoflavone glycosides in leaves of transgenic alfalfa plants expressing an M. truncatula isoflavone synthase gene (MtIFS1) under control of the constitutive cauliflower mosaic virus 35S promoter. HPLC traces show unhydrolyzed leaf extracts of (A) an empty vector control line and (B) an MtIFS1‐expressing line. Peaks with UV spectra similar to the isoflavone genistein that are not present in the control extracts are numbered 1–5. Peaks 1 and 4 were identified by LC/MS as the isoflavone glucosides genistin and sissotrin, respectively. Panel (C) shows an HPLC trace of a leaf extract of the MtIFS1‐expressing line after digestion with b‐glucuronidase. This converts the major endogenous leaf flavonoids, glucoronides of the flavones apigenin and tricin, to their corresponding aglycones (peaks A and T, respectively), better revealing the isoflavone glycosides. HPLC was carried out on an ODS2 reverse‐phase column (5‐mm particle size, 4.6 250 mm2) and eluted in 1% (v/v) phosphoric acid with an increasing gradient of acetonitrile (0–5 min, 5%; 5–10 min, 5–10%; 10–25 min, 10–17%; 25–30 min, 17–23%; 30–65 min, 23–50%; 65–69 min, 50–100%) at a flow rate of 1 ml min1 (Deavours and Dixon, 2005).
100
R. A. DIXON ET AL.
a transgene. NMR fingerprinting has begun to find uses in functional genomics, the study of genetic diversity, the evaluation of the safety of transgenic crops, and determining responses of plants to infection (Charlton et al., 2004; Ward and Beale, 2006). In addition to providing a standard method for assessment of plant quality traits, such as fiber and digestibility (Jung, 1997), NIR techniques have been developed that allow for rapid estimation of specific metabolites, such as ergot alkaloids in endophyte‐infected tall fescue (Roberts et al., 2005).
4. Nonbiased Metabolomics Although it essentially targets those classes of molecules that are subject to the applied extraction protocol(s) and separation methods, the standard GC‐MS approach to plant metabolomics is largely viewed as a nontargeted approach. Its eVectiveness depends on the availablilty of mass spectral libraries to assist the identification of both known and unknown components (Kopka, 2006). In addition to the examples provided in Section D.I above, GC‐MS profiling has become an important tool for functional genomics and analysis of biotic stress responses in legumes (Broeckling et al., 2005; Debrosses et al., 2005). The remarkably high mass resolution power of Fourier transform ion cyclotron mass spectrometry (FT‐MS) allows metabolite profiling without the need for preseparation of metabolites. Complex mixtures can be injected directly into the mass spectrometer, and the components are essentially resolved via the mass discrimination of the instrument (Aharoni et al., 2002). Identification is based on absolute mass measurement. This is a very powerful approach and has been applied to gene discovery associated with nutritional stress in Arabidopsis (Hirai et al., 2004). However, the method can not discriminate between isomers. Extensive metabolomic analysis is facilitated by metabolic pathway databases for the plant species of interest. Several such databases have been developed, and some incorporate features for simultaneous display of gene expression data from microarrays or other formats directly onto the metabolic pathway maps (Krieger et al., 2004; Lange and Ghassemian, 2005; Thimm et al., 2004). The version for M. truncatula is called MedicCyc (Urbanczyk‐Wochniak and Sumner, 2007) and features more than 250 metabolic pathways with associated genes, enzymes, and metabolites. More challenging to construct are databases that actually store the raw data obtained from diVerent types of ‘‘omics’’ approaches. Database of ‘‘omes’’ (DOME) is an early example of such an approach, initially constructed to house transcript, protein, and metabolite data from M. truncatula cell lines responding to biotic and abiotic elicitors (Mehrotra and Mendes, 2006).
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
5.
101
Integrating Transcriptomic and Metabolomic Datasets
Several examples demonstrate the advantages of being able to simultaneously profile the transcriptome and the metabolome. This is a powerful new approach to the analysis of biological systems because it allows statistical analysis tools to be used to analyze, and therefore correlate, both genotype and phenotype (the metabolome; Phelps et al., 2002). However, such an approach presents some problems in whole plant systems, because the existence of diVerent cell types in a complex organism makes it diYcult to correlate transcripts and metabolites that might not be expressed or synthesized in the same cells. Because of this, the first examples of the approach were reported for prokaryotic systems (Phelps et al., 2002), which also have the advantage of being more readily amenable than plants to flux analysis through labeling with isotopic precursors followed by MS or NMR analysis. Likewise, to take advantage of more homogeneous cell populations, the first examples of integrated transcriptomics/metabolomics in plants utilized cell suspension cultures. In a study aimed at elucidating the genes of secondary metabolism, particularly nicotine alkaloid biosynthesis, expressed in tobacco BY2 cell suspension cultures in response to the wound signal MeJA, targeted metabolite profiling by GC‐MS was coupled with cDNA‐AFLP analysis of diVerentially induced transcripts (Goossens et al., 2003). This approach led to the identification of a number of candidate genes for involvement in the nicotine biosynthesis pathway itself, or in its transcriptional regulation. A similar approach, but using oligonucleotide microarrays for transcript profiling and LC‐MS for analysis of secondary metabolites, was used to identify genes encoding glycosyltransferases involved in MeJA‐induced accumulation of triterpene saponins in M. truncatula cell suspension cultures (Achnine et al., 2005). A study of the relationships between the transcriptome and primary and secondary metabolism in Arabidopsis seedlings under conditions of sulfur or nitrogen deprivation revealed the power of this integrated approach for gene discovery, especially when combined with powerful informatic analysis (Hirai et al., 2004). The transcriptome analysis used a macroarray that contained EST clones corresponding to around 9000 Arabidopsis genes, and the metabolome was profiled using extraction in three solvent systems of diVerent polarities followed by FT‐MS analysis. General responses to sulfur and nitrogen deficiency were identified through mathematical analysis of transcriptome and metabolome datasets using PCA and batch‐learning self‐organizing map analysis (Hirai et al., 2004). Using the same techniques, detailed metabolite and transcript profiling of the Arabidopsis PAP1 mutant, which overexpresses anthocyanins (Sections II.F and VI.B.2), revealed the presence of eight novel anthocyanins, and, among the 32 genes that were shown to be specifically upregulated by PAP1, two were identified as specific
102
R. A. DIXON ET AL.
flavonoid‐ and anthyocyanin‐glycosyltransferases (Tohge et al., 2005). Such approaches represent a powerful tool for functional annotation of genes, particularly those that are members of large families that encode enzymes with promiscuous and overlapping in vitro substrate specificities such as glycosyltransferases (Bowles et al., 2006). For such enzymes, correlation of gene and metabolite expression patterns might be the deciding approach for confirming in vivo function.
6. Profiling Technologies and ‘‘Substantial Equivalence’’ During the regulatory process for approval of transgenic plants for commercial use, it is necessary to demonstrate that the regulated product is ‘‘substantially equivalent,’’ from a compositional viewpoint, to its unmodified parent material. Generally, such equivalence is taken as meaning that the nutritional properties of the plant are not altered and that no potentially toxic compounds have been introduced. In a study with herbicide‐resistant alfalfa, field‐grown material was analyzed for fiber content, amino acid and mineral composition, and levels of the potentially estrogenic isoflavonoid coumestrol (McCann et al., 2006) using classical analytical procedures for each parameter. Although the increasing ability to perform more global analysis of transcripts, proteins, and metabolites suggests that ‘‘omics’’ approaches may become standard for demonstration of substantial equivalence, there are also arguments against this. In some respects, ‘‘omics’’ approaches are too sensitive, and it is sometimes the case that variations between diVerent tissues, varieties, or environmental conditions are greater than the changes observed following introduction of a transgene. This is illustrated by the PCA analysis of soluble phenolic compound profiles in control and transgenic alfalfa lines modified in lignin content and composition (Chen et al., 2003). PCA analysis could resolve profiles from transgenics from those from controls for stem extracts, but not for leaf extracts. At the same time, the method resolved diVerences between two nontransgenic cultivars when considering extracts from either leaf or stem tissue (Chen et al., 2003). Nevertheless, metabolite and transcript profiling and fingerprinting have been performed to establish, or refute, substantial equivalence. In one study with peas, NMR fingerprinting showed that expression of a transgene did indeed aVect the metabolite profile, but that this eVect was masked by changes induced by environmental factors such as drought (Charlton et al., 2004). In a study of alfalfa plants ectopically producing isoflavones in the leaves, microarray analysis failed to demonstrate significant changes in transcript levels, other than for the expressed transgene, in pair‐wise comparisons of controls and trangenics, although there was significant interplant variation (Deavours and Dixon, 2005). If the new ‘‘omics’’ technology is to be
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
103
applied to the assessment of substantial equivalence, it is important that a public consensus is reached as to which analytes are of significance for human and animal health considerations and that tolerance intervals are defined that encompass the variations found in commercial populations (Dixon et al., 2006; Ridley et al., 2002).
IV.
MOLECULAR MARKERS A. MARKER TYPES
Following the segregation of Mendelian genetic markers is the most powerful method to understand hereditary transmission (Beckmann and Soller, 1993). The advent of agriculture and domestication began with selection of superior genotypes/lines. Classical plant breeding techniques were mainly based on phenotypic selection (PS) where traits of interest were tagged with markers like seed color, leaf size, and flower color, which could distinguish between genotypes. However, morphological markers are influenced by the environment, may be linked to undesirable traits, and their use for selection is time consuming, requiring large population sizes and space for testing. In the early molecular era, isozyme and protein markers were used to select genotypes in plant breeding programs. These biochemical markers are, however, characterized by low polymorphism, especially between similar or related cultivars. The advent of rapid DNA sequencing led to the discovery of DNA‐based markers (molecular markers) and these have became the marker class of choice. Molecular markers are based on DNA polymorphism as a result of mutation and are highly heritable. Their main advantage is that they are much more numerous and polymorphic than morphological or biochemical markers. The genomes of most plant species contain between 108 and 1010 nucleotides, and thus even a small proportion of polymorphic sites can yield a large number of potential markers (Paterson et al., 1991). Early identification of DNA markers relied on restriction fragment length polymorphisms or RFLPs (Tanksley et al., 1989). RFLP markers segregate as codominant alleles capable of identifying all three morphs, thus being highly informative. The polymerase chain reaction (PCR) technique revolutionized molecular marker technology. Rapid amplification of discrete DNA fragments by PCR enables quick identification of DNA polymorphisms within a genome. The rapid identification of such markers linked to important loci facilitates their integration into plant breeding programs. Randomly amplified polymorphic DNA (RAPD) markers were the first markers of this kind to be developed (Williams et al., 1990). In the past two decades, many
104
R. A. DIXON ET AL.
diVerent molecular marker systems have been developed to serve specific needs, many of which have relied on genome and EST‐sequencing projects. The marker systems currently employed include RFLPs, single strand conformation polymorphisms (SSCPs), sequence‐tagged microsatellite sites (STMSs), RAPDs, sequence characterized amplified regions (SCARs), ESTs, microsatellites, or simple sequence repeats (SSRs), amplified fragment length polymorphisms (AFLPs), sequence‐tagged sites (STSs), cleaved amplified polymorphic sequences (CAPSs), single nucleotide polymorphisms (SNPs), and heteroduplex markers. The ideal marker class should provide more markers per unit DNA, be stable, easily detectable, safe, and cost‐eVective, and have a high degree of polymorphism. Molecular markers are now an indispensable tool for cultivar identification and parentage analysis (Dudley et al., 1992; Sefc et al., 2000), genetic diversity analysis (Mian et al., 2005a; Smith and Smith, 1992), genome mapping, and the tagging of genetically important traits (Cardinal et al., 2003).
B. MOLECULAR GENETIC MAPS Linkage maps are constructed by following the segregation pattern of molecular markers in a population. Markers are placed in linear order based on pair‐wise recombination frequencies between the markers. High marker polymorphism in a population is the key for successful linkage analysis. Backcross, F2, recombinant inbred lines (RILs) and doubled haploids are the most commonly used populations for molecular mapping of self‐pollinated crops (Chen et al., 2001; Eujayl et al., 1998). The pseudo F2 cross (between two heterozygous parents) is the most frequently used population in mapping cross‐pollinated crops (Saha et al., 2005; Van Eck et al., 1993). All of the above populations have both advantages and disadvantages. The F2 and backcross populations show higher segregation, but are not available for subsequent studies. RIL populations can be permanently propagated and oVer unique advantages in quantitative trait loci (QTL) mapping (Burr and Burr, 1991). However, development of RILs is time consuming and very diYcult in self‐incompatible species. In doubled haploid populations, homozygocity for a particular locus can be obtained quickly but segregation distortion is a major problem (Cloutier et al., 1995). Molecular markers are commonly used to generate genetic linkage maps, and have provided a major contribution to the genetic knowledge of many cultivated plants. Over the past two decades, genetic linkage maps have been developed for most of the agriculturally important plant species (Alm et al., 2003; Chen et al., 2001; Eujayl et al., 1998; Gebhardt et al., 1991; Jacobs et al., 1995; Jones et al., 2002; Kuhl et al., 2001; Perez et al., 1999;
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
105
Saha et al., 2005; Tanksley et al., 1992; Warnke et al., 2004; Xu et al., 1995). Such molecular maps have been used to map major genes (Van Eck et al., 1993) and to identify the genetic components of polygenic traits (Bonierbale et al., 1994; Qiu et al., 2006). Genetic linkage maps have been used successfully for the identification of markers linked to a gene of interest (Xu et al., 1999). The availability of genetic maps opened the door for comparative mapping, which allows the comparison of genome organization and orientation of one species to that of a closely or distantly related species through common markers between maps. Comparative mapping has revealed that gene content and order are generally conserved among closely related species (Alm et al., 2003; Jones et al., 2002; Van Deynze et al., 1995). It has also been used for extending genetic information from model organisms to genetically more complex species (Paterson et al., 1995).
C. GENOMICS FOR GENERATION OF MOLECULAR MARKERS Among the diVerent marker classes, SSRs have become the marker class of choice due to their manifold advantages over other marker systems. Single nucleotide polymorphism (SNP) markers are also becoming more popular as genome sequences for agriculturally important crops are becoming available and SNPs are detected at high frequencies. TILLING is also an attractive system for genome analysis.
1.
Simple Sequence Repeats
SSRs, also known as microsatellites, are short stretches of DNA that are repeated many times. The di‐, tri‐, tetra‐, or pentanucleotide tandem repeats are often multialleleic, chromosome‐specific and dispersed throughout the genome (Weber and May, 1989). The basis of polymorphism is due to variation in the number of tandemly repeated nucleotide motifs, and this is thought to arise from slippage of the DNA polymerase during DNA replication. Although these SSRs are highly abundant in animal and plant genomes (Hamada et al., 1982), the dinucleotide repeats are more common in genomic SSRs (Lee et al., 2004), and trinucleotide motifs are the most abundant in EST‐SSRs (Saha et al., 2004). SSR markers are inherited in a Mendelian fashion and are mostly codominant in nature (Saghai‐Maroof et al., 1994). Genomic or EST libraries can be screened for sequences that contain microsatellite motifs in order to develop primers (Panaud et al., 1996). In the early 1990s, SSR markers were mainly developed from genomic libraries, an expensive and ineYcient procedure (Squirrell et al., 2003). The availability of large numbers of ESTs and other DNA sequence data made
R. A. DIXON ET AL.
106
SSR marker development eYcient and cost eVective for many plant species. The development of SSRs from ESTs has been reported in various crop species, including rice (Cho et al., 2000), durum wheat (Eujayl et al., 2002), barley (Thiel et al., 2003), rye (Hackauf and Wehling, 2002), M. truncatula (Eujayl et al., 2004), and tall fescue (Saha et al., 2004). The frequency of SSRs in the EST databases of cereal crops (rice, wheat, maize, barley, and sorghum) varies from 1.5% (maize) to 4.7% (rice) (Kantety et al., 2002), while in tall fescue it is only 1.3% (Saha et al., 2004). The rate of polymorphism of EST‐ SSRs is comparatively lower than that of genomic‐SSRs (Cho et al., 2000; Thiel et al., 2003). SSRs have been used for the construction of linkage maps in a number of species including Arabidopsis (Bell and Ecker, 1994), maize (Senior et al., 1996), wheat (Ro¨der et al., 1995), rice (Panaud et al., 1996), barley (Liu et al., 1996), and soybean (Akkaya et al., 1995). Allelic profiles of genotypes have been studied using SSR markers for the purpose of genotype identification in potato (Schneider and Douches, 1997), soybean (Maughan et al., 1995), grape (Thomas and Scott, 1993), and rapeseed (Kresovich et al., 1995). Selection of agronomic traits was also accomplished using SSR markers (Yu et al., 1994). Map alignment through common markers is important for making mapping studies universally useful within a species (Powell et al., 1996). Specific SSR primers from one species can be used to amplify DNA from another related species. As EST‐SSR markers are derived from transcribed regions of DNA, they are expected to be more conserved and have a higher rate of transferability than genomic SSR markers (Scott et al., 2000). For instance, tomato SSR sequences generated polymorphic and useful alleles in potato (Provan et al., 1996). SSR loci have high rates of transferability across species (>50%) within a genus (Eujayl et al., 2004; Gaita´n‐Solı´s et al., 2002; Thiel et al., 2003). However, the transferability of SSR loci across genera and beyond seems to be low (Roa et al., 2000; Thiel et al., 2003; White and Powell, 1997).
2.
Single Nucleotide Polymorphisms
SNP has emerged as an important molecular marker system. The utility of SNPs in answering a large range of biological questions in a variety of fields is now beyond question. SNPs greatly expedite the understanding of many diseases and genetic variations in humans. SNPs associated with diVerent human conditions, such as risk of cardiovascular disease, and susceptibility to Alzheimer’s, susceptibility to hip osteoarthritis (Mototani et al., 2005), and risk of thrombosis (Ridker et al., 1995), have been identified. In plants, many SNPs have been shown to be associated with useful traits. For example in rice, SNPs for the fragrance trait and starch gelatinization temperature have been identified (Bradbury et al., 2005).
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
107
The advantage of SNP markers is that they occur at a high frequency in genomes of all organisms. However, the frequency is highly dependent on the type of DNA surveyed, for example, coding versus noncoding sequences, genes of choice, and species investigated. In genomic DNA of maize inbred lines, one SNP was identified per 83 bases, while in the barley intronless Isa gene one SNP occurs every 27 bases (Bundock and Henry, 2004), and in sugarcane ESTs one SNP is found per 50 bases (Cordeiro et al., 2006). Significant improvements have been made in SNP detection protocols, including dCAPS (Michaels and Amasino, 1998) and mass spectrometry using MALDI‐TOF MS (Stoerker et al., 2000). There have also been advances in fluorescence‐based technologies, for example AmplifluorÒ (Serological Corporation), TaqManÒ , SnaPshotÒ , and SNPlexÒ (Applied Biosysteems), and IlluminaÒ (Illumina, Inc.), to detect SNPs. Chip‐based technologies, for example Genechips (AVymetrix), and microarray technology (Wang et al., 2005a) have also been used for SNP detection. Dot‐blot‐SNP analysis was described for application in plant breeding and cultivar identification in rice (Shirasawa et al., 2006). A comparison of three SNP genotyping methods including GOOD (Sauer et al., 2000), AmplifluorÒ , and TaqManÒ for three diVerent herbicide resistance genes from A. thaliana found the best results with TaqManÒ for PCR specificity, flexibility in primer design, and success rate (Giancola et al., 2006). However, all three genotyping techniques were successful in discriminating alleles in various plant species. SNPs are very useful as genetic markers for population studies, germplasm fingerprinting and cultivar identification, molecular mapping, genotype/phenotype association, and for positional‐cloning of specific genes. They have practical utility in identifying mutant lines developed from an original cultivar where most of the other marker systems are ineVective (Shirasawa et al., 2006). The addition of SNP markers significantly increased the overall map length and marker density in sunflower (Helianthus annuus L.; Lai et al., 2005). SNP markers are considered useful for gene mapping using populations derived from crosses between closely related lines; molecular markers like AFLPs and SSRs are found to be less polymorphic in these populations. A new breeding method named ‘‘DNA‐selection breeding’’ has been proposed whereby genes associated with diVerent agronomically important traits are selected by SNP analyses and used for selecting superior genotypes (Shirasawa et al., 2006).
3.
Tilling
TILLING (Section II.F) is a high‐throughput, sensitive, cost‐eVective, and rapid means of finding genetic variation in a population. TILLING is eVective in small or large genomes, diploid or hexaploids, and has great
108
R. A. DIXON ET AL.
potential to identify both induced and naturally occurring variation in many species. Thousands of plants or animals can be screened to identify any single base change as well as small indels (insertions and/or deletions) in any gene or genomic regions (Comai et al., 2004). A million base pairs of genomic DNA can be screened per single assay, which makes TILLING a high‐throughput technique (Slade and Knauf, 2005). This technique combines traditional chemical mutagenesis and modern high‐throughput genotyping. DNA from eight mutant lines can be combined in one PCR tube, thus a 96‐well PCR plate can screen 768 genotypes. During electrophoresis, mutant lines reveal polymorphic fragments relative to wild‐type lines. TILLING is suitable for SNP discovery because it is sensitive enough to detect rare SNPs.
D. METABOLOMIC‐BASED ‘‘MARKERS’’ The term ‘‘genetical metabolomics’’ was defined to describe the use of metabolite profiling in QTL mapping (Morrell et al., 2006). If levels of specific metabolites can be used as quantitative traits to define metabolic QTLs (mQTLs) that control levels of specific metabolites, the nature of the genes underlying the mQTLs might be more readily obvious that in classical genetic QTL mapping, since the (probably) known structure of the metabolic pathway under study might suggest regulatory control points. As an example of this approach, flavonoid profiles (from targeted HPLC analysis) of apical tissues were used for mQTL mapping in two full‐sib families of poplar, and three mQTLs tentatively shown to map to enzymes of the flavonoid pathway (Morrell et al., 2006).
E. ADVANTAGES OF MARKER‐ASSISTED BREEDING Marker‐assisted selection (MAS) is a complementary technology which expedites the conventional methods of genetic selection for plant and animal improvement. In classical plant breeding systems, many cycles of selection and backcrossing are required to obtain a desirable genetic gain. Besides, classical breeding is mostly successful for dominant traits which are easily inherited in subsequent generations. However, genetic gains from classical breeding methods in major crop species have reached an apparent plateau. The use of molecular markers associated with qualitative and quantitative traits has been successfully used for the indirect selection of genes of interest. The advantages of MAS include ability to reveal sites of variation in a DNA sequence, and accelerated progress by shortening the breeding cycle. MAS not only gives larger genetic responses but also dramatically increases the frequencies of superior genotypes as compared to PS (Liu et al., 2004). It is
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
109
particularly promising when dominant alleles are present and linked in coupling phase (Berloo and Stam, 1998). Molecular markers can alleviate complications of phenotype‐based selection, provided they cosegregate with the gene of interest. Crop production is significantly aVected by a number of biotic and abiotic stresses, responses to which are mostly controlled by many genes. Resistant cultivar development is the practical solution for many quantitatively inherited traits. Polygenic control, along with large environmental influences, largely limits the eVectiveness of PS for these traits. MAS provides an eYcient way to accelerate development of resistant varieties (Frisch et al., 1999). For example, the submergence tolerance gene Sub1 of rice was tagged with two microsatellite markers, RM219 and RM464A, and several lines were identified that were homozygous for these loci and were genetically similar to the parent M‐202 (Xu et al., 2004). To validate the major QTL for scab resistance in rice, the associated SSR markers were analyzed in the F2:3 lines of one population and in the F3:4 lines of the other (Zhou et al., 2003b). Markers from the original population were also closely associated with scab resistance in both validation populations. MAS was found to be more eVective than PS. However, the most eVective selection strategy was MAS during the seedling stage followed by PS after flowering. In another example, the eating and cooking quality of Zhenshan 97, an elite parent of hybrid rice, was developed by introgressing the Waxy gene region of Minghui63 through MAS breeding (Zhou et al., 2003a). MAS was likewise used for pyramiding three bacterial blight resistance genes (Xa5, Xa13, and Xa21) into indica rice cultivar PR106 (Singh et al., 2001). The gene combination provided a wide spectrum of resistance to the pathogen population that consisted of 23 diVerent Xanthomonas oryzae isolates. In major cereals such as rice, wheat, maize, and barley, molecular markers associated with diVerent qualitative and quantitative traits have been identified and used for MAS. Substantial use of MAS in maize, with a slower pace of uptake in wheat and rice breeding, has been observed. Large‐scale genotyping and MAS programs have been initiated through Rice CAP and Wheat CAP projects with funding from USDA, CSREES. Application of MAS to breeding programs depends on its relative cost and expected economic return. The best prospect for MAS is in multiple‐trait improvement. Excluding costs, multiple‐trait MAS can be used to increase the aggregate breeding values in quantitative characters and is expected to be more eVective than conventional selection or single‐trait MAS (Xie and Xu, 1998). Gains from MAS and PS were compared for quantitative traits in sweet corn (Yousef and Juvik, 2001). A total of 52 paired comparisons were made between MAS and PS composite populations. MAS led to significantly higher gain in 38% of the paired comparisons compared to only 4% for PS. The average gains from MAS and PS were 10.9% and 6.1%, respectively.
R. A. DIXON ET AL.
110
It was also observed that MAS was most appropriate when traits are diYcult and costly to measure and that the higher gain from MAS compensated for the higher costs. It was concluded that ‘‘incorporating DNA markers to traditional breeding programs could expedite selection progress and be cost‐eVective.’’ A RAPD marker associated with common bacterial blight resistance in a common bean population (PC50/XAN159) was transformed into a SCAR marker and used for screening a diVerent population (Yu et al., 2000). The SCAR marker was 94.2% accurate in recognizing the resistant genotype. Cost comparison of MAS with greenhouse screening indicated that MAS was about one‐third less expensive.
V.
TRANSGENESIS
A. TRANSGENESIS AS A TOOL FOR FUNCTIONAL GENOMICS Transgenesis refers to the introduction of heterologous or homologous DNA into a plant genome resulting in its stable integration and expression. The technology has played a critical role in defining the in vivo functions of plant genes. In recent years, with the rapid increase in gene sequence information, systematic transgenic approaches have been taken to characterize large numbers of genes in both reverse and forward genetic studies, particularly in model systems. Predictions of gene function based on sequence homology alone do not necessarily provide information on the exact biological role of the gene in planta (van Enckevort et al., 2005). After completion of the Arabidopsis genome sequence, at least 40% of the initial gene predictions based on computational annotation were subsequently found to be erroneous (Alonso and Ecker, 2006). As one of the key experimental methods in functional genomics, transgenesis has the advantage of revealing the direct link between gene sequence and function; such results not only further the understanding of basic biological questions, but also facilitate exploitation of genomic information for crop improvement. Transgenesis has been widely used for loss‐of‐function and gain‐of‐ function analyses of plant genes. Insertional mutagenesis using T‐DNA is one of the major tools for functional analysis that can provide a phenotype as a clue to gene function (Xu et al., 2005). T‐DNA mutant collections are commonly produced by Agrobacterium‐mediated transformation using a simple Ti plasmid carrying a selectable marker gene. If the T‐DNA inserts within the boundaries of a gene, it can alter or abolish the function of the gene. Because of the disruptive nature of randomly inserted T‐DNA, this type of mutagenesis is commonly associated with loss‐of‐function of endogenous genes. Compared with mutagenesis caused by chemical agents
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
111
[e.g., ethyl methanesulphonate (EMS)] or physical agents (e.g., fast neutrons, g‐radiation), the use of T‐DNA as a mutagen oVers the advantage of easy identification of the mutated gene. The T‐DNA not only disrupts the expression of the gene into which it is inserted but also acts as a marker for subsequent identification of the mutation (Krysan et al., 1999). When coupled with transposons, the introduction of the transposon containing T‐DNA into the plant genome allows for the simultaneous disruption of diVerent loci (Tadege et al., 2005). Large numbers of T‐DNA insertional lines have been produced in Arabidopsis and rice (Alonso and Ecker, 2006; Krysan et al., 1999; Walden, 2002; Xu et al., 2005). So far, more than 360,000 insertion sites have been mapped in the Arabidopsis genome, covering 90% of the genes. One of the most exciting uses of the near complete collection of gene‐indexed Arabidopsis mutations is the ability to carry out genome‐wide forward genetic screens (Alonso and Ecker, 2006). T‐DNA mutagenesis has its limitation in analyzing the function of redundant genes (Xu et al., 2005). Gain‐of‐function approaches such as gene overexpression and T‐DNA activation tagging are straightforward and powerful approaches for elucidating gene function. Transgenic expression of all the cDNAs found in Arabidopsis resulted in the identification of many genes conferring interesting phenotypes. EVorts were also made to overexpress all the TF genes in Arabidopsis. Because of the unique characteristics and modes of action of TFs, this overexpression strategy is considered particularly eVective in revealing gene function (Zhang, 2003). Sometimes the same gene can be found by diVerent approaches. For example, the identification of WIN1, an Arabidopsis ethylene response factor‐type TF that can activate wax deposition, was achieved by systematic overexpression of all gene sequences predicted to encode proteins sharing conserved domains with cognate TFs (Broun et al., 2004). In an independent study, the SHN1 gene, which shares the same sequence as WIN1, was obtained by screening a collection of 2000 transposon activation‐tagged lines (Aharoni et al., 2004). Functional genomics has been broadly defined to include many endeavors on a genome‐wide scale, such as transcriptional profiling to determine gene expression patterns, sequence alignment‐based comparisons to identify homologues between and within organisms, and the use of virus‐induced gene silencing to rapidly detect phenotypic eVects (Xu et al., 2005). Transgenesis studies are generally required to confirm the functions of the genes identified by these methods. While extremely useful, most other approaches to gene function are correlative and do not necessarily prove a causal relationship between gene sequence and function (Krysan et al., 1999). Sometimes unexpected results have been obtained by transgenic analysis. The overexpression of COL9, a member of the CONSTANS‐LIKE gene family, resulted in delayed flowering in Arabidopsis, which is opposite to the role that
112
R. A. DIXON ET AL.
the CONSTANS (CO) gene plays. Further analysis revealed that COL9 delays flowering possibly by antagonistically repressing the expression of CO, and concomitantly reducing FLOWERING LOCUS T (FT ) expression (Cheng and Wang, 2005). In M. truncatula, a TF gene (WXP1) related to wax biosynthesis was identified; overexpression of WXP1 resulted in improved drought tolerance in alfalfa (Zhang et al., 2005). Sequence comparison with WXP1 revealed its homologue in M. truncatula, designated WXP2. Transgenic expression of both WXP1 and WXP2 in Arabidopsis resulted in improved drought tolerance; however, the transgenic plants were opposite in their freezing tolerance, with WXP1 plants more tolerant and WXP2 plants more sensitive to freezing stress (Zhang et al., 2007).
B. CURRENT APPROACHES TO THE GENERATION OF TRANSGENIC PLANTS There are many variations of gene transfer methods to introduce transgenes into the plant genome. The most widely used methods are Agrobacterium‐ mediated gene transfer and biolistic transformation. Both have been applied to legume transformation, although the Agrobacterium‐mediated approach has been the most popular (Somers et al., 2003). A. tumefaciens is a soilborne bacterium that, in nature, is capable of inserting a discrete portion of its DNA into the genome of a wide range of dicotyledonous plants (Valentine, 2003). Most of the machinery necessary for the gene transfer resides on a tumor‐inducing (Ti) plasmid that carries two important genetic components: the T‐DNA delimited by two 25‐bp direct repeats at its ends and the virulence region (Tzfira and Citovsky, 2006). Agrobacterium‐mediated transformation systems take advantage of this natural gene transfer mechanism in plants. Two key advances, the development of binary Ti vectors and of a range of disarmed Agrobacterium strains, have made Agrobacterium transformation the first option in engineering transgenic plants (Hellens and Mullineaux, 2000). Agrobacterium‐ mediated gene transfer oVers the following advantages: (1) a significant portion of the transformants contains single copy transgenes, (2) in planta transformation without the need of tissue culture is possible in Arabidopsis, (3) numerous vector systems are now available, and (4) it is possible to transfer large DNA fragments, including bacterial artificial chromosomes (Herrera‐Estrella et al., 2005). The biolistic method was developed as a necessity to transform species initially considered recalcitrant to Agrobacterium transformation (Herrera‐ Estrella et al., 2005). Biolistics, or microprojectile bombardment, employs high‐velocity gold or tungsten particles to deliver DNA into living cells for stable transformation (Christou, 1992; Sanford, 1988). Gene delivery to plant cells and tissues by microprojectiles has led to the production of
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
113
transgenic plants from many species, particularly monocots. Because biolistic transformation is a physical process and therefore involves only one biological system, it is a fairly reproducible procedure that can be easily adapted from one laboratory to another. Biolistic transformation is the only reliable method for chloroplast transformation. The main disadvantage of this method is the frequently occurring multiple copy integration. Since the successful creation of transgenic Nicotiana and Petunia in the early 1980s, Agrobacterium‐mediated transformation has been the method of choice in producing transgenic plants in a wide range of dicot species. Although it was initially considered impossible to transform monocot species with Agrobacterium, transgenic plants have been obtained with many monocot crops since the mid 1990s, including major cereals like rice, maize, wheat, barley, and a number of forage and turf grasses (Cheng et al., 1997, 2004; Hiei et al., 1994; Ishida et al., 1996; Tingay et al., 1997; Wang and Ge, 2006). The cost associated with meeting regulatory requirements is a substantial impediment for the commercialization of transgenic crops (Bradford et al., 2005). In recent years, public concern about the extent to which transgenic crops might diVer from their traditionally bred counterparts has resulted in revised molecular strategies and choices of genes (Rommens et al., 2004). Due to the nature of popularly used promoters, vectors, and selectable markers for plant transformation, most transgenic plants contain DNA from multiple organisms. It has been proposed to categorize GMOs into diVerent classes based on the genetic distance between the target organism and the source of the transgenes (Nielsen, 2003). In an attempt to address some of the public perception issues relating to introduction of foreign DNA into plants, it has been shown that certain host plant DNA sequences can function in the same way as the Agrobacterium T‐DNA border sequences (Rommens et al., 2004). By incorporating such sequences to guide integration of the inserted transgene, and linking a positive selection for temporary expression of the selectable marker with a negative selection against its integration, it was possible to produce transgenic potato plants with reduced expression of tuber‐specific polyphenol oxidase that contain no foreign DNA (Rommens et al., 2004). Driven by the complexity of intellectual property issues that limit the use of Agrobacterium in both public and private sectors, several species of bacteria outside the Agrobacterium genus have been modified to mediate gene transfer to diVerent plant species (Broothaerts et al., 2005). These plant‐associated symbiotic bacteria, including Rhizobium species NGR234, S. meliloti, and Mesorhizobium loti, were made competent for gene transfer by acquisition of both a disarmed Ti plasmid and a suitable binary vector. Tobacco, Arabidopsis and rice were infected by these bacteria and transgenic plants were obtained. Of the bacteria used, at least S. meliloti is competent to transfer genes into both dicot and monocot plants and into a range of tissues, including leaf tissue, undiVerentiated calli, and immature ovules
114
R. A. DIXON ET AL.
(Broothaerts et al., 2005). The results suggest that non‐Agrobacterium species are capable of the full range of genetic transformation mechanisms shown by their Agrobacterium counterparts. This alternative approach to the Agrobacterium‐mediated gene transfer method, in addition to aVording an ‘‘open source’’ platform for plant biotechnology, may lead to new uses of natural bacteria–plant interactions for crop improvement. Although in general it is desirable to have the transgene integrated into the nuclear DNA, in some cases the plastid genome may be an appropriate target for transformation. The advantage of plastid transformation includes high transgene expression levels, increased biosafety because of maternal inheritance of cytoplasmic genomes in most crops, and lack of gene silencing and position eVects (Bock and Khan, 2004; Maliga, 2004). Transplastomic lines have been mostly produced by biolistic transformation, although direct gene transfer to protoplasts has also been utilized. The expression of a Bacillus thuringiensis (Bt) toxin gene in the tobacco plastid genome yielded high levels of the Bt toxin protein (3–5% of the total soluble protein) and produced plants with high‐level resistance to herbivorous insects (McBride et al., 1995). The expression and accumulation of the human growth hormone somatotropin in transgenic tobacco plastids reached 7% of total soluble protein, and demonstrated the capacity of chloroplasts to allow correct formation of disulfide bonds in a protein of eukaryotic origin (Staub et al., 2000). Although several successful examples of plastid engineering have set a foundation for various future applications, the adaptation of plastid transformation protocols for major food crops has proved significantly more diYcult than initially anticipated (Grevich and Daniell, 2005; Maliga, 2003). After the first successful transformation in tobacco, it took almost 10 years before plastid transformation was achieved in two other Solanaceae species, potato (Sidorov et al., 1999) and tomato (Ruf et al., 2001). Engineering of plastids oVers great promise for the production of edible vaccines, antibodies, and other pharmaceutical proteins in plants.
C. STRATEGIES FOR OVERCOMING RECALCITRANCE OF CROP SPECIES TO GENETIC TRANSFORMATION To date, most transformation procedures involve certain tissue culture steps, particularly callus culture. It is well known that callus induction and plant regeneration from the induced callus is not only time consuming and laborious but also causes somaclonal variation (Bregitzer et al., 1998; Goldman et al., 2004; Spangenberg et al., 1998). Tissue culture‐based methods also generally require considerable training of the practitioner to develop the skills needed to generate suYcient numbers of transgenic plants (Somers et al., 2003). In addition, transformation frequency varies significantly with
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
115
the genotype used, even in extensively studied species such as wheat and maize. In many cases, commercial cultivars may be diYcult to transform, and crossing the initial transgenic with an elite line, followed by significant backcrossing, will be needed for cultivar development. Tremendous eVorts have been made to overcome recalcitrance of crop species to genetic transformation. There have been numerous reports and significant progress on optimizing tissue culture and transformation parameters such as modifying media composition, growth regulators, and culture conditions, and identifying or manipulating more highly virulent Agrobacterium strains. Because the usefulness of the results is often limited to the species or even the genotype tested, it is impractical to summarize such optimization work here. However, it is worth noting that transformation eYcacy can be significantly improved by minimizing tissue culture steps. The use of cotyledonary explants in white clover, soybean, and M. truncatula bypassed the callus formation phase and allowed direct regeneration from the infected explants; such a procedure at least partially solved the genotype dependence problem and allowed rapid production of transgenics (Larkin et al., 1996; Olhoft et al., 2003; Wright et al., 2006). The use of stolons as explants in some grass species also bypassed callus formation and accelerated the process of plant regeneration (Ge et al., 2006; Wang and Ge, 2005). Another strategy to increase transformation frequency is to improve the tissue culture response. In the model grass plant Lolium temulentum, screening of a large number of genotypes revealed a few lines with relatively better callus induction frequency (Wang et al., 2002). Crosses were made between the selected lines, and a significant improvement in tissue culture response of L. temulentum was achieved by the production of haploid and double haploid lines from anthers of F2 plants of the crosses (Wang et al., 2005c). By using the highly tissue culture responsive doubled haploid line, a large number of fertile transgenic L. temulentum plants were produced by Agrobacterium‐ mediated transformation (Ge et al., 2007). The most successful story in plant transformation is the development of the nontissue culture approach for Arabidopsis. Generation of transgenic lines by in planta transformation is simple and routine (Bent, 2000; Clough and Bent, 1998). The impact of the high throughput method on Arabidopsis research has been truly remarkable. Studies on the mechanism of transformation revealed that ovules are the primary target for Arabidopsis in planta transformation by the floral dip method (Bechtold et al., 2000; Desfeux et al., 2000; Ye et al., 1999). On the basis of the lack of success to date, much time and eVort will likely be needed to develop similar transformation methods for other species. A diVerent approach to overcoming recalcitrance to transforamtion is to understand in detail the molecular basis of the T‐DNA transfer process from Agrobacterium to the plant genome. Although outside the scope of this
116
R. A. DIXON ET AL.
article, much progress has been made in this area, especially as regards the proteins that interact with the T‐DNA during its transfer (Anand and Mysore, 2005; Gelvin, 2003a; Tzfira and Citovsky, 2006). Forward genetic screens have been performed in Arabidopsis to identify T‐DNA‐tagged lines that are resistant to Agrobacterium‐mediated transformation (Gelvin, 2003a). As a result, several plant genes have been identified which, if overexpressed, increase transformation frequency. These include histone H2A‐1, VIP1 (necessary from nuclear import of T‐DNA), and a protein that interacts with the Agrobacterium VirB2 (T‐pilus) protein. More details are provided elsewhere (Gelvin, 2003a,b; Tzfira and Citovsky, 2006). It has also been shown that the yeast Rad54 protein, which is involved in chromatin remodeling, improves transformation eYciency when expressed in Arabidopsis (Shaked et al., 2005). These exciting results hold promise for the generation of high transformation eYcient plant lines of many species that currently exhibit recalcitrance to transformation.
D. TRANSGENESIS FOR TRAIT INTEGRATION AND COMMERCIALIZATION Over the last decade, transformed plants have moved from laboratory to the field, where new transgenic cultivars are grown in large acreages throughout the world. The adoption of transgenic crops has experienced double‐digit growth rates every single year since biotech crops were first commercialized in 1996, with the number of biotech countries increasing from 6 to 21 in the same period (James, 2005). The global biotech crop area has seen a remarkable increase of more than 50‐fold in the first decade of commercialization, with 90 million hectares planted in 2005. The accumulated global biotech crop area in its first decade was 475 million ha or 1.17 billion acres (James, 2005). The United States has been the biggest adopter of transgenic crops, with 49.8 million hectares planted in 2005, which represent 55% of the global biotech area. By 2005, herbicide‐tolerant soybeans accounted for 87% of total US soybean acreage, herbicide‐tolerant cotton was planted on 60% of total cotton acreage, insect‐resistant cotton accounted for 52% of cotton acreage, and insect‐resistant corn was planted on 35% of the total acreage (Fernandez‐Cornejo and Caswell, 2006). Farmers continued to choose biotech crops due to significant benefits, including enhanced crop yields, improved insurance against pest problems, reduced pest management costs, decreased pesticide use, and overall increase in grower returns (Sankula, 2006). Planted acreage has mainly concentrated in the following trait–crop combinations: herbicide‐resistant alfalfa, canola, corn, cotton, and soybean; insect‐resistant corn, cotton, rice, and sweet corn; virus‐resistant squash and papaya. Obviously, in the first‐generation transgenic crops, herbicide tolerance has consistently been the dominant trait, followed by insect resistance and virus
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
117
resistance. Some new cultivars have stacked genes for herbicide tolerance and insect resistance. The initial strategies for introducing single gene traits have been very successful in developing novel transgenic cultivars. Current strategies for plant improvement have moved toward engineering more complicated traits, such as stress tolerance, yield potential, and growth rate. With rapid advances in functional genomics, many new genes have been discovered and functionally tested. DiVerent approaches have been developed to manipulate complex traits or engineer metabolic pathways. In some cases, overexpression or disruption of a single gene can lead to the required phenotypic change; for example, drought or cold tolerance can be improved by the expression of a single TF gene (Gilmour et al., 2000; Kasuga et al., 2004; Zhang et al., 2005), or lignin biosynthesis can be modified by downregulation of a gene coding for one of the key enzymes in the pathway (Guo et al., 2001a; Reddy et al., 2005; Section VI.B.I). Other cases, such as the production of b‐carotene in rice, require the introduction of multiple genes (Paine et al., 2005; Ye et al., 2000). The need for multigene transformations has long been cited as a negative factor for the development of metabolically engineered plants. However, this technical obstacle is gradually being overcome. Up to nine transgenes have now been incorporated simulatenously, into rice by a cotransforamtion strategy (Wu et al., 2002), and into Brassica juncea by both stepwise engineering and through an actual nine‐gene construct (Wu et al., 2005). In the latter case, the genes formed a complete biosynthetic pathway to polyunsaturated fatty acids. In cases where constitutive expression of a particular transgene throughout plant growth and development has deleterious eVects, a number of promoter systems are now available for chemically induced transgene expression (Tang et al., 2004). The number of encouraging scientific reports and the range of transgenic materials currently undergoing field testing are truly extensive. Transgenic technologies have proven utility for improving disease resistance, yield potential, abiotic stress (drought, cold, salinity, aluminum) tolerance, nutrient use eYciency, feed quality for animals, processing properties of biofuel crops, and nutritional quality (increased protein and oil content), for delaying ripening, for modifying starch content, and for producing nutraceuticals (vitamins, iron, b‐carotene, flavonoids) and pharmaceuticals.
E. VIRUS‐INDUCED GENE SILENCING AS AN ALTERNATIVE TO STABLE TRANSFORMATION FOR FUNCTIONAL GENOMICS The discovery of the importance of nonprotein‐coding RNAs (ncRNAs) in the regulation of cellular processes has been one of the most important breakthroughs in genetics. One aspect of this field was recognized by the award of the 2006 Nobel Prize for Medicine to Craig Mello and Andrew Fire, for their pioneering work on gene silencing by RNA interference (RNAi),
118
R. A. DIXON ET AL.
a process which involves small RNA molecules to target the destruction of transcripts destined for turnover. In fact, it was research on plant systems that provided some of the first evidence for gene silencing (Jorgensen, 1995; Mueller et al., 1995). These discoveries are encapsulated in two practical approaches for the downregulation of genes: RNAi and virus‐induced gene silencing (VIGS). As used in plants, RNAi usually involves stable transformation with a gene construct that, when expressed, produces a small double‐stranded RNA homologous to a portion of the target gene sequence. This is usually generated via an inverted repeat of the short target sequence interrupted by a plant intron sequence (Wesley et al., 2001). This approach has been widely used for modifying a number of plant traits through targeted downregulation of a specific gene or genes. Examples include engineering altered flower color (Ono et al., 2006) and nutritional quality (Davuluri et al., 2005). VIGS takes advantage of an endogenous plant defense mechanism against virus infection which, in plants simply infected by the virus, targets the viral genome for degradation (Lu et al., 2003). Virus‐based vectors have been designed in which a small portion of the target gene sequence is included; the gene silencing process is then targeted against the corresponding host mRNAs. Although mechanistically similar to RNAi, VIGS has two major advantages over stable RNAi transformation in its high throughput and speed of the response. The one disadvantage is lack of universal application due to species specificity of suitable viral vectors. Most studies with VIGS have used Nicotiana benthamiana as host with a tobacco rattle virus (TRV)‐ based vector. N. benthamiana is universally susceptible to most viruses, and this may be because it lacks one component of the pathway for generating the small silencing RNA molecules (Yang et al., 2004). N. benthamiana is a good model for other Solanaceous species, particularly tomato, although VIGS also works quite well in this species (Ryu et al., 2004). A Brome mosaic virus (BMV)‐based vector has been developed for VIGS applications in grasses such as tall fescue and rice (Ding et al., 2006). This was based on a strain of BMV that was serendipitously discovered in a tall fescue breeding population (Mian et al., 2005b). In classical VIGS, the virus is simply physically inoculated onto the leaves. It is sometimes more eYcient to introduce the virus vector via an Agrobacterium‐ based binary vector, and this process, called ‘‘agroinoculation,’’ is usually performed through leaf infiltration with Agrobacterium harboring the necessary constructs (Dinesh‐Kumar et al., 2003). It was shown that the physical inoculation step can be avoided, and the soil adjacent to the plant roots is simply drenched with the Agrobacterium suspension containing the TRV‐based VIGS vector (Ryu et al., 2004). This ‘‘Agrodrench’’ technique provides a rapid approach for high throughput and large‐scale analysis of gene function, but is currently limited to Solanaceous species (Ryu et al., 2004).
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
119
F. TILLING AS AN ALTERNATIVE TO TRANSGENESIS FOR GENE KNOCKDOWNS TILLING has the unique advantage of allowing the generation of an allelic series for potentially any target gene. It can significantly expedite the crop improvement process. As it is a nontransgenic approach, resulting crop varieties are not subject to the strict regulatory approval process for transgenic crops. It has been used for improving the oil and protein content of soybean with the ultimate target of making allergen‐free soybeans (Comis, 2005). The applicability of TILLING to soybean, maize, romaine and iceberg lettuce, tomato, rice, peanut, bread and durum wheat, and castor has been successfully demonstrated (Slade and Knauf, 2005). However, if a transformation system is available for a crop and only a few genes are targeted for knockout, RNAi is still the method of choice. Besides, RNAi has the advantages of knocking down the expression of multiple related genes with one construct (Lawrence and Pikaard, 2003).
VI.
CASE STUDIES FOR ALFALFA IMPROVEMENT A. INTRODUCTION
Alfalfa is the most widely used forage legume crop in the world today due to its high biomass yield (the record is over 18,000 kg ha1 of forage); high protein, energy, vitamin, and mineral feed quality for livestock; ability to fix atmospheric nitrogen; wide adaptation to various environments; improvement of soil composition when used as a rotation crop in sustainable agricultural systems; utility as a model system for genetic studies of autotetraploid species; and ease of use with the new biotechnologies (Bouton, 2001). The primary center of origin for the genus Medicago is found in the Caucasus, northwestern Iran, and northeastern Turkey. M. sativa is a complex of several perennial subspecies, both diploids and tetraploids that are interfertile and possess a similar karyotype (Quiros and Bauchan, 1988). M. sativa ssp. sativa, M. sativa ssp. falcata, and M. sativa ssp. glutinosa are tetraploid subspecies while M. sativa ssp. coerulea (progenitor to cultivated alfalfa) and some M. sativa ssp. falcata are diploids. Cultivated alfalfa (ssp. sativa) is an autotetraploid with 2n ¼ 4x ¼ 32 (Stanford, 1951). Alfalfa genetics are complex because of the plant’s autotetraploid nature (Stanford, 1951) and an allogamous breeding system that does not tolerate inbreeding. The development and use of molecular markers is limited due to the diYculty of resolving allele dosage and linkage phases in autotetraploids. For this reason in the past, genetic linkage maps have been developed in
120
R. A. DIXON ET AL.
diploid forms of the M. sativa species complex (Brummer et al., 1993; Echt et al., 1993; Kalo et al., 2000; Kiss et al., 1993; Tavoletti et al., 1996). However, the utility of such diploid genetic maps in the breeding of tetraploid alfalfa depend on high synteny across the ploidy levels. In addition to serving as an important framework genetic linkage map in Medicago species, the diploid maps are also useful in transferring unique genes from the diploid level to the cultivated, tetraploid level. The widespread natural occurrence of restitutional 2n gametes (i.e., gametes with the somatic chromosome number) in the Medicago sp. (Bingham, 1968; Harlan and deWet, 1975; Stanford et al., 1972) provided early support for the hypothesis that gene flow across diVerent ploidy levels occurs continuously and naturally via 2n pollen. Such gene transfer via the restitutional gametes can aid in transferring valuable traits from diploid relatives into cultivated alfalfa in a breeding program (Bingham, 1980). The complexities of alfalfa genetics are of less concern for initial trait insertion through transgenic approaches, but do impact segregation frequencies during subsequent introgression into elite cultivars.
B. IMPROVEMENT OF ALUMINUM TOLERANCE 1.
Introduction
The use of alfalfa is mainly confined to the temperate areas of the world and not the tropics. There are several reasons for this, but the main one is an inability to tolerate acid, aluminum toxic soils that are widespread throughout the tropics (Bouton, 2001). Acidity is common in soils where rainfall is high enough to leach appreciable amounts of exchangeable bases from the soil surface layers (Brady, 1974). This leaching eVectively removes the buVering capacity of the soil and causes a drop in pH. Leaching also encourages acidity by allowing percolation of organic acids derived from naturally decomposing organic matter into the soil profile to replace the bases which are then removed by the drainage water. Under very acid conditions, Al becomes soluble in soil and is present in the toxic Al3þ or Al(OH)2þ forms (Brady, 1974). These then become adsorbed, even in preference to hydrogen ions, to clay minerals, with the adsorbed Al coming into equilibrium with the Al ions in the soil solution. The latter also contribute to overall soil acidity. When soil pH is moved toward neutrality with liming, the toxicity of Al is suppressed by changing to less toxic forms such as Al(OH). Al toxicity occurs by definition when the ratio of extractable Al (found in the toxic forms at low pH) to extractable Al plus exchangeable Ca, Mg, and
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
121
K is greater than 60% within 50 cm of the soil surface. On the basis of this definition, Al toxicity is estimated to be present in 56% of the soils in the humid tropics (Buol and Eswaran, 1993). Some general level of Al tolerance will be necessary in most crops if these extensive areas are to be brought into some level of productivity. The most common eVect of Al on plant growth is the reduction of root elongation and proliferation, thereby leading to poor water and nutrient extraction (Buol and Eswaran, 1993). Exposure of plants to Al‐toxic conditions causes inhibition of cell division at the root apex resulting in stunting of primary roots and inhibition of lateral root formation (Ryan et al., 1993; Sivaguru and Horst, 1998). Al‐sensitive plants are thus impaired in nutrient and water uptake, and tend to be drought susceptible with reduced crop yield and quality. Application of expensive soil amendments such as lime and organic acids raises the pH and converts Al into less toxic forms. However, even where liming is practiced, subsoils remain acid and Al‐toxic. A cost‐ eVective alternative is growing Al‐tolerant cultivars in problem soils combined with soil amendments (Foy, 1988).
2.
Marker‐Assisted Breeding for Al Tolerance in Alfalfa
Screening and selecting cultivated alfalfa for tolerance to acidic, Al‐containing soil has been reported (Baligar et al., 1989; Bouton, 1996; Dall’Agnol et al., 1996). Nevertheless, there is no M. sativa subsp. sativa cultivar or plant introduction that does not suVer a decline in performance under acid conditions. Conventional breeding to develop Al‐tolerant germplasm in alfalfa is also limited (Bouton and Parrott, 1997). However, two genomic regions associated with Al tolerance were identified in a diploid (M. sativa subsp. coerulea) genotype using RFLP markers in conjunction with a callus growth bioassay by single marker analysis (Sledge et al., 2002). A study was conducted to identify SSR markers that flank these QTLs using M. truncatula EST‐SSR markers (Section IV.C.1 above) and also to identify additional Al‐tolerance QTLs in a backcross population derived from the cross between Al‐sensitive and Al‐tolerant genotypes of M. sativa subsp. coerulea (Narasimhamoorthy et al., 2007). The ultimate goal underlying QTL mapping is often to identify the specific genes responsible for phenotypic variation. One method for doing this is the placement of candidate genes associated with a desirable phenotype from other species on to genetic maps to look for coincidence of map position. The M. truncatula EST and genome databases were mined to identify DNA sequences with high homology to Al tolerance genes identified in other plant species, to be used as candidate genes for genetic mapping in diploid alfalfa (Narasimhamoorthy et al., 2007). Fifteen candidate genes selected for
122
R. A. DIXON ET AL.
candidate gene mapping included those that coded for proteins responsible for organic acid synthesis, genes involved in signal transduction, and genes that code for enzymes that alleviate oxidative stress. Evidence from other crop species supports Al‐activated release from roots of carboxylates such as citric and malic acids as a major resistance mechanism (Kochian et al., 2005). An intron‐targeted mapping strategy was adopted for two specific genes involved in Al‐activated root carboxylate release, namely citrate synthase (CS) and malate dehydrogenase. Six candidate gene markers designed from M. truncatula ESTs that showed homology to known Al‐tolerance genes identified in other plant species were placed on the QTL map. Three putative QTLs on linkage groups LG I, LG II, and LG III, explaining 38%, 16%, and 27% of the phenotypic variation, respectively, were identified. A marker designed from a candidate gene involved in malic acid release mapped near a marginally significant QTL on LG I. In order to move the Al tolerance QTL from the diploid (2x) M. sativa subsp. coerulea genotype to the cultivated tetraploid (4x) M. sativa subsp. sativa, 2x‐4x crosses were made, using the diploid as the seed parent, and the tetraploid as the pollen parent. Alfalfa is essentially a bivalent forming autotetraploid where regular meiotic stages predominate in normal plants. Whereas the predominating gametes are mostly normal, restitutional 2n gametes occur at a low frequency in diploid alfalfa plants and are usually functional in fertilization events that involve tetraploid forms of cultivated alfalfa. The 2x‐4x hybrids were genotyped to confirm the presence of markers linked to Al tolerance QTLs and phenotyped for tolerance to Al stress for further confirmation of their tolerance (Fig. 4A). These tetraploid hybrids were backcrossed to selected genotypes from the nondormant alfalfa cultivar ‘‘CUF 101.’’ The BC1F1 plants were genotyped to select plants carrying the markers for Al tolerance QTLs which can be further backcrossed to elite clones. In addition to the markers linked to Al tolerance QTLs, a genome scan approach that randomly selects genetic markers spread over each linkage group to select for the cultivated alfalfa background should reduce the time involved in rigorous PS for cultivated alfalfa phenotypes during further backcrossing to introgress the Al tolerance QTLs. A synthetic population can be developed after 3–4 backcrosses to select plants that carry the Al tolerance QTL from the diploid alfalfa into the elite tetraploid cultivar background. These can then be tested in both greenhouse and field for acid soil tolerance.
3. A Transgenic Approach to Al Tolerance in Alfalfa As outlined earlier, Al‐induced secretion of organic acids from the roots has been proposed as a mechanism for Al tolerance (Delhaize and Ryan, 1995; Kochian, 1995; Ma, 2000). Two general patterns of Al‐stimulated eZux of organic acids have been reported. In Pattern I, no discernible
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
123
Figure 4 Improvement of aluminum tolerance in alfalfa. (A) The root systems of 8‐week‐old plants of diVerent alfalfa clones obtained through marker‐assisted breeding. (a) CUF 101 in limed soil, (b) CUF 101 in unlimed soil, (c) 2x‐4x hybrid in limed soil, and (d) 2x‐4x hybrid in unlimed soil. (B) Root systems of control and transgenic alfalfa lines after 8‐week growth. (a) Regen‐SY (nontransformed) in unlimed soil, (b) Regen‐SY in limed soil, (c) Regen‐SY transformed with a CS gene in unlimed soil, and (d) Regen‐SY transformed with a CS gene in limed soil.
delay is observed between the addition of Al and the onset of organic acid release, suggesting that Al activates a preexisting mechanism without a need for induction of novel proteins. In Pattern II, organic acid secretion is delayed for several hours after exposure to Al3þ, indicating that protein induction is required (Ma et al., 2001).
R. A. DIXON ET AL.
124
Alfalfa is very sensitive to Al3þ and its yield and stand duration in acid soils are compromised due to both inhibited root system development and reduced symbiotic nitrogen fixation. Increasing production of Al‐chelating compounds, particularly organic acids, in plant roots through a transgenic approach could enhance tolerance to Al toxicity. In the first report of this approach, a CS gene from Pseudomonas aeruginosa was expressed in tobacco and papaya plants (da la Fuente et al., 1997); this led to increased citric acid production and tolerance to Al toxicity. Likewise, in alfalfa, the overexpression of malate dehydrogenase (MDH) in transgenic plants enhanced tolerance to Al toxicity through increased organic acid synthesis (Tesfaye et al., 2001). In a similar approach, the Arabidopsis ACT2 constitutive promoter or the tobacco RB7 root‐specific promoter were used to drive the CS gene in alfalfa (Rosellini et al., 2003). Transgenic plants expressing the CS gene possessed better root growth and total dry matter yield than control plants in Al‐toxic soils (Fig. 4B), and also had longer roots when grown in a medium containing Al. This approach can be fine‐tuned to increase the production of citric acid in specific root tissues, and the CS transgene can be pyramided with Al tolerance QTLs by crossing the transgenic plants with plants known to possess molecular marker‐tagged Al tolerance QTLs.
C. GENE DISCOVERY AND METABOLIC ENGINEERING FOR FORAGE QUALITY ENHANCEMENT 1.
Introduction
Forage quality is a major but complex trait for plant improvement. Generally speaking, quality decreases as plants mature and enter the flowering stage, primarily as a result of lignification of secondary cell walls (Jung and Vogel, 1986). Other quality traits include protein content and amino acid composition, protein and nutrient bioavailabilty, presence of chemical antifeedants, and potential for causing pasture bloat. We here describe studies designed to better understand, and improve, forage quality traits in alfalfa. This provides another illustration of how some of the postgenomics technologies and resources described earlier, in this case those derived for/from Arabidopsis and M. truncatula, can be applied for variety improvement.
2.
Improved Forage Digestibility
Feeding and grazing studies have shown that only small changes in forage digestibility can have significant eVects on animal performance (Casler and Vogel, 1999). Improving digestibility is therefore an important goal of forage
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
125
breeding programs, and has, in the past, been addressed primarily through standard approaches of crossing and selection (Jung et al., 1994). At the same time, this research has led to an understanding of relationships between various forage quality parameters, such as neutral detergent fiber (NDF), acid detergent fiber (ADF) and acid detergent lignin (ADL), and forage quality (Jung, 1997; Jung et al., 1997). There is a general consensus that high lignin levels reduce digestibility (Jung and Deetz, 1993). However, most of the studies from which this conclusion is based used materials with diVerent lignin contents and/or compositions as a result of divergent selection for forage quality traits (Me´chin et al., 2000), natural genetic variation of plant accessions (Casler, 1987), delignification (Jung et al., 1992), or diVerent maturity of plant tissues (Reeves, 1987). The results of such studies will be complicated by the many uncontrolled developmental and genetic variables that could potentially aVect digestibility (Titgemayer et al., 1996). Because alfalfa is an outbreeding autotetraploid, inbred lines are not available. Generating isogenic transgenic lines in which lignin content or composition are modified by altering expression of a target gene in the lignin pathway provides both a new approach to trait improvement and a strategy for better elucidating the lignin/digestibility relationship in alfalfa. The value of such an approach, for both basic and applied research, becomes more apparent when considering the potential variations that can exist in lignin structure. Lignin is a polymer of hydroxylated and methoxylated phenylpropane units (monolignols) linked via oxidative coupling (Boudet et al., 1995). There is a vigorous ongoing debate as to the extent, or lack, of orderliness in the polymer (Davin and Lewis, 2005; Ralph et al., 2007), with the prevailing view being that lignin is assembled by a relatively random free radical‐mediated process, that is nevertheless under developmental control at the level of substrate supply (Boerjan et al., 2003). Angiosperm lignin contains two major monolignols, mono‐methoxylated guaiacyl (G) and di‐methoxylated syringyl (S) units, polymerized through at least five diVerent linkage types (Boerjan et al., 2003). It also contains low levels of p‐hydroxyphenyl, or H units (Fig. 5). In many forage crops, lignin content and S/G ratio increase with stem maturity (Buxton and Russell, 1988; Jung and Vogel, 1986), and both content and S/G ratio therefore correlate negatively with forage digestibility in ruminant animals (Albrecht et al., 1987; Buxton and Russell, 1988; Grabber et al., 1992; Jung et al., 1997; Sewalt et al., 1996). However, the relationship between lignin composition and digestibility is far from clear, since the amount of G lignin has also been linked with reduced cell wall degradability in forages (Jung and Deetz, 1993), and studies with synthetic lignins (Grabber et al., 1997) have yielded results that question eVects of lignin composition alone on forage digestibility. The biosynthesis of the monolignol building blocks of lignin is believed to proceed essentially according to the pathway in Fig. 5 (HoVmann et al., 2003;
O
126
H2N
A
Phenylalanine
OH
Cinnamic acids
PAL O
O
C4H HO
OH
Cinnanic acid
Cinnamoyl CoAs O
4CL HO
OH
Cinnamyl aldehydes O
CCR HO
SCoA
CH2OH
CAD HO
H
p-Coumaroyl CoA p-Coumaryl aldehyde HCT
p-Coumaric acid
O HO
HO
H lignin units
p-Coumaryl alcohol
C
Shikimate
C3H
O Shikimate
HO
HCT O TC108382
TC106914
TC108383
TC94321
TC107427
TC107428
HO
SCoA
HO 16 14 12 10 8 6 4 2 0
O HO MeO
O
CCRHO MeO
Feruloyl CoA
H
d c D ell c ro ug ultu ht re st re (in ss cl Le ud Flo i w n a Fu f a g s er ng nd ee Ph al co ds os ly i ty ) ph nfe led In ate cte on se -s d ct tar lea -d ve am d f ag lea ed f le af N od ul Ro a o Fu M yc ted t n N ga or em lly hi roo at in zal t o fe r N de- cte oot it in d Ph rog fec ro os en- ted ot s U pha tar roo Vv t irr e-s ed t ad ta r ia rve oot te d dr se oo M e dl t Jin el Vir gs u ic ite s-i nf St d ro ec em ot te su d l sp ea en f si on
TC108382
HO MeO
CH2OH
MeO
COMT
ds
Po
HO
H
MeO MeO
Coniferyl alcohol
HO
O
HO
G lignin units
MeO
F5H
F5H HO
CH2OH
CAD HO
Coniferyl aldehyde
TC94321
ite ic El
SCoA
COMT MeO
O H
CAD HO
Sinapyl aldehyde
MeO
CH2OH
Sinapyl alcohol
S lignin units
R. A. DIXON ET AL.
B
Cinnamyl alcohols
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
127
Humphreys and Chapple, 2002; Humphreys et al., 1999; Nair et al., 2004; Schoch et al., 2001), although it is possible that there are variations between species, especially as regards pathway regulation and whether linear or parallel pathways exist for G and S lignin synthesis (Chen et al., 2006b; Parvathi et al., 2001). As a result of the genome‐sequencing projects outlined in Section II, and a number of EST programs in other species, genes encoding all of the enzymes in Fig. 5 have been identified from representative monocots and dicots. Most important for forage species are the EST collections from Medicago, tall fescue, and perennial ryegrass. Because of the very high sequence identity between orthologous genes in M. truncatula and M. sativa (Aziz et al., 2005), the genetic resources from M. truncatula have proven very useful for identifying lignin pathway gene sequences that can be applied directly for genetic modification of lignin in alfalfa. An example is provided in Fig. 5. Searching the DFCI MtGI (http:// compbio.dfci.harvard.edu/tgi/cgi‐bin/tgi/gimain.pl?gudb¼medicago) reveals five TC sequences whose BLAST annotation suggests that the gene might encode a caVeic acid 3‐O‐methyltransferase (COMT), the enzyme that carries out the final methylation step in the formation of S lignin (Fig. 5). By counting the number of individual ESTs corresponding to each TC in each of the more than 60 cDNA libraries sequenced to date, it is possible to obtain an estimate of the relative degree of expression of each of the TCs in diVerent tissues (such an approach is called an ‘‘in silico Northern,’’ after the ‘‘Northern’’ blot hybridization technique for measuring transcript levels). Similar information could be obtained from microarray analysis of specific tissue types. It is clear from Fig. 5B that only TC 94321 is strongly represented in stem tissue. This TC is therefore most likely the true COMT involved in lignification. The same approach was taken to identify M. truncatula genes encoding the three cytochrome P450 enzymes of the lignin pathway, namely cinnamate 4‐hydroxylase (C4H), coumaroyl shikimate 3‐hydroxylase [also known as coumarate 3‐hydroxylase (C3H)] and coniferaldehyde 5‐hydroxylase [also known as ferulate 5‐hydroxylase (F5H)] (Reddy et al., 2005). These three
Figure 5 Application of genomics/transgenesis to lignin modification in alfalfa. (A) Currently accepted model of the lignin biosynthetic pathway. Enzymes are: PAL, L‐phenylalanine ammonia‐lyase; C4H, cinnamate 4‐hydroxylase; 4CL, 4‐coumarate:CoA ligase; CCR, cinnamoyl CoA reductase; CAD, cinnamyl alcohol dehydrogenase; HCT, hydroxycinnamoyl CoA:shikimate/quinate hydroxycinnamoyl transferase; C3H, ‘‘coumarate 3‐hydroxylase;’’ CCoAOMT, caVeoyl CoA 3‐O‐methyltransferase; F5H, ‘‘ferulate 5‐hydroxylase;’’ COMT, ‘‘caVeic acid 3‐O‐methyltransferase.’’ (B) cDNA library‐specific EST counts for all TCs annotated as encoding caVeic acid O‐methyltransferase. Note that only one TC is strongly expressed in stems, the major site of lignification. This sequence was therefore chosen for antisense and RNAi‐mediated downregulation, and the corresponding vectors introduced into alfalfa by Agrobacterium‐mediated transformation and regeneration via somatic embryogenesis (C) (Chen et al., 2006b).
128
R. A. DIXON ET AL.
enzymes catalyze strategically placed reactions in the formation of all monolignols, G‐units, and S‐units, respectively (Fig. 5). The close sequence identity between orthologous genes in the two closely related Medicago species makes it possible to make gene constructs using M. truncatula sequences to target alfalfa genes for downregulation. Downregulation of C4H, C3H, or F5H in alfalfa therefore used the corresponding M. truncatula sequences expressed in the antisense orientation (Reddy et al., 2005). The transgenes were driven by the bean phenylalanine ammonia‐lyase PAL2 promoter (Liang et al., 1989), which is expressed in most of the vascular tissues of alfalfa (Guo et al., 2001a). Transformation used eYcient methods based on cocultivation of leaf discs with Agrobacterium followed by regeneration via somatic embryogenesis (Samac and Temple, 2006) (Fig. 5C). Plants with validated reductions in target transcript and/or enzyme activity levels had either reduced lignin levels with relatively normal lignin composition (C4H transgenics), lignin rich in p‐hydroxyphenyl (H) units (C3H transgenics), or lignin rich in G units with reduced S content (F5H transgenics) (Reddy et al., 2005). Previous studies had used similar antisense technology to downregulate (COMT) and/or caVeoyl CoA 3‐O‐methyltransferase (CCoAOMT) in the same alfalfa genetic background (Guo et al., 2001a). COMT downregulation reduced both lignin content and S/G ratio, whereas lignin content was reduced, but S lignin levels remained unaltered, in CCoAOMT downregulated plants (Guo et al., 2001a). The availability of sets of transgenic alfalfa plants with various combinations of altered lignin content and composition allowed a determination to be made of the relative importance for forage digestibility of altered lignin content or composition in the same genetic background. Plants were grown to the early bud stage, harvested, and analyzed for a number of forage quality parameters, including in situ digestibility in the rumens of fistulated steers (Guo et al., 2001b; Reddy et al., 2005). These studies clearly indicated that lignin content, rather than composition, impacted digestibility, with the greatest improvement in digestibility (up to 15% in plants downregulated in C3H) being observed with the plants with the most reduced lignin levels (Reddy et al., 2005). Earlier studies, in which transgenic forage had been analyzed in fistulated sheep, also indicated that downregulation of CAD improved digestibility in alfalfa (Baucher et al., 1999), although to a lesser extent than with the transgenes described earlier. Although downregulation of C3H, and the enzyme preceding it (HCT), gives the largest digestibility improvements in alfalfa, there is no free lunch here, as these particular transgenic plants suVer from yield depression (Reddy et al., 2005; Shadle et al., 2007). The reason for this is not totally clear, although distorted vascular tissues are observed in the HCT lines (Shadle et al., 2007), suggesting that water relations may be disturbed. The digestibility increases in COMT and CCoAOMT lines of around 5% through
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
129
a single transformation event, but not associated with negative agronomic performance, represent an economically beneficial transgenic improvement. Such lines have now been extensively field tested with a view to commercialization (Temple et al., 2004). Additional research may lead to an understanding of the reduced growth phenotype of plants with strong lignin downregulation, such that this can be ameliorated either by targeting the lignin reduction to a more narrow set of cell types, or through introduction of compensatory mechanisms. In both cases, it is likely that postgenomics approaches will hold the key, by providing either promoters with a more specific cell type expression or high throughput platforms for trait identification.
3.
Bloat Resistance
Although this may appear paradoxical in view of the above discussion of improving biomass digestibility in alfalfa, the high protein content of alfalfa can cause problems for ruminant animals because it is too rapidly digested by rumen microogransims (Marshall et al., 1980). This leads to: (1) excess methane production in the rumen, exacerbated by foaming caused by a combination of the high protein content and presence of other agents such as saponins and leading to the condition known as pasture bloat; (2) increased urinary nitrogen excretion; and (3) reduced levels of ‘‘by‐pass’’ protein not exiting the rumen and therefore not contributing to the nitrogen nutrition of the animal. Designing a ‘‘bloat safe alfalfa,’’ with these other additional benefits, has been a major goal for alfalfa breeders (Coulman et al., 2000). Studies in sheep have demonstrated significant improvements in performance and reduction in bloat, if the animals are fed forages contain reasonable levels of flavonoid polymers known as condensed tannins [CTs, also called proanthocyanidins (Aerts et al., 1999)]. CTs bind to proteins and reduce their rate of microbial degradation. In laboratory studies, treatment of feed proteins with modest amounts of CTs (around 2–4% of dry matter) reduced both proteolysis during ensiling and rumen fermentation. In studies with sheep, increasing dietary CTs (from only trace amounts to 4% of dry matter) increased by‐pass protein, and a diet containing only 2% CT increased absorption of essential amino acids by the small intestine (Douglas et al., 1999). Low concentrations of CTs also help counter intestinal parasites in lambs, and, as described earlier, confer bloat safety (Aerts et al., 1999). Levels of CTs for bloat reduction are at the lower end of the range needed to significantly improve the nitrogen nutrition of the animal (Li et al., 1996). The above properties of CTs are the main driving force behind eVorts to genetically introduce the CT pathway into forage crops (Aerts et al., 1999; Reed, 1995). However, high concentrations of
130
R. A. DIXON ET AL.
CTs (from 6% to 12% of dry matter) reduce the palatability of forages, and can negatively impact nutritive value, including digestibility, by directly binding bacterial enzymes or forming complexes with cell wall polysaccharides and thereby reducing their accessibility to degrading enzymes (Aerts et al., 1999; Reed, 1995; Smulikowska et al., 2001). Alfalfa is a bloat‐causing forage because its aerial portions do not contain measurable levels of CTs. These compounds do, however, accumulate to quite high levels in the seed coat (Koupai‐Abyazani et al., 1993). Thus, alfalfa contains all the genes necessary for CT biosynthesis. The trick is to express these genes ectopically in the aerial portions of the plant. To identify the necessary genes for engineering CTs in alfalfa, it is first important to consider the chemical structures of these compounds. CTs are oligomeric and polymeric end products of the flavonoid biosynthetic pathway (Fig. 6). The past 3 years have seen important breakthroughs
Phenylalanine PAL C4H 4CL 4-Coumaroyl-CoA + 3 Malonyl-CoA CHS Naringenin chalcone CHI Chalcone F3H F3′H Dihydroflavonols DFR Leucoanthocyanidins
FLS LAR
Flavonols Catechins Proanthocyanidins
ANS Anthocyanidins
ANR
Epicatechins
Anthocyanins Figure 6 An overview of the biosynthesis of anthocyanins and condensed tannins. Enzymes are: ANR, anthocyanidin reductase or BANYULS; ANS, anthocyanidin synthase (LDOX leucoanthocyanidin dioxygenase); CHI, chalcone isomerase; CHS, chalcone synthase; 4CL, 4‐coumarate:CoA ligase; C4H, cinnamate 4‐hydroxylase; DFR, dihydroflavonol reductase; F3H, flavonoid‐3‐hydroxylase; F30 H, flavonoid‐30 ‐hydroxylase; FLS, flavonol synthase; LAR, leucoanthocyanidin reductase; PAL, phenylalanine ammonia‐lyase. The PAP1 myb TF controls expression of the genes encoding the enzymes of the anthocyanin pathway from PAL to ANS in Arabidopsis.
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
131
in our understanding of the biosynthesis of the building blocks of CTs, the flavan‐3‐ols (þ)‐catechin, and ()‐epicatechin (Fig. 6; Dixon et al., 2005; Tanner et al., 2003; Xie et al., 2003). However, virtually nothing is known about the ways in which these units are assembled into the corresponding oligomers in vivo (Xie and Dixon, 2005). Molecular genetic approaches are leading to an understanding of the regulatory genes that control CT biosynthesis, and this information, together with the increased knowledge of the enzymes specific for the pathway, will facilitate the genetic engineering of plants for introduction of value added forage quality traits. The major discovery engine for the genes specific for CT biosynthesis and its control has been the use of forward genetics in Arabidopsis. Interruption of CT biosynthesis at any stage results in the formation of a transparent testa (tt) phenotype. Many tt mutants of Arabidopsis have now been characterized, and the cloned genes that had been disrupted comprise biosynthetic enzymes of the CT pathway, TFs controlling both the pathway and endothelial cell development, transporters, a proton pump, and an oxidase (Lepiniec et al., 2006). On the basis of our results, at least three genes appear necessary for introducing the CT pathway into tissues that do not naturally make these compounds: these encode MYB family TFs functionally orthologous to Arabidopsis PAP1, which, when ectopically expressed, leads to massive accumulation of anthocyanin pigments (Borevitz et al., 2000); the MYB family TF TT2 (Nesi et al., 2001), which appears to regulate genes encoding late steps in CT biosynthesis; and the enzyme anthocyanidin reductase (ANR), which is encoded by the BANYULS gene of Arabidopsis, and converts anthocyanidins into their corresponding 2,3‐cis‐flavan‐3‐ols (e.g., cyandin to ()‐epicatechin, Fig. 6; Xie et al., 2003, 2004). PAP1 was found by a T‐DNA activation tagging approach (Borevitz et al., 2000), and was one of the first genes to be discovered in this way in view of the obvious purple‐red phenotype of plants overexpressing this gene. Discovery of the function of the Arabidopsis BANYULS gene, by analysis of the catalytic activity of the recombinant protein expressed in vitro, was soon followed by the isolation of the functional orthologue from Medicago, by utilizing EST information from a cDNA library representing transcripts from developing seeds (Xie et al., 2004). Expressing PAP1, TT2, and ANR together appeared insuYcient to allow for constitutive accumulation of CTs in Arabidopsis leaves and stems (Sharma and Dixon, 2005). In contrast, coexpression of Arabidopsis PAP1 and Arabidopsis or Medicago ANR leads to production of CTs in tobacco leaves and flowers, at levels that would be protective for bloat if they were in alfalfa (Xie et al., 2006). One limitation to moving this technology directly to alfalfa is the finding that Arabidopsis PAP1 does not appear to function well in legumes (G. J. Peel
R. A. DIXON ET AL.
132
and R. A. Dixon, unpublished results). This problem can be partially circumvented in M. truncatula, the leaves of which contain a central red spot rich in anthocyanins; simply expressing the ANR gene in these leaves therefore results in a detectable accumulation of CTs (Xie et al., 2006). We have identified two novel MYB TFs by informatic analysis of the M. truncatula genome sequence, and these confer a strong purple pigment phenotype when transformed into alfalfa, M. truncatula, or clover (G. J. Peel, E. Wright, Z. Y. Wang, and R. A. Dixon, unpublished results). Thus, we believe that the development of a bloat‐safe alfalfa by tannin engineering will soon become a reality. Previous studies have hinted at the potential for accumulating tannins in alfalfa foliage after transformation with flavonoid pathway TFs, but this accumulation required that the plants were placed under stress conditions such as cold or high light intensity (Ray et al., 2003).
D.
ISSUES FOR MOLECULAR DEVELOPMENT OF ALFALFA
In self‐pollinated crops, introgressing desirable exotic alleles from wild to cultivated backgrounds with the aid of molecular markers is a straightforward process. However, in tetraploid alfalfa with no extant inbred lines and an autoteraploid nature, such an eVort presents a much more complicated challenge. Although QTLs and transgenes for Al tolerance, and transgene sets for lignin modification and introduction of condensed tannins, are currently being introgressed into elite cultivar backgrounds, their deployment requires estimation of the QTL or transgene eVects in commercial breeding populations. In particular, the marker and QTL eVects must be estimated on a regular basis to improve accuracy and to guard against unfavorable associations with other traits and against epistatic eVects with the background genome or environment. The deployment of transgenes also requires further research to ensure stable expression and understand the eVect of the transgenes in combination with the QTLs. These considerations are, of course, apart from the evolving web of patent and regulatory issues.
VII. THE FUTURE: BRIDGING THE GAP FROM MODELS TO CROPS A long‐range goal of translational genomics is to utilize bioinformatics to leverage genomic information from model and reference organisms for the economic benefit of important crop species. Given the abundant genomic
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
133
information available for model and crops species, it is perhaps surprising that a significant gap still exists between the wealth of available knowledge about genome structure and content and its utilization by plant breeders whose programs are directed toward agronomic improvement of crop species. For example, an informal survey at the 2006 International Plant Breeding Symposium revealed that very few of the breeders polled access plant genomics resources. What are required are intuitive, web‐based tools that enable plant breeders to retrieve genomic, genetic, and phenotypic data and information that is relevant to them. An example of such an online database is the Soybean Breeder’s Toolbox (http://soybeanbreederstoolbox.org/) that allows exploration of the genomic resources through easily retrieved information. The toolbox provides information about molecular markers on genetic maps, diseases and pests that damage soybean crops, and data associated with soybean quantitative traits such as the resistance of diVerent soybean genotypes to biotic and abiotic stresses. Using comparative genomics, information from model plant species can accelerate the discovery of genes responsible for disease and pest resistance, tolerance to plant stresses such as drought, and enhanced nutritional value including production of antioxidants and anticancer compounds. A sequenced and annotated genome can accelerate the identification of candidate genetic loci underlying phenotypes of interest. Because sequence and function of genes are largely conserved among related species, comparative genomics can leverage information and knowledge gained from a sequenced model, or reference species, to make hypotheses about the relationship between genotype and phenotype for related species. With the exception of Medicago, Lotus, and now soybean, crop legumes have not been sequenced because their large and complex polyploid genomes make genome sequencing endeavors cost‐prohibitive. In regions of the genome where syntenic relationships exist between Medicago or Lotus and a crop legume, the annotated genomic sequence from these species can be leveraged to identify candidate genes of interest in other legumes. The following is an example of how the Legume Information System (LIS; http://www.comparative‐legumes.org) (Gonzales et al., 2005) can be used to find candidate genes for sudden death syndrome (SDS) in soybean via the Medicago genome. SDS, caused by Fusarium solani f. sp. glycines, creates toxins in the roots resulting in root rot and leaf scorch that severely reduces soybean production each year. SDS is a major concern and has become the focus for breeders and scientists interested in producing a more resistant soybean plant. QTLs for SDS in soybean have been previously identified and mapped (Njiti et al., 2002). Once QTL regions have been located, the actual genetic elements responsible for the phenotype may perhaps be identified. Specifically, genetic maps, physical maps, and annotated TC and EST
134
R. A. DIXON ET AL.
sequences from soybean and M. truncatula can be compared. In addition, the recently published M. truncatula genomic sequences can be used to identify M. truncatula candidate genes in a genomic region syntenic to a QTL region for SDS in soybean. Genomic sequences of candidate genes from M. truncatula can then be used to identify ESTs with sequence similarities from soybean for DNA marker development and cloning of potential soybean disease‐causing alleles. By using the CMap module of LIS to query and display soybean SDS QTLs on genetic maps, the soybean linkage maps are compared to M. truncatula maps to identify syntenic regions containing SDS QTLs. Once genomic markers in M. truncatula have been identified as syntenic to the SDS QTL region, the M. truncatula physical maps are used to identify the sequenced genomic clones within comparable regions. These genomic sequences within the physical region are then analyzed for candidate genes using annotations displayed in the LIS Comparative Functional Genomics Browser (Fig. 7). Finally, consensus sequences aligned to genomic sequence can be analyzed using the existing annotations to isolate candidate soybean EST sequences that may confer SDS resistance in soybean.
VIII. THE FUTURE TECHNOLOGIES, OPPORTUNITIES, AND CHALLENGES ‘‘Science is built up with facts, as a house is with stones. But a collection of facts is no more a science than a heap of stones is a house.’’ Jules Henri Poincare´ (1854–1912) La Science et l’hypothe´se Plant biologists who focus on crop species typically rely on integrative and comparative analyses using model and reference species. However, these scientists are increasingly challenged in translating diverse genetic, genomic (read ‘‘‐omics’’), and phenotypic information to address their long‐term research goals. These diverse data types are dispersed within a growing number of independently evolving, web‐based information resources. Quan et al. (2003) explain it well: ‘‘. . .many barriers exist between [scientists] and their data, which is scattered over dozens of machines in incompatible data stores in a myriad of formats.’’ In addition, the researcher is commonly faced with data from diVerent resources that are essentially equivalent from the biologist’s point of view. For instance, various unigene/gene index/cluster sets, produced using diVerent protocols, are available to organize transcriptome information. Second, researchers find web‐based information often requires careful data management practices on the client side (desktop).
60
Barrel medic Young (U. Minn.) 2004 4
Barrel medic Cook/Kim MtGenome v3 (UC-Davis) 1009
62
90
[86]
[52]
53
DK447R
APX 001E07_2
001E07
120
APX 30
44
021K24
DK413L 025M19 60
35
150
094F23
90
− Mag + −1−157.88 cM
26
SDS 8-2
30
[200] A121_1 Pgi A122_1 Bng132_2 0P_E08 mQ086_1 Satt281 Satt291 A655_1 Sat_153 Satt 170 GMAC7L_SSR Bng014_1 L059_1 Sat_213 Satt643 L148_1 Sat_076 A063_2 BLT053_9 K011_3 Satt365 Satt289 Sat_312 Satt319 AG36 Sat_251 Sat_142 K418_4 Satt460 K365_1 BLT029_1 Sct_028 A538_1 K474_1 Satt202 Satt433 K455_1 0005_250
120
8
17
018F06
016C13
− Mag + −1−63.80 cM
− Mag + 0−145 bands
135
Figure 7 A candidate SDS resistance gene is identified in soybean using Medicago annotated genomic sequence in an area where syntenic relationships exist between Medicago and soybean.
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
Soybean composite genetic map 2003 C2
136
R. A. DIXON ET AL.
It is diYcult to cross the boundaries between these resources beyond what is explicitly provided through links from one information resource to another, much less to download a combined set of data for follow‐on analysis or experimental testing. As biology becomes more and more an information science, addressing the problems of integrating independently managed data resources is of extreme importance. Compounding this issue will be the flood of data as next‐generation technologies, such as massively high throughput DNA sequence analysis, greatly increase the volume of data while dramatically reducing its unit cost. GenBank, for example, is a comprehensive nucleotide sequence public repository that contains DNA sequence information for more than 205,000 organisms with more than 3000 new species being added per month. GenBank has doubled in size approximately every 18 months since its inception. As of August 2006, GenBank warehouses more than 145 billion bases of nucleotide sequence. Hendler makes a compelling case that ‘‘as modern science grows in complexity and scope, there is an increasing need for more collaboration between scientists at diVerent institutions, in diVerent subareas, and across scientific disciplines’’ (Hendler, 2003). The frustrations of biological researchers suggest that the current World Wide Web is not suYcient for the needs of collaboration across scientific disciplines or, ultimately, for their daily discovery activities. For example, a Google search using the term ‘‘gene’’ will return web pages for both the scientific journal Gene and for the Hollywood actor Gene Hackman. What are required are mechanisms that imply the correct semantic meaning to search terms as intended by the plant biologist. That said, the current distributed, dynamic nature of the web is particularly suited to the emerging, and ever‐changing data types and needs of the research community. Emerging semantic web and web services technologies (Schiltz et al., 2004; Wilkinson et al., 2003) appear to provide architecture for discovery of, and access to, distributed biological data sources and analysis services. Building on XML, semantic web service technologies underlie an emerging approach to provide not only data integration, but also data interoperability, of distributed web resources. Semantic web technologies are designed to scale and evolve using computer algorithms instead of human‐developed ‘‘parsers’’ to identify, configure, compare, and combine data resources on the web. The specific requirements of a semantic web application include the use of geographically distributed information with diverse ownership (i.e., no control of evolution). The application should make use of heterogeneous information and data sources in other ways than intended by the original authors. Importantly, the application would adopt formal descriptions of the meaning of the information. Finally, the application should have a combination of static and dynamic knowledge. The Virtual Plant Information Network (VPIN; http://vpin.ncgr.org) is a National Science Foundation‐funded
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
137
collaborative project to further develop the technology framework for a web‐ based, distributed, virtual plant network, into a single semantic web services (SWS) platform. This platform allows partners to share data and invoke web processes to operate on that data. As a SWS platform, VPIN allows partners to describe their data and services, and to find data and services, based on suitable definitions understood by both computers and humans. Nucleic acid sequencing is a rapidly advancing field that has phenomenal potential to transform crop improvement strategies. Next‐generation, nonelectrophoretic DNA sequencing technologies generate data at more than 3000‐fold the rate but at 1/80th the cost of conventional capillary sequencing (Table IV). For example, Solexa’s technology generates approximately one billion bases of DNA sequence per 3‐day run with a reagent cost of 3500 US dollars. To put this into perspective, it would be possible to generate with 10 Solexa instruments the equivalent of all of the data currently in GenBank in a little more than 6 weeks! Of these new technologies, 454 Life Sciences Corporation, Branford, CT, developed the first DNA sequencing platform to employ picoliter volumes in a highly multiplexed, flow‐through array (Margulies et al., 2005). Sequencing is performed on randomly fragmented cDNA using microbead‐based pyrosequencing chemistry. This platform provides significant improvements in cost‐eVectiveness, ease of use and speed and has significant potential to fundamentally change DNA sequencing strategies for crop species. It has been used to sequence pooled RNA samples of M. truncatula (Cheung et al., 2006). DNA sequencing is used in four principal applications: (1) de novo genome sequencing to create a reference set of sequences that render a species genomically tractable; (2) gene or genome resequencing in which genes, genome segments, or entire genomes are sequenced in individuals within a population in order to undertake association studies; (3) RNA profiling or transcriptome sequencing, in which an RNA sample is converted to complementary DNA (cDNA) and sequenced in order to determine the sequence or abundance of transcripts (ESTs), for correlation with phenotypes; and Table IV Comparisons of Current and Next‐Generation DNA Sequencing Technologies ABI 3730 l Read length Reads per run Bases per run Run per day Cost per base a
To be determined.
400 96 38,400 72 0.125¢
ABI SOLiD
Solexa
454 Life sciences
2 20 (paired ends) 40 million 1 billion 1 per 5 days TBDa
35 30 million 1 billion 1 per 3 days 0.0004¢
200 0.5 million 100 million 1 per 2 days 0.012¢
138
R. A. DIXON ET AL.
(4) metagenomics, in which DNA from mixed environmental samples is sequenced in order to evaluate correlation between environmental variables and species’ abundances. De novo genome sequencing has thus far been the focus of a substantial proportion of sequencing resources. Nonelectrophoretic DNA sequencing technologies oVer cost and throughput advantages in de novo genome sequencing applications. By circumventing the need to propagate clone libraries in a living host, they avoid cloning bias associated with cloning artifacts. The emerging paradigm for use of nonelectrophoretic sequencing technologies in de novo genome‐sequencing projects is a hybrid approach that uses a combination of capillary and pyrosequencing technologies. Hybrid assemblies of Sanger and pyrosequencing reads were shown to be feasible and cost‐eVective for development of draft and finished prokaryotic genome sequences (Goldberg et al., 2006). A hybrid approach is likely to become the fundamental strategy for future analyses of crop species’ genomes. The goal of gene expression profiling experiments is typically to understand the dynamics of transcript abundance between states or temporal events in networks and pathways. Usually, this involves the identification of a set of transcripts whose expression diVers as an external parameter is varied (e.g., developmental stage, genotype, stress). Several sequence‐based transcript‐ profiling methods have been described that provide absolute counts of the number of times a transcript occurs in a sample (Kuo et al., 2006; Mikkilineni et al., 2004). These approaches appear to be extensible to nonelectrophoretic sequencing technologies (Mikkilineni et al., 2004). An RNA profiling application where nonelectrophoretic sequencing instruments are clearly the platform of choice is in the identification and characterization of small RNA molecules. Several publications have demonstrated the broad utility of 454 pyrosequencing for identifying and profiling various classes of micro and small interfering RNA molecules (Henderson et al., 2006; Lu et al., 2006). Next‐generation sequencing approaches have considerable potential to impact crop species EST, genomic and resequencing eVorts. Studies have provided an opportunity for benchmarking of a new paradigm in sequence technology for eYcient and cost‐eVective genome analysis. In the case of DNA sequence analysis, the bottleneck is no longer data generation, but data management and integration for crop improvement.
ACKNOWLEDGMENTS We thank Cuc Ly for assistance with artwork, and Dr. Yongzhen Pang for providing Fig. 7. Work described from the authors’ laboratories was supported by the Samuel Roberts Noble Foundation, and by grants from the
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
139
NSF Plant Genome Program, US Department of Energy, Oklahoma Center for the Advancement of Science and Technology, Forage Genetics International and Halliburton Energy Services.
REFERENCES Achnine, L., Huhman, D. V., Farag, M. A., Sumner, L. W., Blount, J. W., and Dixon, R. A. (2005). Genomics‐based selection and functional characterization of triterpene glycosyltransferases from the model legume Medicago truncatula. Plant J. 41, 875–887. Aebersold, R. H., and Mann, M. (2003). Mass spectrometry‐based proteomics. Nature 422, 198–207. Aerts, R. J., Barry, T. N., and McNabb, W. C. (1999). Polyphenols and agriculture: Beneficial eVects of proanthocyanidins in forages. Agric. Ecosyst. Environ. 75, 1–12. Agrawal, G. K., and Thelen, J. J. (2006). Large scale identification and quantitative profiling of phosphoproteins expressed during seed filling in oilseed rape. Mol. Cell. Proteomics 5, 2044–2059. Aharoni, A., de Vos, C. H. R., Verhoeven, H. A., Maliepaard, C. A., Kruppa, G., Bino, R., and Goodenowe, D. B. (2002). Non‐targeted metabolomic profiling using Fourier transform ion cyclotron mass spectrometry (FTMS). OMICS J. Integrat. Biol. 6, 217–234. Aharoni, A., Dixit, S., Jetter, R., Thoenes, E., van Arkel, G., and Pereira, A. (2004). The SHINE clade of AP2 domain transcription factors activates wax biosynthesis, alters cuticle properties, and confers drought tolerance when overexpressed in Arabidopsis. Plant Cell 16, 2463–2480. Akkaya, M. S., Shoemaker, R. C., Specht, J. E., Bhagwat, A. A., and Cregan, P. B. (1995). Integration of simple sequence repeat (SSR) DNA markers into a soybean linkage map. Crop Sci. 35, 1439–1445. Albrecht, K. A., Wedin, W. F., and Buxton, D. R. (1987). Cell‐wall composition and digestibility of alfalfa stems and leaves. Crop Sci. 27, 735–741. Alm, V., Fang, C., Busso, C. S., Devos, K. M., Vollan, K., Grieg, Z., and Rognli, O. A. (2003). A linkage map of meadow fescue (Festuca pratensis Huds.) and comparative mapping with other Poaceae species. Theor. Appl. Genet. 108, 25–40. Alonso, J. M., and Ecker, J. R. (2006). Moving forward in reverse: Genetic technologies to enable genome‐wide phenomic screens in Arabidopsis. Nat. Rev. Genet. 7, 524–536. Alonso, J. M., Stepanova, A. N., Leisse, T. J., Kim, C. J., Chen, H., Shinn, P., Stevenson, D. K., Zimmerman, J., Barajas, P., Cheuk, R., Gadrinab, C., Heller, C., et al. (2003). Genome‐ wide insertional mutagenesis of Arabidopsis thaliana. Science 301, 653–657. Anand, A., and Mysore, K. S. (2005). Current advances in Agrobacterium‐plant interactions and their implications in agricultural biotechnology. In ‘‘Advances in Plant Physiology’’ (A. Hemantaranjan, Ed.), pp. 221–242. Scientific Publishers, Jodhpur, India. Arabidopsis Genome Initiative. (2000). Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature 408, 796–815. Asamizu, E., Watanabe, M., and Tabata, S. (2000). Large scale structural analysis of cDNAs in the model legume, Lotus japonicus. J. Plant Res. 113, 451–455. Ausubel, F. M. (2002). Summaries of National Science Foundation‐sponsored Arabidopsis 2010 projects and National Science Foundation‐sponsored plant genome projects that are generating Arabidopsis resources for the community. Plant Physiol. 129, 394–437. Aziz, N., Paiva, N. L., May, G. D., and Dixon, R. A. (2005). Profiling the transcriptome of alfalfa glandular trichomes. Planta 221, 28–38.
140
R. A. DIXON ET AL.
Azpiroz‐Leehan, R., and Feldmann, K. A. (1997). T‐DNA insertion mutagenesis in Arabidopsis: Going back and forth. Trends Genet. 13, 152–156. Bachem, C. W. B., Oomen, R. J. E. F., and Visser, R. G. F. (1998). Transcript imaging with cDNA‐AFLP: A step‐by‐step protocol. Plant Mol. Biol. Rep. 16, 157–173. Baldwin, D., Crane, V., and Rice, D. (1999). A comparison of gel‐based, nylon filter and microarray techniques to detect diVerential RNA expression in plants. Curr. Opin. Plant Biol. 2, 96–103. Baligar, V. C., Elgin, J. H., and Foy, C. D. (1989). Variability in alfalfa for growth and mineral uptake and eYciency ratios under aluminum stress. Agron. J. 81, 223–229. Barnett, M. J., Toman, C. J., Fisher, R. F., and Long, S. R. (2004). A dual‐genome symbiosis chop for coordinate study of singal exchange and development in a prokaryote‐host interaction. Proc. Natl. Acad. Sci. USA 101, 16636–16641. Bartel, B., and Bartel, D. P. (2003). MicroRNAs: At the root of plant development? Plant Physiol. 132, 709–717. Baucher, M., BernardVailhe, M. A., Chabbert, B., Besle, J. M., Opsomer, C., VanMontagu, M., and Botterman, J. (1999). Down‐regulation of cinnamyl alcohol dehydrogenase in transgenic alfalfa (Medicago sativa L.) and the eVect on lignin composition and digestibility. Plant Mol. Biol. 39, 437–447. Bechtold, N., Jaudeau, B., Jolivet, S., Maba, B., Vezon, D., Voisin, R., and Pelletier, G. (2000). The maternal chromosome set is the target of the T‐DNA in the in planta transformation of Arabidopsis thaliana. Genetics 155, 1875–1887. Beckmann, J. S., and Soller, M. (1993). Restriction fragment length polymorphisms in genetic improvement: Methodologies, mapping and costs. Theor. Appl. Genet. 67, 35–43. Bell, C. J., and Ecker, J. R. (1994). Assignment of 30 microsatellite loci to the linkage map of Arabidopsis. Genomics 19, 137–144. Bent, A. F. (2000). Arabidopsis in planta transformation. Uses, mechanisms, and prospects for transformation of other species. Plant Physiol. 124, 1540–1547. Berloo, R., and Stam, P. (1998). Marker‐assisted selection in autogamous RIL populations: A simulation study. Theor. Appl. Genet. 96, 147–154. Bevan, M., Bancroft, I., Mewes, H. W., Martienssen, R., and McCombie, R. (1999). Clearing a path through the jungle: Progress in Arabidopsis genomics. Bioessays 21, 110–120. Bevan, M., Mayer, K., White, O., Eisen, J. A., Preuss, D., Bureau, T., Salzberg, S. L., and Mewes, H. W. (2001). Sequence and analysis of the Arabidopsis genome. Curr. Opin. Plant Biol. 4, 105–110. Bingham, E. T. (1968). Transfer of diploid Medicago spp. germplasm to tetraploid M. sativa L. in 4x‐2x crosses. Crop Sci. 8, 760–762. Bingham, E. T. (1980). Maximizing heterozygosity in autopolyploids. In Polyploidy, Biological Relevance (W. H. Lewis, Ed.), pp. 471–490. Plenum Press, New York. Bock, R., and Khan, M. S. (2004). Taming plastids for a green future. Trends Biotechnol. 22, 311–318. Boerjan, W., Ralph, J., and Baucher, M. (2003). Lignin biosynthesis. Annu. Rev. Plant Biol. 54, 519–546. Bonierbale, M. W., Plaisted, R. L., Pineda, O., and Tanksley, S. D. (1994). QTL analysis of trichome‐mediated insect resistance in potato. Theor. Appl. Genet. 87, 973–987. Borevitz, J., Xia, Y., Blount, J. W., Dixon, R. A., and Lamb, C. (2000). Activation tagging identifies a conserved MYB regulator of phenylpropanoid biosynthesis. Plant Cell 12, 2383–2393. Boudet, A. M., Lapierre, C., and Grima‐Pettenati, J. (1995). Tansley Review No. 80. Biochemistry and molecular biology of lignification. New Phytol. 129, 203–236. Bouton, J. H. (1996). Screening the alfalfa core collection for acid soil tolerance. Crop Sci. 36, 198–200.
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
141
Bouton, J. H. (2001). Alfalfa. In Proceedings of the XIX International Grassland Congress (J. A. Gomide, W. R. S. Mattos, and S. C. da Silva, Eds.), pp. 545–547. FEALQ, Piracicaba, Sao Pedro, Sao Paulo, Brazil. Bouton, J. H., and Parrott, W. A. (1997). Salinity and aluminum stress. In ‘‘Biotechnology and the Improvement of Forage Legumes’’ (B. D. McKersie and D. C. W. Brown, Eds.), pp. 203–226. CAB International, Wallingford, New York. Bowles, D., Lim, E., Poppenberger, B., and Vaistij, F. (2006). Glycosyltransferases of lipophilic small molecules. Annu. Rev. Plant Biol. 57, 567–597. Bradbury, L. M. T., Fitzgerald, T. L., Henry, R. J., Jin, Q. S., and Waters, D. L. E. (2005). The gene for fragrance in rice. Plant Biotechnol. J. 3, 363–370. Bradford, K. J., Van Deynze, A., Gutterson, N., Parrott, W., and Strauss, S. H. (2005). Regulating transgenic crops sensibly: Lessons from plant breeding, biotechnology and genomics. Nat. Biotechnol. 23, 439–444. Brady, N. C. (1974). ‘‘The Nature and Property of Soils.’’ Macmillian Publishing Co. Inc., New York. Brazma, A., Hingamp, P., Quackenbush, J., Sherlock, G., Spellman, P., Stoeckert, C. J. J., Aach, J., Ansorge, W., Ball, C. A., Causton, H. C., Gaasterland, T., Glenisson, P., et al. (2001). Minimum information about a microarray experiment (MIAME): Toward standards for microarray data. Nat. Genet. 29, 365–371. Bregitzer, P., Halbert, S. E., and Lemaux, P. G. (1998). Somaclonal variation in the progeny of transgenic barley. Theor. Appl. Genet. 96, 421–425. Brenner, S., Johnson, M., Bridgham, J., Golda, G., Lloyd, D. H., Johnson, D., Luo, S., McCurdy, S., Foy, M., Ewan, M., Roth, R., George, D., et al. (2000). Gene expression analysis by massively parallel signature sequencing (MPSS) on microbead arrays. Nat. Biotechnol. 18, 630–634. Broeckling, C. D., Huhman, D. V., Farag, M., Smith, J. T., May, G. D., Mendes, P., Dixon, R. A., and Sumner, L. W. (2005). Metabolic profiling of Medicago truncatula cell cultures reveals eVects of biotic and abiotic elicitors on primary metabolism. J. Exp. Bot. 56, 323–336. Broothaerts, W., Mitchell, H. J., Weir, B., Kaines, S., Smith, L. M. A., Wei, Y., Mayer, J. E., Roa‐Rodrı´guez, C., and JeVerson, R. A. (2005). Gene transfer to plants by diverse species of bacteria. Nature 433, 629–633. Broun, P., Poindexter, P., Osborne, E., Jiang, C.‐Z., and Riechmann, J. L. (2004). WIN1, a transcriptional activator of epidermal wax accumulation in Arabidopsis. Proc. Natl. Acad. Sci. USA 101, 4706–4711. Brummer, E. C., Bouton, J. H., and Kochert, G. (1993). Development of an RFLP map in diploid alfalfa. Theor. Appl. Genet. 86, 329–332. Bundock, P. C., and Henry, R. J. (2004). Single nucleotide polymorphism haplotype diversity and recombination in the Isa gene of barley. Theor. Appl. Genet. 109, 543–551. Buol, S. W., and Eswaran, H. (1993). Assessment and conquest of poor soils. In ‘‘Adaptation of Plants to Soil Stresses’’ (J. W. Maranville, Ed.), pp. 17–27. Intsormil Publication No. 94–2, University of Nebraska, Lincoln. Burr, B., and Burr, F. A. (1991). Recombinant inbreds for molecular mapping in maize. Trends Genet. 7, 55–60. Buxton, D. R., and Russell, J. R. (1988). Lignin constituents and cell‐wall digestibility of grass and legume stems. Crop Sci. 28, 553–558. Cannon, S. B., Sterck, L., Rombauts, S., Sato, S., Cheung, F., Gouzy, J. P., Wang, X., Mudge, J., Vasdewani, J., Scheix, T., Spannagl, M., Nicholson, C., et al. (2006). Legume genome evolution viewed through the Medicago truncatula and Lotus japonicus genomes. Proc. Natl. Acad. Sci. USA 103, 14959–14964. Cardinal, J., Lee, M., and Moore, J. (2003). Genetic mapping and analysis of quantitative trait loci aVecting fiber and lignin content in maize. Theor. Appl. Genet. 106, 866–874.
142
R. A. DIXON ET AL.
Casati, P., and Walbot, V. (2003). Gene expression profiling in response to ultraviolet radiation in maize genotypes with varying flavonoid content. Plant Physiol. 132, 1739–1754. Casati, P., Zhang, X. C., Burlingame, A. L., and Walbot, V. (2005). Analysis of leaf proteome after UV‐B irradiation in maize lines diVering in sensitivity. Mol. Cell. Proteomics 4, 1673–1685. Casler, M. D. (1987). In vitro digestibility of dry matter and cell wall constituents of smooth bromegrass forage. Crop Sci. 27, 931–934. Casler, M. D., and Vogel, K. P. (1999). Accomplishments and impact from breeding for increased forage nutritional value. Crop Sci. 39, 12–20. Charlton, A., Godward, J., Oehlschlager, S., Ambrose, M., Bowen, M., Welham, T., Mullineaux, P., and Domoney, C. (2004). Investigation of the metabolome of leaves from a pea germplasm array and from transgenic pea plants exposed to environmental stress. In ‘‘Proceedings of the Fifth European Conference on Grain Legumes,’’ pp. 189–190, Dijon, France. Chen, E. I., Hewel, J., Felding‐Habermann, B., and Yates, J. R., III (2006a). Large scale protein profiling by combination of protein fractionation and multidimensional protein identification technology (MudPIT). Mol. Cell. Proteomics 5, 53–56. Chen, F., Duran, A. L., Blount, J. W., Sumner, L. W., and Dixon, R. A. (2003). Profiling phenolic metabolites in transgenic alfalfa modified in lignin biosynthesis. Phytochemistry 64, 1013–1021. Chen, F., Reddy, M. S. S., Temple, S., Jackson, L., Shadle, G. L., and Dixon, R. A. (2006b). Multi‐site genetic modulation of monolignol biosynthesis suggests new routes for the formation of syringyl lignin and wall‐bound ferulic acid in alfalfa (Medicago sativa L.). Plant J. 48, 113–124. Chen, X., Salamini, F., and Gebhardt, C. (2001). A potato molecular‐function map for carbohydrate metabolism and transport. Theor. Appl. Genet. 102, 284–295. Cheng, M., Fry, J. E., Pang, S., Zhou, H., Hironaka, C. M., Duncan, D. R., Conner, T. W., Wan, Y., Cheng, M., Pang, S. Z., Zhou, H. P., and Wan, Y. C. (1997). Genetic transformation of wheat mediated by Agrobacterium tumefaciens. Plant Physiol. 115, 971–980. Cheng, M., Lowe, B. A., Spencer, T. M., Ye, X. D., and Armstrong, C. L. (2004). Factors influencing Agrobacterium‐mediated transformation of monocotyledonous species. In Vitro Cell. Dev. Biol.‐Plant 40, 31–45. Cheng, X.‐F., and Wang, Z.‐Y. (2005). Overexpression of COL9, a CONSTANS‐LIKE gene, delays flowering by reducing CO and FT expression in Arabidopsis thaliana. Plant J. 43, 758–768. Cheung, F., Haas, B. J., Goldberg, S. M. D., May, G. D., Xiao, Y., and Town, C. D. (2006). Sequencing Medicago truncatula expressed sequenced tags using 454 Life Sciences technology. BMC Genomics 7, 272–282. Cho, Y. G., Ishii, T., Temnykh, S., Chen, X., Lopovich, L., McCouch, S. R., Park, W. D., Ayres, N., and Cartinhour, S. (2000). Diversity of microsatellites derived from genomic libraries. Theor. Appl. Genet. 100, 713–722. Chory, J., Ecker, J. R., Briggs, S., Caboche, M., Coruzzi, G. M., Cook, D., Dangl, J., Grant, S., Guerinot, M. L., HenikoV, S., Martienssen, R., Okada, K., et al. (2000). National Science Foundation‐Sponsored Workshop report: ‘The 2010 Project’: Functional genomics and the virtual plant. A blueprint for understanding how plants are built and how to improve them. Plant Physiol. 123, 423–425. Christou, P. (1992). Genetic transformation of crop plants using microprojectile bombardment. Plant J. 2, 275–281. Claverie, J.‐M. (2005). Fewer genes, more non‐coding RNA. Science 309, 1529–1530. Clough, S. J., and Bent, A. F. (1998). Floral dip: A simplified method for Agrobacterium‐ mediated transformation of Arabidopsis thaliana. Plant J. 16, 735–743.
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
143
Cloutier, S., Cappadocia, M., and Landry, B. S. (1995). Study of microspore‐culture responsiveness in oilseed rape (Brassica napus L.) by comparative mapping of a F2 population and two microspore‐derived populations. Theor. Appl. Genet. 91, 841–847. Colebatch, G., Desbrosses, G., Ott, T., Krusell, L., Montanari, O., Kloska, S., Kopka, J., and Udvardi, M. K. (2004). Global changes in transcription orchestrate metabolic diVerentiation during symbiotic nitrogen fixation in Lotus japonicus. Plant J. 39, 487–512. Comai, L., Young, K., Till, B. J., Reynolds, S. H., Greene, E. A., Codomo, C. A., Enns, L. C., Johnson, J., Burtner, C., and Oden, A. R. (2004). EYcient discovery of DNA polymorphisms in natural populations by Ecotilling. Plant J. 37, 778–786. Comis, D. (2005). TILLING genes to improve soybeans. Agric. Res. 53, 4–5. Cook, D. R. (1999). Medicago truncatula: A model in the making! Curr. Opin. Plant Biol. 2, 301–304. Cordeiro, G. M., Eliott, F., McIntyre, C. L., Casu, R. E., and Henry, R. J. (2006). Characterisation of single nucleotide polymorphisms in sugarcane ESTs. Theor. Appl. Genet. 113, 331–343. Costa, M. A., Collins, R. E., Anterola, A. M., Cochrane, F. C., Davin, L. B., and Lewis, N. G. (2003). An in silico assessment of gene function and organization of the phenylpropanoid pathway metabolic networks in Arabidopsis thaliana and limitations thereof. Phytochemistry 64, 1097–1112. Coulman, B., Goplen, B., Majak, W., McAllister, T., Cheng, K. J., Berg, B., Hall, J., McCartney, D., and Acharya, S. (2000). A review of the development of a bloat‐reduced alfalfa cultivar. Can. J. Plant Sci. 80, 487–491. Czechowski, T., Bari, R. P., Stitt, M., Scheible, W. R., and Udvardi, M. K. (2004). Real‐time RT‐PCR profiling of over 1400 Arabidopsis transcription factors: Unprecedented sensitivity reveals novel root‐ and shoot‐specific genes. Plant J. 38, 366–379. da la Fuente, J. M., Ramirez‐Rodriguez, V., Cabrera‐Ponce, J. L., and Herrera‐Estrella, L. (1997). Aluminum tolerance in transgenic plants by alteration of citrate synthesis. Science 276, 1566–1568. Dall’Agnol, M., Bouton, J. H., and Parrott, W. A. (1996). Screening methods to develop alfalfa germplasms tolerant of acid, aluminum toxic soils. Crop Sci. 36, 64–70. Davin, L. B., and Lewis, N. G. (2005). Dirigent phenoxy radical coupling: Advances and challenges. Curr. Opin. Biotechnol. 16, 398–406. Davuluri, G. R., van Tuinen, A., Fraser, P. D., Manfredonia, A., Newman, R., Burgess, D., Brummell, D. A., King, S. R., Palys, J., Uhlig, J., Bramley, P. M., Pennings, H. M. J., et al. (2005). Fruit‐specific RNAi‐mediated suppression of DET1 enhances tomato nutritional quality. Nat. Biotechnol. 23, 890–895. De Paepe, A., Vuylsteke, M., Van Hummelen, P., Zabeau, M., and Van Der Straeten, D. (2004). Transcriptional profiling by cDNA‐AFLP and microarray analysis reveals novel insights into the early response to ethylene in Arabidopsis. Plant J. 39, 537–559. Deavours, B. E., and Dixon, R. A. (2005). Metabolic engineering of isoflavonoid biosynthesis in alfalfa (Medicago sativa L.). Plant Physiol. 138, 2245–2259. Debrosses, G. G., Kopka, J., and Udvardi, M. K. (2005). Lotus japonicus metabolic profiling. Development of gas chromatography mass spectrometry resources for the study of plant‐ microbe interactions. Plant Physiol. 137, 1302–1318. Delhaize, E., and Ryan, P. R. (1995). Aluminum toxicity and tolerance in plants. Plant Physiol. 107, 315–321. d’Erfurth, I., Cosson, V., Eschstruth, A., Lucas, H., Kondorosi, A., and Ratet, P. (2003). EYcient transposition of the Tnt1tobacco retrotransposon in the model legume Medicago truncatula. Plant J. 34, 95–106. Desfeux, C., Clough, S. J., and Bent, A. F. (2000). Female reproductive tissues are the primary target of Agrobacterium‐mediated transformation by the Arabidopsis floral‐dip method. Plant Physiol. 123, 895–904.
144
R. A. DIXON ET AL.
Devos, K. M., and Gale, M. D. (2000). Genome relationships: The grass model in current research. Plant Cell 12, 636–646. Dhugga, K. S., Barreiro, R., Whitten, B., Stecca, K., Hazebroek, J., Randhawa, G. S., Dolan, M., Kinney, A. J., Tomes, D., Nichols, S., and Anderson, P. (2004). Guar seed b‐ mannan synthase is a member of the cellulose synthase super gene family. Science 303, 363–366. Dinesh‐Kumar, S. P., Anandalakshmi, R., Marathe, R., SchiV, M., and Liu, Y. (2003). Virus‐ induced gene silencing. In ‘‘Plant Functional Genomics’’ (E. Grotewold, Ed.), pp. 287–293. Humana Press Inc., Totowa, NJ. Ding, X. S., Schneider, W. L., Chaluvadi, S. R., Mian, R., and Nelson, R. S. (2006). Characterization of a Brome mosaic virus strain and its use as a vector for gene silencing in monocotyledonous hosts. Mol. Plant Microbe Interact. 19, 1229–1239. Dixon, R. A., Achnine, L., Kota, P., Liu, C.‐J., Reddy, M. S., and Wang, L. (2002). The phenylpropanoid pathway and plant defense: A genomics perspective. Mol. Plant Pathol. 3, 371–390. Dixon, R. A., Sharma, S. B., and Xie, D. (2005). Proanthocyanidins: A final frontier in flavonoid research? New Phytologist 165, 9–28. Dixon, R. A., Gang, D. R., Charlton, A. J., Fiehn, O., Kuiper, H. A., Reynolds, T. L., Tjeerdema, R. S., JeVrey, E. H., German, J. B., Ridley, W. P., and Seiber, J. N. (2006). Applications of metabolomics to agriculture. J. Agric. Food Chem. 54, 8984–8994. Douglas, G. B., Stienezen, M., Waghorn, G. C., Foote, A. G., and Purchas, R. W. (1999). EVect of condensed tannins in birdsfoot trefoil (Lotus corniculatus) and sulla (Hedysarum coronarium) on body weight, carcass fat depth, and wool growth of lambs in New Zealand. NZ J. Agric. Res. 42, 55–64. Downie, A. (1997). Fixing a symbiotic circle. Nature 387, 352–354. Dudley, J. W., Saghai Maroof, A., and Rufener, G. K. (1992). Molecular marker information and selection of parents in corn breeding programs. Crop Sci. 32, 301–304. Echt, C. S., Kidwell, K. K., Knapp, S. J., Osborn, T. C., and McCoy, T. J. (1993). Linkage mapping in diploid alfalfa (Medicago sativa). Genome 37, 61–71. Eujayl, I., Baum, M., Powell, W., Erskine, W., and Pehu, E. (1998). A genetic linkage map of lentil (Lens sp.) based on RAPD and AFLP markers using recombinant inbred lines. Theor. Appl. Genet. 97, 83–89. Eujayl, I., Sorrells, M. E., Baum, M., Wolters, P., and Powell, W. (2002). Isolation of EST‐ derived microsatellite markers for genotyping the A and B genomes of wheat. Theor. Appl. Genet. 104, 399–407. Eujayl, I., Sledge, M. K., Wang, L., May, G. D., Chekhovskiy, K., Zwonitzer, J. C., and Mian, M. A. R. (2004). Medicago truncatula EST‐SSRs reveal cross‐species genetic markers for Medicago spp. Theor. Appl. Genet. 108, 414–422. Fei, Z., Tang, X., Alba, R., and Giovannoni, J. (2006). Tomato expression database (TED): A suite of data presentation and analysis tools. Nucleic Acids Res. 34, D770–D776 (database issue). Fernandez‐Cornejo, J., and Caswell, M. (2006). The first decade of genetically engineered crops in the United States (USDA Economic Information Bulletin Number 11. www.ers.usda. gov). Fiehn, O. (2002). Metabolomics: The link between genotypes and phenotypes. Plant Mol. Biol. 48, 155–171. Firnhaber, C., Pu¨hler, A., and Ku¨ster, H. (2005). EST sequencing and time course microarray hybridizations identify more than 700 Medicago truncatula genes with develomental expression regulation in flowers and pods. Planta 222, 269–283. Foy, C. D. (1988). Plant adaptation to acid, aluminum toxic soils. Comm. Soil Sci. Plant Anal. 19, 958–987.
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
145
Fraser, P. D., Pinto, M. E. S., Holloway, D. E., and Bramley, P. M. (2000). Application of high‐ performance liquid chromatography with photodiode array detection to the metabolic profiling of plant isoprenoids. Plant J. 24, 551–558. Freeling, M. (2001). Grasses as a single genetic system. Reassessment 2001. Plant Physiol. 125, 1191–1197. Fridman, E., Wang, J., Iijima, Y., Froehlich, J. E., Gang, D. R., Ohlrogge, J., and Pichersky, E. (2005). Metabolic, genomic, and biochemical analyses of glandular trichomes from the wild tomato species Lycopersicon hirsutum identify a key enzyme in the biosynthesis of methylketones. Plant Cell 17, 1252–1267. Frisch, M., Bohn, M., and Melchinger, A. E. (1999). Comparison of selection strategies for marker‐assisted backcrossing of a gene. Crop Sci. 39, 1295–1301. Fuell, H., Hardy, N., Beckmann, M., Draper, J., Smith, A. R., Taylor, T., Fiehn, O., Goodacre, R., Bino, R., Hall, R., Kopka, J., Lange, B. M., et al. (2004). A proposed framework for the description of plant metabolomics experiments and their results. Nat. Biotechnol. 22, 1601–1606. Gaita´n‐Solı´s, E., Duque, M. C., Edwards, K. J., and Tohme, J. (2002). Microsatellite repeats in common bean (Phaseolus vulgaris): Isolation, characterization, and cross‐species amplification in Phaseolus ssp. Crop Sci. 42, 2128–2136. Gale, M. D., and Devos, K. M. (1998). Comparative genetics in the grasses. Proc. Natl. Acad. Sci. USA 95, 1971–1974. Gang, D. R., Wang, J., Dudareva, N., Nam, K. H., Simon, J. E., Lewinsohn, E., and Pichersky, E. (2001). An investigation of the storage and biosynthesis of phenylpropenes in sweet basil. Plant Physiol. 125, 539–555. Ge, Y., Norton, T., and Wang, Z.‐Y. (2006). Transgenic zoysiagrass (Zoysia japonica) plants obtained by Agrobacterium‐mediated transformation. Plant Cell Rep. 25(8), 792–798. Ge, Y., Cheng, X.‐F., Hopkins, A., and Wang, Z.‐Y. (2007). Generation of transgenic Lolium temulentum plants by Agrobacterium tumefaciens‐mediated transformation. Plant Cell Rep. (in press). Gebhardt, C., Ritter, E., Barone, A., Debener, T., Walkemeier, B., Schachtschabel, E., Kaufman, H., Thompson, R. D., Bonierbale, M. W., Ganal, M. W., Tanksley, S. D., and Salamini, F. (1991). RFLP maps of potato and their alignment with the homologous tomato genome. Theor. Appl. Genet. 83, 49–57. Gelvin, S. B. (2003a). Agrobacterium‐mediated plant transformation: The biology behind the ‘‘gene‐jockeying’’ tool. Microbiol. Mol. Biol. Rev. 67, 16–37. Gelvin, S. B. (2003b). Improving plant genetic engineering by manipulating the host. Trends Biotechnol. 21, 95–98. Georgieva, I. D. (1998). Relation between glandular trichome phenolic compounds in tomato and resistance to whitefly (Trialeurodes vaporariorum West). Genet. Breed. 29, 61–66. Giancola, S., McKhann, H. I., Be´rard, A., Camilleri, C., Durand, S., Libeau, P., Roux, F., Reboud, X., Gut, I. G., and Brunel, D. (2006). Utilization of the three high‐throughput SNP genotyping methods, the GOOD assay, Amplifluor and TaqMan, in diploid and polyploidy plants. Theor. Appl. Genet. 112, 1115–1124. Gilmour, S. J., Sebolt, A. M., Salazar, M. P., Everard, J. D., and Thomashow, M. F. (2000). Overexpression of the Arabidopsis CBF3 transcriptional activator mimics multiple biochemical changes associated with cold acclimation. Plant Physiol. 124, 1854–1865. GoV, S. A., Ricke, D., Lan, T.‐H., Presting, G., Wang, R. D., Dunn, M., Glazebrook, J., Sessions, A., Oeller, P., Varma, H., Hadley, D., Hutchison, D., et al. (2002). A draft sequence of the rice genome (Oryza sativa L. ssp. japonica). Science 296, 92–100. Goldberg, S. M., Johnson, J., Busam, D., Feldblyum, T., Ferriera, S., Friedman, R., Halpern, A., Khouri, H., Kravitz, S. A., Lauro, F. M., Li, K., Rogers, Y.‐H., et al. (2006). A Sanger/
146
R. A. DIXON ET AL.
pyrosequencing et alhybrid approach for the generation of high‐quality draft assemblies of marine microbial genomes. Proc. Natl. Acad. Sci. USA 103, 11240–11245. Goldman, J. J., Hanna, W. W., Fleming, G. H., and Ozias‐Akins, P. (2004). Ploidy variation among herbicide‐resistant bermudagrass plants of cv. TifEagle transformed with the bar gene. Plant Cell Rep. 22, 553–560. Gonzales, M. D., Archuleta, E., Farmer, A., Gajendran, K., Grant, D., Shoemaker, R., Beavis, W. D., and Waugh, M. E. (2005). The Legume Information System (LIS): An integrated information resource for comparative legume biology. Nucleic Acids Res. 33, D660–D665. Goossens, A., Ha¨kkinen, S., Laakso, I., Seppa¨nen‐Laakso, T., Biondi, S., De Sutter, V., Lammertyn, F., Nuutila, A. M., So¨derlund, H., Zabeau, M., Inze´, D., and Oksman‐ Caldentey, K. M. (2003). A functional genomics approach toward the understanding of secondary metabolism in plant cells. Proc. Natl. Acad. Sci. USA 100, 8595–8600. Grabber, J. H., Jung, G. A., Abrams, S. M., and Howard, D. B. (1992). Digestion kinetics of parenchyma and sclerenchyma cell walls isolated from orchardgrass and switchgrass. Crop Sci. 32, 806–810. Grabber, J. H., Ralph, J., Hatfield, R. D., and Quideau, S. (1997). p‐Hydroxyphenyl, guaiacyl, and syringyl lignins have similar inhibitory eVects on wall degradability. J. Agric. Food Chem. 45, 2530–2532. Graham, T. L. (1991). A rapid, high resolution high performance liquid chromatography profiling procedure for plant and microbial aromatic secondary metabolites. Plant Physiol. 95, 584–593. Grevich, J. J., and Daniell, H. (2005). Chloroplast genetic engineering: Recent advances and future perspectives. Crit. Rev. Plant Sci. 24, 83–107. Guo, D., Chen, F., Inoue, K., Blount, J. W., and Dixon, R. A. (2001a). Down‐regulation of caVeic acid 3‐O‐methyltransferase and caVeoyl CoA 3‐O‐methyltransferase in transgenic alfalfa (Medicago sativa L.): Impacts on lignin structure and implications for the biosynthesis of G and S lignin. Plant Cell 13, 73–88. Guo, D., Chen, F., Wheeler, J., Winder, J., Selman, S., Peterson, M., and Dixon, R. A. (2001b). Improvement of in‐rumen digestibility of alfalfa forage by genetic manipulation of lignin O‐methyltransferases Transgenic Res. 10, 457–464. Hackauf, B., and Wehling, P. (2002). Identification of microsatellite polymorphisms in an expressed portion of the rye genome. Plant Breed. 121, 17–25. Hall, R., Beale, M., Fiehn, O., Hardy, N., Sumner, L. W., and Bino, R. (2002). Plant metabolomics as the missing link in functional genomics strategies. Plant Cell 14, 1437–1440. Hamada, H., Petrino, M. G., and Kakunaga, T. (1982). A novel repeated element with Z‐DNA forming potential is widely found in evolutionarily diverse eukaryotic genomes. Proc. Natl. Acad. Sci. USA 79, 6465–6469. Harlan, J. R., and deWet, J. M. J. (1975). The origins of polyploids. Bot. Rev. 41, 361–390. Hass, W., Faherty, B. K., Gerber, S. A., Elias, J. E., Beausoleil, S. A., Bakalarski, C. E., Li, X., Villen, J., and Gygi, S. P. (2006). Optimization and use of peptide mass measurement accuracy in shotgun proteomics. Mol. Cell. Proteomics 5, 1326–1337. He, X. G. (2000). On‐line identification of phytochemical constituents in botanical extracts by combined high‐performance liquid chromatographic‐diode array detection‐mass spectrometric techniques. J. Chromatogr. A 880, 203–232. Hellens, R., and Mullineaux, P. (2000). A guide to Agrobacterium binary Ti vectors. Trends Plant Sci. 5, 446–451. Henderson, I. R., Zhang, X., Lu, C., Johnson, L., Meyers, B. C., Green, P. J., and Jacobsen, S. E. (2006). Dissecting Arabidopsis thaliana DICER function in small RNA processing, gene silencing and DNA methylation patterning. Nat. Genet. 38, 721–725. Hendler, J. (2003). Communication. Science and the semantic web. Science 299, 520–521.
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
147
HenikoV, S., Till, B. J., and Comai, L. (2004). TILLING. Traditional mutagenesis meets functional genomics. Plant Physiol. 135, 630–636. Herrera‐Estrella, L., Simpson, J., and Martinez‐Trujillo, M. (2005). Transgenic plants: An historical perspective. In ‘‘Transgenic Plants: Methods and Protocols’’ (L. Pena, Ed.), pp. 3–32. Humana Press, Totowa, NJ. Hertzberg, M., Sievertzon, M., Aspeborg, H., Nilsson, P., Sandberg, G., and Lundeberg, J. (2001). cDNA microarray analysis of small plant tissue samples using a cDNA tag target amplification protocol. Plant J. 25, 585–591. Hiei, Y., Ohta, S., Komari, T., and Kumashiro, T. (1994). EYcient transformation of rice (Oryza sativa L.) mediated by Agrobacterium and sequence analysis of the boundaries of the T‐DNA. Plant J. 6, 271–282. Hirai, M. Y., Yano, M., Goodenowe, D. B., Kanaya, S., Kimura, T., Awazuhara, M., Arita, M., Fujiwara, T., and Saito, K. (2004). Integration of transcriptomics and metabolomics for understanding of global responses to nutritional stresses in Arabidopsis thaliana. Proc. Natl. Acad. Sci. USA 101, 10205–10210. Hirochika, H., Guiderdoni, E., An, G., Hsing, Y. I., Eun, M. Y., Han, C. D., Upadhaya, N., Ramachandran, S., Zhang, Q., Pereira, A., Sundaresan, V., and Leung, H. (2004). Rice mutant resources for gene discovery. Plant Mol. Biol. 54, 325–334. HoVmann, L., Maury, S., Martz, F., GeoVroy, P., and Legrand, M. (2003). Purification, cloning, and properties of an acyltransferase controlling shikimate and quinate ester intermediates in phenylpropanoid metabolism. J. Biol. Chem. 278, 95–103. Huang, S. (2000). The practical problems of post‐genomic biology. Nat. Biotechnol. 18, 471–472. Huhman, D. V., and Sumner, L. W. (2002). Metabolic profiling of saponin glycosides in Medicago sativa and Medicago truncatula using HPLC coupled to an electrospary ion‐ trap mass spectrometer. Phytochemistry 59, 347–360. Humphreys, J. M., and Chapple, C. (2002). Rewriting the lignin roadmap. Curr. Opin. Plant Biol. 5, 224–229. Humphreys, J. M., Hemm, M. R., and Chapple, C. (1999). New routes for lignin biosynthesis defined by biochemical characterization of recombinant ferulate 5‐hydroxylase, a multifunctional cytochrome P450‐dependent monooxygenase. Proc. Natl. Acad. Sci. USA 96, 10045–10050. Ishida, Y., Saito, H., Ohta, S., Hiei, Y., Komari, T., and Kumashiro, T. (1996). High eYciency transformation of maize (Zea mays L.) mediated by Agrobacterium tumefaciens. Nat. Biotechnol. 14, 745–750. Ishii, M., Hashimoto, S., Tsutsumi, S., Wada, Y., Matsushima, K., Kodama, T., and Aburatani, H. (2000). Direct comparison of genechip and SAGE on the quantitative accuracy in transcript profiling analysis. Genomics 68, 136–143. Jackson, S., Rokhsar, D., Stacey, G., Shoemaker, S., Schmutz, J., and Grimwood, J. (2006). Toward a reference sequence of the soybean genome: A multiagency eVort. Crop Sci. 46, 55–61. Jacobs, J. M. E., Van Eck, H. J., Arens, P., Verkerk‐Bakker, B., Linel‐Hekkert, B., Bastiaanssen, H. J. M., El‐Kharbotly, A., Pereira, A., Jacobsen, E., and Stiekema, W. J. (1995). A genetic map of potato (Solanum tuberosum) integrating molecular markers, including transposons, and classical markers. Theor. Appl. Genet. 91, 289–300. James, C. (2005). Global Status of Commercialized Biotech/GM Crops: 2005 (The International Service for the Acquisition of Agri‐biotech Applications (ISAAA). http://www.isaaa.org). Jones, E. S., Mahoney, N. L., Hayward, M. D., Armstead, I. P., Jones, J. G., Humphreys, M. O., King, I. P., Kishida, T., Yamada, T., Balfourier, F., Charmet, G., and Forster, J. W. (2002). An enhanced molecular marker‐based genetic map of perennial ryegrass (Lolium perenne) reveals comparative relationships with other Poaceae genomes. Genome 45, 282–295.
148
R. A. DIXON ET AL.
Jorgensen, R. A. (1995). Cosuppression, flower color patterns, and metastable gene expression states. Science 268, 686–691. Jung, H. G. (1997). Analysis of forage fiber and cell walls in ruminant nutrition. J. Nutr. 127, S810–S813. Jung, H. G., and Deetz, D. A. (1993). Cell wall lignification and degradability. In ‘‘Forage Cell Wall Structure and Digestibility’’ (H. G. Jung, D. R. Buxton, R. D. Hatfield, and J. Ralph, Eds.), pp. 315–346. ASA‐CSSA‐SSSA, Madison. Jung, H. G., and Vogel, K. P. (1986). Influence of lignin on digestibility of forage cell wall material. J. Anim. Sci. 62, 1703–1712. Jung, H. G., Valdez, F. R., Hartfield, R. D., and Blanchette, R. A. (1992). Cell wall composition and degradability of forage stems following chemical and biological delignification. J. Sci. Food. Agric. 58, 347–355. Jung, H. G., Smith, R. R., and Endres, C. C. (1994). Cell wall composition and degradability of stem tissue from lucerne divergently selected for lignin and in vitro dry matter disappearance. Grass Forage Sci. 49, 295–304. Jung, H. G., Mertens, D. R., and Payne, A. J. (1997). Correlation of acid detergent lignin and Klason lignin with digestibility of forage dry matter and neutral detergent fiber. J. Dairy Sci. 80, 1622–1628. Kale´n, G., Samuelsson, R., and Torma, E. (1992). A method for the determination of alkaloids in reed canarygrass (Phalaris arundinaceae L.) using HPLC and solid phase extraction. Plant Sci. 42, 224–229. Kalo, P., Endre, G., Zimanyi, L., Csandi, G., and Kiss, G. B. (2000). Construction of an improved linkage map of diploid alfalfa. Theor. Appl. Genet. 100, 641–657. Kantety, R. V., Rota, M. L., Matthews, D. E., and Sorrells, M. E. (2002). Data mining for simple sequence repeats in expressed sequence tags from barley, maize, rice, sorghum and wheat. Plant Mol. Biol. 48, 501–510. Kasuga, M., Miura, S., Shinozaki, K., and Yamaguchi‐Shinozaki, K. (2004). A combination of the Arabidopsis DREB1A gene and stress‐inducible rd29A promoter improved drought‐ and low‐temperature stress tolerance in tobacco by gene transfer. Plant Cell Physiol. 45, 346–350. Kawasaki, S., and Murakami, Y. (2000). Genome analysis of Lotus japonicus. J. Plant Res. 113, 497–506. Kehoe, D. M., Villand, P., and Somerville, S. (1999). DNA microarrays for studies of higher plants and other photosynthetic organisms. Trends Plant Sci. 4, 38–41. Kerek, N. M., Ceserani, T., Tausta, S. L., Sussex, I. M., and Nelson, T. M. (2003). Laser capture microdissection of cells from plant tissues. Plant Physiol. 132, 27–35. Kiss, G. B., Csanadi, G., Kalman, K., Kalo, P., and Okresz, L. (1993). Construction of a basic genetic map of Medicago using RFLP, RAPD, isozyme and morphological markers. Mol. Gen. Genet. 238, 129–137. Kochian, L. V. (1995). Cellular mechanism of aluminum toxicity and resistance in plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 46, 237–260. Kochian, L. V., Pineros, M. A., and Hoekenga, O. A. (2005). The physiology, genetics and molecular biology of plant aluminum resistance and toxicity. Plant Soil 274, 175–195. KoornneV, M., Alonso‐Blanco, C., and Vreugdenhil, D. (2004). Naturally occurring genetic variation in Arabidopsis thaliana. Annu. Rev. Plant Biol. 55, 141–172. Kopka, J. (2006). Gas chromatography mass spectrometry. In ‘‘Plant Metabolomics’’ (K. Saito, R. A. Dixon, and L. Willmitzer, Eds.), pp. 3–20. Springer‐Verlag, Berlin, Heidelberg. Koupai‐Abyazani, M. R., McCallum, J., Muir, A. D., Lees, G. L., Bohm, B. A., Towers, G. H. N., and Gruber, M. Y. (1993). Purification and characterization of a proanthocyanidin polymer from seed of alfalfa (Medicago sativa cv. beaver). J. Agric. Food Chem. 41, 565–569.
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
149
Kresovich, S., Szewe‐McFadden, A. K., Bliek, S. M., and McFerson, J. R. (1995). Abundance and characterization of simple sequence repeats (SSRs) isolated from a size‐fractionated genomic library of Brassica napus L. Theor. Appl. Genet. 91, 206–211. Krieger, C. J., Zhang, P., Mueller, L. A., Wang, A., Paley, S., Arnaud, M., Pick, J., Rhee, S. Y., and Karp, P. D. (2004). MetaCyc: A multiorganism database of metabolic pathways and enzymes. Nucleic Acids Res. 32, D438–D442. Krysan, P. J., Young, J. C., and Sussman, M. R. (1999). T‐DNA as an insertional mutagen in Arabidopsis. Plant Cell 11, 2283–2290. Kuenne, C., Lange, M., Funke, T., Miehe, H., Thiel, T., Grosse, I., and Scholz, U. (2005). CR‐EST: A resource for crop ESTs. Nucleic Acids Res. 33, D619–D621. Kuhl, J. C., Hanneman, R. E., and Havey, M. J. (2001). Characterization and mapping of Rpi1, a late‐blight resistance locus from diploid (1EBN) Mexican Solanum pinnatisectum. Mol. Genet. Genomics 265, 977–985. Kuo, W. P., Liu, F., Trimarchi, J., Punzo, C., Lombardi, M., Sarang, J., Whipple, M. E., Maysuria, M., Serikawa, K., Lee, S. Y., McCrann, D., Kang, J., et al. (2006). A sequence‐oriented comparison of gene expression measurements across diVerent hybridization‐based technologies. Nat. Biotechnol. 24, 832–840. Lai, Z., Livingstone, K., Zou, Y., Church, S. A., Knapp, S. J., Andrews, J., and Rieseberg, L. H. (2005). Identification and mapping of SNPs from ESTs in sunflower. Theor. Appl. Genet. 111, 1532–1544. Lange, B. M., and Ghassemian, M. (2005). Comprehensive post‐genomic data analysis approaches integrating biochemical pathway maps. Phytochemistry 66, 413–451. Lange, B. M., Wildung, M. R., Stauber, E. J., Sanchez, C., Pouchnik, D., and Croteau, R. (2000). Probing essential oil biosynthesis and secretion by functional evaluation of expressed sequence tags from mint glandular trichomes. Proc. Natl. Acad. Sci. USA 97, 2934–2939. Larkin, P. J., Gibson, J. M., Mathesius, U., Weinman, J. J., Gartner, E., Hall, E., Tanner, G. J., Rolfe, B. G., and Djordjevic, M. A. (1996). Transgenic white clover: Studies with the auxin‐ responsive promoter, GH3, in root gravitropism and lateral root development. Transgenic Res. 5, 325–335. Larson, T. R., and Graham, I. A. (2001). A novel technique for the sensitive quantification of acyl CoA esters from plant tissues. Plant J. 25, 115–125. Lawrence, R. J., and Pikaard, C. S. (2003). Transgene‐induced RNA interference: A strategy for overcoming gene redundancy in polyploids to generate loss‐of‐function mutations. Plant J. 36, 114–121. Lee, J. M., Nahm, S. H., Kim, Y. M., and Kim, B. D. (2004). Characterization and molecular genetic mapping of microsatellite loci in pepper. Theor. Appl. Genet. 108, 619–627. Lei, Z., Elmer, A., Watson, B. S., Dixon, R. A., Mendes, P., and Sumner, L. W. (2005). A two dimensional electrophoresis proteomic reference map and systematic identification of 1367 proteins from a cell suspension culture of the model legume Medicago truncatula. Mol. Cell. Proteomics 4, 1812–1825. Lepiniec, L., Debeaujon, I., Routaboul, J.‐M., Baudry, A., Pourcel, L., Nesi, N., and Caboche, M. (2006). Genetics and biochemistry of seed flavonoids. Annu. Rev. Plant Biol. 57, 405–430. Li, Y. G., Tanner, G., and Larkin, P. (1996). The DMACA‐HCl protocol and the threshold proanthocyanidin content for bloat safety in forage legumes. J. Sci. Food Agric. 70, 89–101. Liang, X., Dron, M., Schmid, J., Dixon, R. A., and Lamb, C. J. (1989). Developmental and environmental regulation of a phenylalanine ammonia‐lyase b‐glucuronidase gene fusion in transgenic tobacco plants. Proc. Natl. Acad. Sci. USA 86, 9284–9288. Lim, J., Jung, J. W., Lim, C. E., Lee, M. H., Kim, B. J., Kim, M., Bruce, W. B., and Benfey, P. N. (2005). Conservation and diversification of SCARECROW in maize. Plant Mol. Biol. 59, 619–630.
150
R. A. DIXON ET AL.
Liu, C.‐J., and Dixon, R. A. (2001). Elicitor‐induced association of isoflavone O‐methyltransferase with endomembranes prevents formation and 7‐O‐methylation of daidzein during isoflavonoid phytoalexin biosynthesis. Plant Cell 13, 2643–2658. Liu, P., Zhu, J., and Lu, Y. (2004). Marker‐assisted selection in segregating generations of self‐fertilizing crops. Theor. Appl. Genet. 109, 370–376. Liu, Z. W., Biyashev, R. M., and Saghai‐Maroof, M. A. (1996). Development of simple sequence repeat DNA markers and their integration into a barley linkage map. Theor. Appl. Genet. 93, 869–876. Lorenz, W. W., and Dean, J. F. D. (2002). SAGE profiling and demonstration of diVerential gene expression along the axial developmental gradient of lignifying xylem in loblolly pine (Pinus taeda). Tree Physiol. 22, 301–310. Lovinger, A., Liewehr, D., and Lamp, W. O. (2000). Glandular trichomes on alfalfa impede searching behavior of the potato leafhopper parasitoid. Biol. Control 18, 187–192. Lu, C., Kulkarni, K., Souret, F. F., MuthuValliappan, R., Tej, S. S., Poethig, R. S., Henderson, I. R., Jacobsen, S. E., Wang, W., Green, P. J., and Meyers, B. C. (2006). MicroRNAs and other small RNAs enriched in the Arabidopsis RNA‐dependent RNA polymerase‐2 mutant. Genome Res. 16, 1276–1288. Lu, R., Martin, H. A. M., Peart, J. R., Malcuit, I., and Baulcombe, D. C. (2003). Virus‐induced gene silencing in plants. Methods 30, 296–303. Ma, J. F. (2000). Role of organic acids in detoxification of Al in higher plants. Plant Cell Physiol. 44, 383–390. Ma, J. F., Ryan, P. R., and Delhaize, E. (2001). Aluminum tolerance in plants and the complexing role of organic acids. Trends Plant Sci. 6, 273–278. Majeran, W., Cai, Y., Sun, Q., and van Wijk, K. J. (2005). Functional diVerentiation of bundle sheath and mesophyll maize chloroplasts determined by comparative proteomics. Plant Cell 17, 3111–3140. Maliga, P. (2003). Progress towards commercialization of plastid transformation technology. Trends Biotechnol. 21, 20–28. Maliga, P. (2004). Plastid transformation in higher plants. Annu. Rev. Plant Biol. 55, 289–313. Maluf, W. R., Campos, G. A., and das Gracas, C. M. (2001). Relationships between trichome types and spider mite (Tetranychus evansi) repellence in tomatoes with respect to foliar zingiberene contents. Euphytica 121, 73–80. Mao, C., Wang, S., Jia, Q., and Wu, P. (2006). OsEIL1, a rice homolog of the Arabidopsis EIN3 regulates the ethylene response as a positive component. Plant Mol. Biol. 61, 141–152. Margulies, M., Egholm, M., Altman, W. E., Attiya, S., Bader, J. S., Bemben, L. A., Berka, J., Braverman, M. S., Chen, Y. J., Chen, Z., Dewell, S. B., Du, L., et al. (2005). Genome sequencing in microfabricated high‐density picolitre reactors. Photosystem I reaction center polypeptides of spinach are synthesized on thylakoid‐bound ribosomes. Nature 437, 376–380. Marshall, D. R., Broue´, P., Grace, J., and Munday, J. (1980). Tannins in pasture legumes: The annual and perennial Medicago species. Aust. J. Exp. Agric. Anim. Husb. 21, 47–50. Matsumura, H., Nirasawa, S., and Terauchi, R. (1999). Transcript profiling in rice (Oryza sativa L.) seedlings using serial analysis of gene expression (SAGE). Plant J. 20, 719–726. Mattick, J. S. (2005). The functional genomics of noncoding RNA. Science 309, 1527–1528. Maughan, P. J., Saghai, M. A. M., and Buss, G. R. (1995). Microsatellite and amplified sequence length polymorphisms in cultivated and wild soybean. Genome 38, 715–723. May, G. D., and Dixon, R. A. (2004). Quick Guide: Medicago truncatula. Curr. Biol. 14, R180–R181. McBride, K. E., Svab, Z., Schaaf, D. J., Hogan, P. S., Stalker, D. M., and Maliga, P. (1995). Amplification of a chimeric Bacillus gene in chloroplasts leads to an extraordinary level of an insecticidal protein in tobacco. Bio. Technol. 13, 362–365.
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
151
McCallum, C. M., Comai, L., Greene, E. A., and HenikoV, S. (2000). Targeting Induced Local Lesions IN Genomes (TILLING) for plant functional genomics. Plant Physiol. 123, 339–442. McCann, M. C., Rogan, G. J., Fitzpatrick, S., Trujillo, W. A., Sorbet, R., Hartnell, G. F., Riodan, S. G., and Nemeth, M. A. (2006). Glyphosate‐tolerant alfalfa is compositionally equivalent to conventional alfalfa. J. Agric. Food. Chem. 54, 7187–7192. McGrath, K., Dombrecht, B., Manners, J. M., Schenk, P. M., Edgar, C. I., Maclean, D. J., Scheible, W. R., Udvardi, M. K., and Kazan, K. (2005). Repressor‐ and activator‐type ethylene response factors function in jasmonate signaling and disease resistance identified via a genome‐wide screen of Arabidopsis transcription factor gene expression. Plant Physiol. 139, 949–959. Me´chin, V., Argillier, O., Menanteau, V., Barrie´re, Y., Mila, I., Pollet, B., and Lapierre, C. (2000). Relationship of cell wall composition to in vitro cell wall digestibility of maize inbred line stems. J. Sci. Food Agric. 80, 574–580. Mehrotra, B., and Mendes, P. (2006). Bioinformatics approaches to integrate metabolomics and other systems biology data. In ‘‘Plant Metabolomics’’ (K. Saito, R. A. Dixon, and L. Willmitzer, Eds.), pp. 105–115. Springer‐Verlag, Berlin, Heidelberg. Mian, M. A. R., Zwonitzer, J. C., Chen, Y., Saha, M. C., and Hopkins, A. A. (2005a). AFLP diversity within and among hardinggrass populations. Crop Sci. 45, 2591–2597. Mian, M. A. R., Zwonitzer, J. C., Hopkins, A. A., Ding, X. S., and Nelson, R. S. (2005b). Response of tall fescue genotypes to a new strain of Brome mosaic virus. Plant Dis. 89, 224–227. Michaels, S. D., and Amasino, R. M. (1998). A robust method for detecting single‐nucleotide changes as polymorphic markers by PCR. Plant J. 14, 381–385. Mikkilineni, V. R., Mitra, D., Merritt, J., DiTonno, J. R., Church, G. M., Ogunnaike, B., and Edwards, J. S. (2004). Digital quantitative measurements of gene expression. Biotechnol. Bioeng. 86, 117–124. Mitra, R. M., Gleason, C. A., Edwards, A., Hadfield, J., Downie, J. A., Oldroyd, G. E. D., and Long, S. R. (2004). A Ca2þ/calmodulin‐dependent protein kinase required for symbiotic nodule development: Gene identification by transcript‐based cloning. Proc. Natl. Acad. Sci. USA 101, 4701–4705. Modolo, L., Blount, J. W., Achnine, L., Naoumkina, M., Wang, X., and Dixon, R. A. (2007). A functional genomics approach to (iso)flavonoid glycosylation in the model legume Medicago truncatula. Plant Mol. Biol. (in press). Moore, S., Payton, P., Wright, M., Tanksley, S., and Giovannoni, J. (2005). Utilization of tomato microarrays for comparative gene expression analysis in the Solanaceae. J. Exp. Bot. 56, 2885–2895. Morreel, K., Ralph, J., Lu, F., Goeminne, G., Busson, R., Herdewijn, P., Goeman, J. L., Van der Eycken, J., Boerjan, W., and Messens, E. (2004). Phenolic profiling of caVeic acid O‐methyltransferase‐deficient poplar reveals novel benzodioxane oligolignols. Plant Physiol. 136, 4023–4036. Morrell, K., Goeminne, G., Storme, V., Sterck, L., Ralph, J., Coppieters, W., Breyne, P., Steenackers, M., Georges, M., Messens, E., and Boerjan, W. (2006). Genetical metabolomics of flavonoid biosynthesis in Populus: A case study. Plant J. 47, 224–237. Mototani, H., Mabuchi, A., Saito, S., Fujioka, M., Iida, A., Takatori, Y., Kotani, A., Kubo, T., Nakamura, K., Sekine, A., Murakami, Y., Tsunoda, T., et al. (2005). A functional single nucleotide polymorphism in the core promoter region of CALM1 is associated with hip osteoarthritis in Japanese. Hum. Mol. Genet. 14, 1009–1017. Mudge, J., Cannon, S. B., Kalo, P., Oldroyd, G. E. D., Roe, B. A., Town, C. D., and Young, N. D. (2005). Highly syntenic regions in the genomes of soybean, Medicago truncatula, and Arabidopsis thaliana. BMC Plant Biol. 5, 15–30.
152
R. A. DIXON ET AL.
Mueller, E., Gilbert, J., Davenport, G., Brigneti, G., and Baulcombe, D. C. (1995). Homology‐ dependent resistance: Transgenic virus resistance in plants related to homology‐dependent gene silencing. Plant J. 7, 1001–1013. Nair, R. B., Bastress, K. L., Ruegger, M. O., Denault, J. W., and Chapple, C. (2004). The Arabidopsis thaliana reduced epidermal fluorescence1 gene encodes an aldehyde dehydrogenase involved in ferulic acid and sinapic acid biosynthesis. Plant Cell 16, 544–554. Naoumkina, M., Torres‐Jerez, I., Allen, S., He, J., Zhao, P., Dixon, R. A., and May, G. D. (2007). Analysis of cDNA libraries from developing seeds of guar (Cyamopsis tetragonoloba (L.) Taub). BMC Plant Biol. (in review). Narasimhamoorthy, B., Bouton, J. H., Olsen, K. M., and Sledge, M. K. (2007). Quantitative trait loci and candidate gene mapping of aluminum tolerance in diploid alfalfa. Theor. Appl. Genet. 114, 901–913. Nelson, C. J., Hegeman, A. D., Harms, A. C., and Sussman, M. R. (2006). A quantitative analysis of Arabidopsis plasma membrane using trypsin‐catalyzed 18O‐labeling. Mol. Cell. Proteomics 5, 1382–1395. Nesi, N., Jond, C., Debeaujon, I., Caboche, M., and Lepiniec, L. (2001). The Arabidopsis TT2 gene encodes an R2R3 MYB domain protein that acts as a key determinant for proanthocyanidin accumulation in developing seed. Plant Cell 13, 2099–2114. Nielsen, K. M. (2003). Transgenic organisms—time for conceptual diversification? Nat. Biotechnol. 21, 227–228. Njiti, V. N., Meksem, K., Iqbal, M. J., Johnson, J. E., Kassem, M. A., Zobrist, K. F., Kilo, V. Y., and Lightfoot, D. A. (2002). Common loci underlie field resistance to soybean sudden death syndrome in Forrest, Pyramid, Essex, and Douglas. Theor. Appl. Genet. 104, 294–300. Oldroyd, G. E., and Geurts, R. (2001). Medicago truncatula, going where no plant has gone before. Trends Plant Sci. 6, 552–554. Olhoft, P. M., Flagel, L. E., Donovan, C. M., and Somers, D. A. (2003). EYcient soybean transformation using hygromycin B selection in the cotyledonary‐node method. Planta 216, 723–735. Ono, E., Fukuchi‐Mizutani, M., Nakamura, N., Fukui, Y., Yonekura‐Sakakibara, K., Yamaguchi, M., Nakayama, T., Tanaka, T., Kusumi, T., and Tanaka, Y. (2006). Yellow flowers generated by expression of the aurone biosynthetic pathway. Proc. Natl. Acad. Sci. USA 103, 11075–11080. Paine, J. A., Shipton, C. A., Chaggar, S., Howells, R. M., Kennedy, M. J., Vernon, G., Wright, S. Y., HinchliVe, E., Adams, J. L., Silverstone, A. L., and Drake, R. (2005). Improving the nutritional value of Golden Rice through increased pro‐vitamin A content. Nat. Biotechnol. 23, 482–487. Panaud, O., Chen, X., and McCouch, S. R. (1996). Development of microsatellite markers and characterization of simple‐sequence length polymorphism (SSLP) in rice (Oryza sativa). Mol. Gen. Genet. 252, 597–607. Parvathi, K., Chen, F., Guo, D., Blount, J. W., and Dixon, R. A. (2001). Substrate preferences of O‐methyltransferases in alfalfa suggest new pathways for 3‐O‐methylation of monolignols. Plant J. 25, 193–202. Paterson, A. H., Damon, S., Hewitt, J. D., Zamir, D., Rabinowitch, H. D., Lincoln, S. E., Lander, E. S., and Tanksley, S. D. (1991). Mendelian factors underlying quantitative traits in tomato: Comparison across species, generations, and environments. Genetics 127, 181–197. Paterson, A. H., Lin, Y. R., Li, S., Schertz, K. F., Doebley, J. F., Pinson, S. R. M., Liu, S. C., Stansel, J. W., and Irvine, J. E. (1995). Convergent domestication of cereal crops by independent mutations at corresponding genetic loci. Science 269, 1714–1717. Penmetsa, R. V., and Cook, D. R. (2000). Production and characterization of diverse developmental mutants of Medicago truncatula. Plant Physiol. 123, 1387–1397.
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
153
Perez, F., Menendez, A., Dehal, P., and Quiros, C. F. (1999). Genomic structural diVerentiation in Solanum: Comparative mapping of the A‐ and E‐genomes. Theor. Appl. Genet. 98, 1183–1193. Perry, J. A., Wang, T. L., Welham, T. J., Gardner, S., Pike, J. M., Yoshida, S., and Parniske, M. (2003). A TILLING reverse gentics tool and a web‐accessible collection of mutants of the legume Lotus japonicus. Plant Physiol. 131, 866–871. Pfeil, B., Schlueter, J., Shoemaker, R. C., and Doyle, J. (2005). Placing palepolyploidy in relation to taxon divergence: A phylogenetic analysis in legumes using 39 gene families. Syst. Biol. 54, 441–454. Phelps, T. J., Palumbo, A. V., and Beliaev, A. S. (2002). Metabolomics and microarrays for improved understanding of phenotypic characteristics controlled by both genomics and environmental constraints. Curr. Opin. Biotechnol. 13, 20–24. Powell, W., Morgante, M., Andre, C., Hanafey, M., Vogel, J., Tingey, S., and Rafalski, A. (1996). The utility of RFLP, RAPD, AFLP, and SSR (microsatellite) markers for germplasm analysis. Mol. Breed. 2, 225–238. Provan, J., Powell, W., and Waugh, R. (1996). Microsatellite analysis of relationships within cultivated potato (Solanum tuberosum). Theor. Appl. Genet. 92, 1078–1084. Qiu, D., Morgan, C., Shi, J., Long, Y., Liu, J., Li, R., Zhuang, X., Wang, Y., Tan, X., Dietrich, E., Weihmann, T., Everett, C., et al. (2006). A comparative linkage map of oilseed rape and its use for QTL analysis of seed oil and erucic acid content. Theor. Appl. Genet. 114, 67–80. Quackenbush, J., Liang, F., Holt, I., Pertea, G., and Upton, J. (2000). The TIGR gene indices: Reconstruction and representation of expressed gene sequences. Nucleic Acids Res. 28, 141–145. Quan, D., Martin, S., and Grossman, D. (2003). Applying Semantic Web Techniques to Bioinformatics. http://theory.csail.mit.edu/dquan/iswc2003‐bioinformatics.pdf. Quiros, C. F., and Bauchan, G. R. (1988). The genus Medicago and the origin of the Medicago sativa complex. In ‘‘Alfalfa and Alfalfa Improvement’’ (A. A. Hanson, D. K. Barnes, and R. R. Hill, Eds.), pp. 93–124. Agronomy Monographs, American Society of Agronomy, Madison, WI. Ralph, J., Brunow, G., Harris, P., Dixon, R. A., and Boerjan, W. (2007). Lignification: Are lignins biosynthesized via simple combinatorial chemistry or via proteinaceous control and template replication? In ‘‘Recent Advances in Polyphenols Research’’ (F. Daayf and V. Lattanzio, Eds.) Blackwell Publishing, Malden, MA (in press). Ray, H., Yu, M., Auser, P., Blahut‐Beatty, L., McKersie, B., Bowley, S., Westcott, N., Coulman, B., Lloyd, A., and Gruber, M. Y. (2003). Expression of anthocyanins and proanthocyanidins after transformation of alfalfa with maize Lc. Plant Physiol. 132, 1448–1463. Reddy, M. S. S., Chen, F., Shadle, G. L., Jackson, L., Aljoe, H., and Dixon, R. A. (2005). Targeted down‐regulation of cytochrome P450 enzymes for forage quality improvement in alfalfa (Medicago sativa L.). Proc. Natl. Acad. Sci. USA 102, 16573–16578. Reed, J. D. (1995). Nutritional toxicology of tannins and related polyphenols in forage lugumes. J. Anim. Sci. 73, 1516–1528. Reeves, J. B. (1987). Lignin and fiber compositional changes in forages over a growing season and their eVects on in vitro digestibility. J. Dairy Sci. 70, 1583–1594. Rensink, W. A., and Buell, C. R. (2005). Microarray expression profiling resources for plant genomics. Trends Plant Sci. 10, 603–609. Ridker, P. M., Hennekens, D. H., Lindpainter, K., Stampfer, M. J., Eisenbeerg, P. R., and Miletich, J. P. (1995). Mutation in the gene coding for coagulation factor V and the risk of myocardial infarction, stroke, and venous thrombosis in apparently healthy men. N. Engl. J. Med. 332, 912–917.
154
R. A. DIXON ET AL.
Ridley, W. P., Sidhu, R. S., Pyla, P. D., Nemeth, M. A., Breeze, M. L., and Astwood, J. D. (2002). A comparison of the nutritional profile of Roundup Ready corn event NK603 to that of conventional corn (Zea mays L.). J. Agric. Food Chem. 50, 7235–7243. Roa, A. C., Chavarriaga‐Aguirre, P., Duque, M. C., Maya, M. M., Bonierbale, M. W., Iglesias, C., and Tohme, J. (2000). Cross‐species amplification of cassava (Manihot esculenta) (Euphorbiaceae) microsatellites: Allelic polymorphism and degree of relationship. Am. J. Bot. 87, 1647–1655. Roberts, C. A., Benedict, H. R., Hill, N. S., Kallenbach, R. L., and Rottinghaus, G. E. (2005). Determination of ergot alkaloid content in tall fescue by near‐infrared spectroscopy. Crop Sci. 45, 778–783. Robinson, P. N., Wollstein, A., Bo¨hme, U., and Beattie, B. (2004). Ontologizing gene‐ expression microarray data: Characterizing clusters with gene ontology. Bioinformatics 20, 979–981. Ro¨der, M., Plaschke, J., Konig, S., Borner, A., Sorrells, M., Tanksley, S., and Ganal, M. (1995). Abundance, variability and chromosomal location of microsatellites in wheat. Mol. Gen. Genet. 246, 327–333. Roessner, U., Wagner, C., Kopka, J., Trethewey, R. N., and Willmitzer, L. (2000). Simultaneous analysis of metabolites in potato tuber by gas chromatography‐mass spectrometry. Plant J. 23, 131–142. Roessner, U., Leudemann, A., Brust, D., Fiehn, O., Linke, T., Willmitzer, L., and Fernie, A. R. (2001). Metabolic profiling allows comprehensive phenotyping of genetically or environmentally modified plant systems. Plant Cell 13, 11–29. Rommens, C. M., Humara, J. M., Ye, J., Yan, H., Richael, C., Zhang, L., Perry, R., and Swords, K. (2004). Crop improvement through modification of the plant’s own genome. Plant Physiol. 135, 421–431. Rosellini, D., Barone, P., Bouton, J. H., LaFayette, P., Sledge, M. K., Veronesi, F., and Parrott, W. A. (2003). Alfalfa for acid soils: A biotech approach. Czech. J. Plant Breed. 30, 168–172. Rudd, S. (2003). Expressed sequence tags: Alternative or complement to whole genome sequences? Trends Plant Sci. 8, 321–329. Ruf, S., Hermann, M., Berger, I. J., Carrer, H., and Bock, R. (2001). Stable genetic transformation of tomato plastids and expression of a foreign protein in fruit. Nat. Biotechnol. 19, 870–875. Ryan, P. R., DiTomaso, J. M., and Kochian, L. V. (1993). Aluminum toxicity in roots: An investigation of spatial sensitivity and the role of the root cap. J. Exp. Bot. 44, 437–446. Ryu, C. M., Anand, A., and Mysore, K. S. (2004). Agrodrench: A novel and eVective agroinocualtion method for virus‐induced gene silencing in roots and diverse Solanaceous species. Plant J. 40, 322–331. Saghai‐Maroof, M. A., Biyashev, R. M., Yang, G. P., Zhang, Q., and Allard, R. W. (1994). Extraordinarily polymorphic microsatellite DNA in barley: Species diversity, chromosomal locations, and population dynamics. Proc. Natl. Acad. Sci. USA 91, 5466–5470. Saha, M. C., Mian, M. A. R., Eujayl, I., Zwonitzer, J. C., Wang, L., and May, G. D. (2004). Tall fescue EST‐SSR markers with transferability across several grass species. Theor. Appl. Genet. 109, 783–791. Saha, M. C., Mian, M. A. R., Zwonitzer, J. C., Chekhovskiy, K., and Hopkins, A. A. (2005). An SSR‐ and AFLP‐based genetic linkage map of tall fescue (Festuca arundinacea Schreb.). Theor. Appl. Genet. 110, 323–336. Samac, D. A., and Temple, S. J. (2006). Development and utilization of transformation in Medicago species. In ‘‘Genetically Modified Crops’’ (G. H. Liang and D. Z. Skinner, Eds.), pp. 165–202. The Haworth Press, Inc., Binghamton, NY. Sanford, J. C. (1988). The biolistic process. Trends Biotechnol. 6, 299–302.
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
155
Sankula, S. (2006). Quantification of the Impacts on US Agriculture of Biotechnology‐ Derived Crops Planted in 2005 (National Center for Food and Agricultural Policy. www. ncfap.org). Sasaki‐Sekimoto, Y., Taki, N., Obayashi, T., Aono, M., Matsumoto, F., Sakurai, N., Suzuki, H., Hirai, M. Y., Noji, M., Saito, K., Masuda, T., Takamiya, K., et al. (2005). Coordinated activation of metabolic pathways for antioxidants and defence compounds by jasmonates and their roles in stress tolerance in Arabidopsis. Plant J. 44, 653–668. Sauer, S., Lechner, D., Berlin, K., Plancon, C., Heuermann, A., Lehrach, H., and Gut, I. G. (2000). Full flexibility genotyping of single nucleotide polymorphisms by the GOOD assay. Nucleic Acids Res. 28, E100. Schiltz, G., Gessler, D., and Stein, L. (2004). Semantic MOBY.Position Paper, W3C Workshop on Semantic Web for Life Sciences. http://lists.w3.org/Archives/Public/public‐swls‐ws/ 2004Sep/att‐0036/smoby‐w3c‐sw‐ls.pdf. Schneider, K., and Douches, D. S. (1997). Assessment of PCR‐based simple sequence repeats to fingerprint North American potato cultivars. Am. Potato J. 74, 149–160. Schoch, G., Goepfert, S., Morant, M., Hehn, A., Meyer, D., Ullmann, P., and Werck‐ Reichart, D. (2001). CYP98A3 from Arabidopsis thaliana is a 30 ‐hydroxylase of phenolic esters, a missing link in the phenylpropanoid pathway. J. Biol. Chem. 276, 36566–36574. Scott, K. D., Eggler, P., Seaton, G., Rossetto, M., Ablett, E. M., Lee, L. S., and Henry, R. J. (2000). Analysis of SSRs derived from grape ESTs. Theor. Appl. Genet. 100, 723–726. Sefc, K. M., Lopes, M. S., Lefort, F., Botta, R., Roubelakis‐Angelakis, K. A., Iba´n˜ez, P., Wagner, H. W., Glo¨ssl, J., and Steinkellner, H. (2000). Microsatellite variability in grapevine cultivars from diVerent European regions and evaluation of assignment testing to assess the geographic origin of cultivars. Theor. Appl. Genet. 100, 498–505. Senior, M. L., Chin, E. C. L., Lee, M., Smith, J. S. C., and Stuber, C. W. (1996). Simple sequence repeat markers developed from maize sequences found in the GenBank database: Map construction. Crop Sci. 36, 1676–1683. Sewalt, V. J. H., Glasser, W. G., Fontenot, J. P., and Allen, V. G. (1996). Lignin impact on fiber degradation. 1. Quinone methide intermediates formed from lignin during in vitro fermentation of corn stover. J. Sci. Food Agric. 71, 195–203. Shadle, G., Chen, F., Reddy, M. S. S., Jackson, L., Nakashima, J., and Dixon, R. A. (2007). Down‐regulation of hydroxycinnamoyl CoA: Shikimate hydroxy cinnamoyl transferase in transgenic alfalfa impacts lignification, development and forage quality. Phytochemistry (in press). Shaked, H., Melamed‐Bessudo, C., and Levy, A. A. (2005). High frequency gene targeting in Arabidopsis plants expressing the yeast RAD54 gene. Proc. Natl. Acad. Sci. USA 102, 12265–12269. Shanmugam, K. T., O’Gara, F., Andersen, K., and Valentine, R. C. (1978). Biological nitrogen fixation. Annu. Rev. Plant Physiol. 29, 263–276. Sharma, S. B., and Dixon, R. A. (2005). Metabolic engineering of proanthocyanidins by ectopic expression of transcription factors in Arabidopsis thaliana. Plant J. 44, 62–75. Shen, Y., Zhang, R., Moore, R., Kim, J., Metz, T., Hixson, K., Zhao, R., Livesay, E., Udseth, H., and Smith, R. (2005). Automated 20 kpsi RPLC‐MS and MS/MS with chromatographic peak capacities of 1000–1500 and capabilities in proteomics and metabolomics. Anal. Chem. 77, 3090–3100. Shirasawa, K., Shiokai, S., Yamaguchi, M., Kishitani, S., and Nishio, T. (2006). Dot‐blot‐SNP analysis for practical plant breeding and cultivar identification in rice. Theor. Appl. Genet. 113, 147–155. Sidorov, V. A., Kasten, D., ShengZhi, P., Hajdukiewicz, P. T. J., Staub, J. M., and Nehra, N. S. (1999). Stable chloroplast transformation in potato: Use of green fluorescent protein as a plastid marker. Plant J. 19, 209–216.
156
R. A. DIXON ET AL.
Singh, S., Sidhu, J. S., Huang, N., Vikal, Y., Li, Z., Brar, D. S., Dhaliwal, H. S., and Khush, G. S. (2001). Pyramiding three bacterial blight resistance genes (Xa5, Xa13 and Xa21) using marker‐ assisted selection into indica rice cultivar PR106. Theor. Appl. Genet. 102, 1011–1015. Sivaguru, M., and Horst, W. J. (1998). The distal part of the transition zone is the most aluminum‐sensitive apical root zone of maize. Plant Physiol. 116, 155–163. Slade, A. J., and Knauf, V. C. (2005). TILLING moves beyond functional genomics into crop improvement. Transgenic Res. 14, 109–115. Sledge, M. K., Bouton, J. H., Dall’Agnoll, M., Parrott, W. A., and Kochert, G. (2002). Identification and confirmation of aluminum tolerance QTL in diploid Medicago sativa subsp. coerulea. Crop Sci. 42, 1121–1128. Smith, O. S., and Smith, J. S. C. (1992). Measurement of genetic diversity among maize hybrids: A comparison of isozymic, RFLP, pedigree, and heterosis data. Maydica 37, 53–60. Smulikowska, S., Pastuszewska, B., Swiech, E., Ochtabinska, A., Mieczkowska, A., Nguyen, V. C., and Buraczewska, L. (2001). Tannin content aVects negatively nutritive value of pea for monogastrics. J. Anim. Feed Sci. 10, 511–523. Somers, D. A., Samac, D. A., and Olhoft, P. M. (2003). Recent advances in legume transformation. Plant Physiol. 131, 892–899. Spangenberg, G., Wang, Z.‐Y., and Potrykus, I. (1998). ‘‘Biotechnology in Forage and Turf Grass Improvement.’’ Springer, Berlin. Squirrell, J., Hollingsworth, P. M., Woodhead, M., Russell, J., Lowe, A. J., Gibby, M., and Powell, W. (2003). How much eVort is required to isolate nuclear microsatellites from plants? Mol. Ecol. 12, 1339–1348. Stacey, G., Libault, M., Brechenmacher, L., Wan, J., and May, G. D. (2006). Genetics and functional genomics of legume nodulation. Curr. Opin. Plant Biol. 9, 110–121. Stanford, E. H. (1951). Tetrasomic inheritance in alfalfa. Agron. J. 43, 2222–2225. Stanford, E. H., Clement, W. M., and Bingham, E. T. (1972). Cytology and evolution of the Medicago sativa‐coerulea‐falcata complex. In ‘‘Alfalfa Science and Technology’’ (C. H. Hanson, Ed.), pp. 87–102. American Society of Agronomy, Madison, WI. Stashenko, E. E., Acosta, R., and Martinez, J. R. (2000). High‐resolution gas‐chromatographic analysis of the secondary metabolites obtained by subcritical‐fluid extraction from Colombian rue (Ruta graveolens L.). J. Biochem. Biophys. Methods 43, 379–390. Staub, J. M., Garcia, B., Graves, J., Hajdukiewicz, P. T. J., Hunter, P., Nehra, N., Paradkar, V., Schlittler, M., Carroll, J. A., Spatola, L., Ward, D., Ye, G.‐N., et al. (2000). High‐yield production of a human therapeutic protein in tobacco chloroplasts. Nat. Biotechnol. 18, 333–338. Stoerker, J., Mayo, J. D., TetzlaV, C. N., Sarracino, D. A., Schwope, I., and Richert, C. (2000). Rapid genotyping by MALDI‐monitored nuclease selection from probe libraries. Nat. Biotechnol. 18, 1213–1216. Sumner, L. W., Duran, A. L., Huhman, D. V., and Smith, J. T. (2002). Metabolomics: A developing and integral component in functional genomic studies of Medicago truncatula. In ‘‘Phytochemistry in the Genomics and Post‐Genomics Eras’’ (J. T. Romeo and R. A. Dixon, Eds.), pp. 31–61. Elsevier, Oxford. Sumner, L. W., Mendes, P., and Dixon, R. A. (2003). Plant metabolomics: Large‐scale phytochemistry in the functional genomics era. Phytochemistry 62, 817–836. Suzuki, H., Reddy, M. S. S., Naoumkina, M., Aziz, N., May, G. D., Huhman, D. V., Sumner, L. W., Blount, J. W., Mendes, P., and Dixon, R. A. (2005). Methyl jasmonate and yeast elicitor induce diVerential genetic and metabolic re‐programming in cell suspension cultures of the model legume Medicago truncatula. Planta 220, 698–707. Suzuki, M., Kao, C. Y., Cocciolone, S., and McCarty, D. R. (2001). Maize VP1 complements Arabidopsis abi3 and confers a novel ABA/auxin interaction in roots. Plant J. 28, 409–418.
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
157
Tadege, M., Ratet, P., and Mysore, K. S. (2005). Insertional mutagenesis: A Swiss Army Knife for functional genomics of Medicago truncatula. Trends Plant Sci. 10, 229–235. Tang, W., Luo, X., and Samuels, V. (2004). Regulated gene expression with promoters responding to inducers. Plant Sci. 166, 827–834. Tanksley, S. D., Young, N. D., Paterson, A. H., and Bonierbale, M. W. (1989). RFLP mapping in plant breeding. New tools for and old science. Biotechnology 7, 257–264. Tanksley, S. D., Ganal, M. W., Prince, J. P., de Vicente, M. C., Bonierbale, M. W., Broun, P., Fulton, T. M., Giovannoni, J. J., Grandillo, S., Martin, G. B., Messeguer, R., Miller, J. C., et al. (1992). High density molecular linkage maps of the tomato and potato genomes. Genetics 132, 1141–1160. Tanner, G. J., Francki, K. T., Abrahams, S., Watson, J. M., Larkin, P. J., and Ashton, A. R. (2003). Proanthocyanidin biosynthesis in plants. Purification of legume leucoanthocyanidin reductase and molecular cloning of its cDNA. J. Biol. Chem. 278, 31647–31656. Tavoletti, S., Veronesi, F., and Osborn, T. C. (1996). RFLP linkage map of an alfalfa meiotic mutant based on an F1 population. J. Hered. 87, 167–170. Taylor, N. L., Heazlewood, J. L., Day, D. A., and Millar, A. H. (2005). DiVerential impact of environmental stresses on the pea mitochondrial proteome. Mol. Cell. Proteomics 4, 1122–1133. Temple, S. J., Whalen, D., Drummond, B. J., Dixon, R. A., and McCaslin, M. (2004). Improving alfalfa digestibility: Field evaluation of transgenic alfalfa engineered for reduced lignin. Abstracts of the National Alfalfa Improvement Conference. Quebec City, Quebec. Tesfaye, M., Temple, S. J., Allan, D. J., Vance, C. P., and Samac, D. A. (2001). Overexpression of malate dehydrogenase in transgenic alfalfa enhances organic acid synthesis and confers tolerance to aluminum. Plant Physiol. 127, 1836–1844. Thiel, T., Michalek, W., Varshney, R. K., and Graner, A. (2003). Exploiting EST databases for the development and characterization of gene‐derived SSR‐markers in barley (Hordeum vulgare L.). Theor. Appl. Genet. 106, 411–422. Thiellement, H., Bahrman, N., Damerval, C., Plomion, C., Rossignol, M., Santoni, V., de Vienne, D., and Zivy, M. (1999). Proteomics for genetic and physioloigcal studies in plants. Electrophoresis 20, 2013–2026. Thimm, O., Blaesing, O., Gibon, Y., Nagel, A., Meyer, S., Krueger, P., Selbig, J., Mueller, L., Rhee, S. Y., and Stitt, M. (2004). MAPMAN: A user‐driven tool to display genomics data sets onto diagrams of metabolic pathways and other biological processes. Plant J. 37, 914–939. Thomas, M. R., and Scott, N. S. (1993). Microsatellite repeats in grapevine reveal DNA polymorphisms when analysed as sequence tagged sites (STSs). Theor. Appl. Genet. 86, 985–990. Tingay, S., McElroy, D., Kalla, R., Fieg, S., Wang, M., Thornton, S., Brettell, R., and Wang, M. B. (1997). Agrobacterium tumefaciens‐mediated barley transformation. Plant J. 11, 1369–1376. Titgemayer, E. C., Cochran, R. C., Towne, E. G., Armendariz, C. K., and Olson, K. C. (1996). Elucidation of factors associated with the maturity‐related decline in degradability of big bluestem cell wall. J. Anim. Sci. 74, 648–657. Tohge, T., Nishiyama, Y., Hirai, M. Y., Yano, M., Nakajima, J.‐I., Awazuhara, M., Inoue, E., Takahashi, H., Goodenowe, D. B., Kitayama, M., Noji, M., Yamazaki, M., et al. (2005). Functional genomics by integrated analysis of metabolome and transcriptome of Arabidopsis plants over‐expressing an MYB transcription factor. Plant J. 42, 218–235. Tuskan, G. A., DiFazio, S., Jansson, S., Bohlmann, J., Grigoriev, I., Hellsten, U., Putnam, N., Ralph, S., Rombauts, S., Salamov, A., Schein, J., Sterck, L., et al. (2006). The genome of black cottonwood Populus trichocarpa (Torr. & Gray). Science 313, 1596–1604.
158
R. A. DIXON ET AL.
Tyagi, A. K., Mohanty, A., Bajaj, S., Chaudhury, A., and Maheshwari, S. C. (1999). Transgenic rice: A valuable monocot system for crop improvement and gene research. Crit. Rev. Biotechnol. 19, 41–79. Tzfira, T., and Citovsky, V. (2006). Agrobacterium‐mediated genetic transformation of plants: Biology and biotechnology. Curr. Opin. Biotechnol. 17, 147–154. Udvardi, M. K., Tabata, S., Parniske, M., and Stougaard, J. (2005). Lotus japonicus: Legume research in the fast lane. Trends Plant Sci. 10, 222–228. Urbanczyk‐Wochniak, E., and Sumner, L. W. (2007). MedicCyc: A biochemical pathway database for Medicago truncatula. Bioinformatics (in press). Valentine, L. (2003). Agrobacterium tumefaciens and the plant: The David and Goliath of modern genetics. Plant Physiol. 133, 948–955. van de Peppel, J., Kemmeren, P., van Bakel, H., Radonjic, M., van Leenen, M., and Holstege, F. C. P. (2003). Monitoring global messenger RNA changes in externally controlled microarray experiments. EMBO J. 4, 387–393. Van Deynze, A. E., Nelson, J. C., O’Donoughue, L. S., Ahn, S. N., Siripoonwiwat, W., Harrington, S. E., Yglesias, E. S., Braga, D. P., McCouch, S. R., and Sorrells, M. E. (1995). Comparative mapping in grasses. Oat relationships. Mol. Gen. Genet. 249, 349–356. Van Eck, H. J., Jacobs, J. M. E., van Dijk, J., Stiekema, W. J., and Jacobsen, E. (1993). Identification and mapping of three flower colour loci of potato (Solanum tuberosum L.) by RFLP analysis. Theor. Appl. Genet. 86, 295–300. van Enckevort, L. J. G., Droc, G., PiVanelli, P., Greco, R., Gagneur, C., Weber, C., Gonza´lez, V. M., Cabot, P., Fornara, F., Berri, S., Miro, B., Ping, L., et al. (2005). EU‐ OSTID: A collection of transposon insertional mutants for functional genomics in rice. Plant Mol. Biol. 59, 99–110. VandenBosch, K. A., and Stacey, G. (2003). Summaries of legume genomics projects from around the globe. Community resources for crops and models. Plant Physiol. 131, 840–865. Velculescu, V. E. (1999). Tantalizing transcriptomes—SAGE and its use in global gene expression analysis. Science 286, 1491–1492. Velculescu, V. E., Vogelstein, B., and Kinzler, K. W. (2000). Analysing uncharted transcriptomes with SAGE. Trends Genet. 16, 423–425. Wagner, G. J. (1991). Secreting glandular trichomes: More than just hairs. Plant Physiol. 96, 675–679. Wagner, G. J., Wang, E., and Shepherd, R. W. (2004). New approaches for studying and exploiting an old protuberance, the plant trichome. Ann. Bot. 93, 3–11. Walden, R. (2002). T‐DNA tagging in a genomics era. Crit. Rev. Plant Sci. 21, 143–165. Wang, H., Li, G., and Chen, R. (2006). Fast neutron bombardment (FNB) induced deletion mutagenesis for forward and reverse genetic studies in plants. In ‘‘Floriculture, Ornamental and Plant Biotechnology: Advances and Topical Issues’’ (J. T. da Silva, Ed.), pp. 629–639. Global Science Books, London, UK. Wang, H. Y., Luo, M., Tereshchenko, I. V., Frikker, D. M., Cui, X., Li, J. Y., Hu, G., Chu, Y., Azaro, M. A., Lin, Y., Shen, L., Yang, Q., et al. (2005a). A genotyping system capable of simultaneously analyzing >1000 single‐nucleotide polymorphisms in a haploid genome. Genome Res. 15, 276–283. Wang, Z., Dai, L., Jiang, Z., Peng, W., Zhang, L., Wang, G., and Xie, D. (2005b). GmCOI1, a soybean F‐box protein gene, shows ability to mediate jasmonate‐regulated plant defense and fertility in Arabidopsis. Mol. Plant Microbe Interact. 18, 1285–1295. Wang, Z.‐Y., and Ge, Y. (2005). Rapid and eYcient production of transgenic bermudagrass and creeping bentgrass bypassing the callus formation phase. Funct. Plant Biol. 32, 769–776. Wang, Z.‐Y., and Ge, Y. (2006). Recent advances in genetic transformation of forage and turf grasses. In Vitro Cell. Dev. Biol. Plant 42, 1–18.
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
159
Wang, Z.‐Y., Hopkins, A., and Scott, M. (2002). Plant regeneration from embryogenic cell suspension cultures of Lolium temulentum. In Vitro Cell. Dev. Biol. Plant 38, 446–450. Wang, Z.‐Y., Ge, Y., Mian, R., and Baker, J. (2005c). Development of highly tissue culture responsive lines of Lolium temulentum by anther culture. Plant Sci. 168, 203–211. Ward, J. L., and Beale, M. H. (2006). NMR spectroscopy in plant metabolomics. In ‘‘Plant Metabolomics’’ (K. Saito, R. A. Dixon, and L. Willmitzer, Eds.), pp. 81–91. Springer‐ Verlag, Berlin, Heidelberg. Warnke, S. E., Barker, R. E., Jung, G., Sim, S.‐C., Mian, M. A. R., Saha, M. C., Brilman, L. A., Dupal, M. P., and Forster, J. W. (2004). Genetic linkage mapping of an annual perennial ryegrass population. Theor. Appl. Genet. 109, 294–304. Watson, B. S., Asirvatham, V. S., Wang, L., and Sumner, L. W. (2003). Mapping the proteome of barrel medic (Medicato truncatula). Plant Physiol. 131, 1104–1123. Weber, J. L., and May, P. E. (1989). Abundant class of human DNA polymorphisms which can be typed using the polymerase chain reaction. Am. J. Hum. Genet. 44, 388–396. Weigel, D., Ahn, J. H., Blazquez, M. A., Borewitz, J., Christensen, S. K., Fankhauser, C., Ferrandiz, C., Kardailsky, I., Malanchaurvil, E. J., NeV, M. M., Nguyen, J. T., Sato, S., et al. (2000). Activation tagging in Arabidopsis. Plant Physiol. 122, 1003–1013. Wesley, V. S., Helliwell, C. A., Smith, N. A., Wang, M. B., Rouse, D. T., Liu, Q., Gooding, P. S., Singh, S. P., Abbott, D., Stoutjesdijk, P. A., Robinson, S. P., Gleave, A. P., et al. (2001). Construct design for eYcient, eVective and high‐throughput gene silencing in plants. Plant J. 27, 581–590. White, G., and Powell, W. (1997). Isolation and characterization of microsatellite loci in Swietenia humilis (Meliaceae): An endangered tropical hardwood species. Mol. Ecol. 6, 851–860. Wienkoop, S., Zoeller, D., Ebert, B., Simon‐Rosin, U., Fisahn, J., Glinski, M., and Weckwerth, W. (2004). Cell‐specific protein profiling in Arabidopsis thaliana trichomes: Identification of trichome‐located proteins involved in sulfur metabolism and detoxification. Phytochemistry 65, 1641–1649. Wilkinson, M., Gessler, D. D., Farmer, A., and Stein, L. (2003). The biomoby project explores open‐source, simple, extensible protocols for enabling biological database interoperability. In ‘‘Proceedings of the Virtual Conference on Genomics and Bioinformatics,’’ pp. 16–26. Williams, J. G. K., Kubelik, A. R., Livak, K. J., Rafalski, J. A., and Tingey, S. V. (1990). DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Res. 18, 6531–6535. WolV, S., Otto, A., Albrecht, D., Zeng, J. S., Buttner, K., Gluckmann, M., Hecker, M., and Becher, D. (2006). Gel‐free and gel‐based proteomics in Bacillus subtilis. Mol. Cell. Proteomics 5, 1183–1192. Wright, E., Dixon, R. A., and Wang, Z.‐Y. (2006). Medicago truncatula transformation using cotyledon explants. In ‘‘Agrobacterium Protocols’’ (K. Wang, Ed.), 2nd ed., pp. 129–135. Humana Press, Totowa, New Jersey. Wu, G., Truksa, M., Datla, N., Vrinten, P., Bauer, J., Zank, T., Cirpus, P., Heinz, E., and Qiu, X. (2005). Stepwise engineering to produce high yields of very long‐chain polyunsaturated fatty acids in plants. Nat. Biotechnol. 23, 1013–1017. Wu, L., Nandi, S., Chen, L., Rodriguez, R. L., and Huang, N. (2002). Expression and inheritance of nine transgenes in rice. Transgenic Res. 11, 533–541. Wu, S. H., Ramonell, K., Gollub, J., and Somerville, S. (2001). Plant gene expression profiling with DNA microarrays. Plant Physiol. Biochem. 39, 917–926. Xie, C., and Xu, S. (1998). EYciency of multistage marker‐assisted selection in the improvement of multiple quantitative traits. Heredity 80, 489–498. Xie, D.‐Y., and Dixon, R. A. (2005). Proanthocyanidin biosynthesis—still more questions than answers? Phytochemistry 66, 2126–2143.
160
R. A. DIXON ET AL.
Xie, D. Y., Sharma, S. R., Paiva, N. L., Ferreira, D., and Dixon, R. A. (2003). Role of anthocyanidin reductase, encoded by BANYULS in plant flavonoid biosynthesis. Science 299, 396–399. Xie, D.‐Y., Sharma, S. B., and Dixon, R. A. (2004). Anthocyanidin reductases from Medicago truncatula and Arabidopsis thaliana. Arch. Biochem. Biophys. 422, 91–102. Xie, D.‐Y., Sharma, S. B., Wright, E., Wang, Z.‐Y., and Dixon, R. A. (2006). Engineering plants for introduction of health‐beneficial proanthocyanidins. Plant J. 45, 895–907. Xu, K., Deb, R., and Mackill, D. J. (2004). A microsatellite marker and a codominant PCR‐ based marker for marker‐assisted selection of submergence tolerance in rice. Crop Sci. 44, 248–253. Xu, M. L., Melchinger, A. E., Xia, X. C., and Lu¨bberstedt, T. (1999). High‐resolution mapping of loci conferring resistance to sugarcane mosaic virus in maize using RFLP, SSR, and AFLP markers. Mol. Gen. Genet. 261, 574–581. Xu, W. W., Sleper, D. A., and Chao, S. (1995). Genome mapping of tall fescue (Festuca arundinacea Schreb.) with RFLP markers. Theor. Appl. Genet. 91, 947–955. Xu, Y. B., McCouch, S. R., and Zhang, Q. (2005). How can we use genomics to improve cereals with rice as a reference genome? Plant Mol. Biol. 59, 7–26. Yamada, K., Lim, J., Dale, J. M., Chen, H., Shinn, P., Palm, C. J., Southwick, A. M., Wu, H. C., Kim, C., Nguyen, M., Pham, P., Cheuk, R., et al. (2003). Empirical analysis of transcriptional activity in the Arabidopsis genome. Science 302, 842–846. Yan, S.‐P., Zhang, Q.‐Y., Tang, Z.‐C., Su, W.‐A., and Sun, W.‐N. (2006). Comparative proteomic analysis provides new insights into chilling stress responses in rice. Mol. Cell. Proteomics 5, 484–496. Yang, S.‐J., Carter, S. A., Cole, A. B., Cheng, N.‐H., and Nelson, R. S. (2004). A natural variant of a host RNA‐dependent RNA polymerase is associated with increased susceptibility to viruses by Nicotiana benthamiana. Proc. Natl. Acad. Sci. USA 101, 6297–6302. Yates, J. R., III (1998). Mass spectrometry and the age of the proteome. J. Mass Spectrom. 33, 1–19. Ye, G. N., Stone, D., Pang, S. Z., Creely, W., Gonzalez, K., and Hinchee, M. (1999). Arabidopsis ovule is the target for Agrobacterium in planta vacuum infiltration transformation. Plant J. 19, 249–257. Ye, X., Salim, A.‐B., Klo¨ti, A., Zhang, J., Lucca, P., Beyer, P., and Potrykus, I. (2000). Engineering the provitamin A (beta‐carotene) biosynthetic pathway into (carotenoid‐free) rice endosperm. Science 287, 303–305. Yousef, G. G., and Juvik, J. A. (2001). Comparison of phenotypic and marker‐assisted selection for quantitative traits in sweet corn. Crop Sci. 41, 645–656. Ytterberg, A. J., Peltier, J. B., and Van Wijk, K. J. (2006). Protein profiling of plastoglobules in chloroplasts and chromoplasts. A surprising site for diVerential accumulation of metabolic enzymes. Plant Physiol. 140, 984–997. Yu, J., Hu, S., Wang, J., Wong, G. K.‐S., Li, S., Liu, B., Deng, Y., Dai, L., Zhou, Y., Zhang, X., Cao, M., Liu, J., et al. (2002). A draft sequence of the rice genome (Oryza sativa L. ssp. indica). Science 296, 79–92. Yu, K., Park, S. J., and Poysa, V. (2000). Marker‐assisted selection of common beans for resistance to common bacterial blight: EYciency and economics. Plant Breed. 119, 411–415. Yu, Y. G., Saghai‐Maroof, M. A., Buss, G. R., Maughan, P. J., and Tolin, S. A. (1994). RFLP and microsatellite mapping of a gene for soybean mosaic virus resistance. Phytopathology 84, 60–64. Zamore, P. D., and Haley, B. (2005). Ribo‐gnome: The big world of small RNAs. Science 309, 1519–1524. Zhang, J.‐Y., Broeckling, C. D., Blancaflor, E. B., Sledge, M., Sumner, L. W., and Wang, Z.‐Y. (2005). Overexpression of WXP1, a putative Medicago truncatula AP2 domain‐containing
BEYOND STRUCTURAL GENOMICS FOR PLANT SCIENCE
161
transcription factor gene, increases cuticular wax accumulation and enhances drought tolerance in transgenic alfalfa (Medicago sativa). Plant J. 42, 689–707. Zhang, J.‐Y., Broeckling, C., Sumner, L. W., and Wang, Z.‐Y. (2007). Heterologous expression of two putative Medicago truncatula ERF transcription factor genes, WXP1 and WXP2, in Arabidopsis led to increased leaf wax accumulation and improved drought tolerance, but diVerential response in freezing tolerance. Plant Mol. Biol. 64, 265–278. Zhang, J. Z. (2003). Overexpression analysis of plant transcription factors. Curr. Opin. Plant Biol. 6, 430–440. Zhou, P. H., Tan, Y. F., He, Y. Q., Xu, C. G., and Zhang, Q. (2003a). Simultaneous improvement for four quality traits of Zhenshan 97, an elite parent of hybrid rice, by molecular marker‐assisted selection. Theor. Appl. Genet. 106, 326–331. Zhou, W. C., Kolb, F. L., Bai, G. H., Domier, L. L., Boze, L. K., and Smith, N. J. (2003b). Validation of a major QTL for scab resistance with SSR markers and use of marker‐assisted selection in wheat. Plant Breed. 122, 40–46. Zhu, H., Kim, D. J., Baek, J.‐M., Choi, H.‐K., Ellis, L. C., Kuester, H., McCombie, W. R., Peng, H.‐M., and Cook, D. R. (2003). Syntenic relationships between Medicago truncatula and Arabidopsis reveal extensive divergence of genome organization. Plant Physiol. 131, 1018–1026.
THE MOLECULARIZATION OF PUBLIC SECTOR CROP BREEDING: PROGRESS, PROBLEMS, AND PROSPECTS Sangam L. Dwivedi,1, Jonathan H. Crouch,2 David J. Mackill,3 Yunbi Xu,2 Matthew W. Blair,4 Michel Ragot,5 Hari D. Upadhyaya6 and Rodomiro Ortiz2 1
Agricultural Science Center at Clovis, 2346, SR288, Clovis, New Mexico 88101 2 International Maize and Wheat Improvement Center (CIMMYT), Apdo 0660 Mexico, D.F., Mexico 3 International Rice Research Institute (IRRI), DAPO, Box 3777, Metro Manila, Philippines 4 Centro Internacional de Agricultura Tropical (CIAT), AA6713, Cali, Colombia 5 Syngenta Seeds Inc., Stanton, Minnesota 55018 6 International Crops Research Institute for the Semi‐Arid Tropics (ICRISAT), Patancheru 502324, Andhra Pradesh, India
I. Introduction to Global Food Production and Major Breeding Challenges II. Development of Markers for Assisting Selection A. Genetic Resources B. Genomic Resources C. Genetic Linkage Map D. Marker‐Trait Associations from Analysis of Diverse Germplasm III. Marker Validation and Refinement A. Markers for Simply Inherited Traits B. QTL Marker for Complex Traits IV. Successful Applications of Marker‐Assisted Genetic Enhancement in Public Sector Breeding Programs A. Resistance to Biotic Stresses B. Tolerance to Abiotic Stresses C. Agronomic and Seed Quality Traits D. Specific Challenges for Alien Gene Introgression V. Successful Application of Marker‐Assisted Genetic Enhancement in Private Sector Breeding Programs VI. Impact of Marker‐Assisted Genetic Enhancement A. Enhanced Selection Power Present Address: International Crops Research Institute for the Semi‐Arid Tropics (ICRISAT), Patancheru 502324, Andhra Pradesh, India.
163 Advances in Agronomy, Volume 95 Copyright 2007, Elsevier Inc. All rights reserved. 0065-2113/07 $35.00 DOI: 10.1016/S0065-2113(07)95003-8
164
S. L. DWIVEDI ET AL. B. Reduced Cost, Increased Feasibility, Time Savings, and Parental Selection C. Overview of Products from Molecular Breeding VII. Approaches to Enhance the EYciency and Scope of Molecular Breeding A. Studying the Molecular Basis of Heterosis B. Fine‐Mapping, Cloning, and Pyramiding of QTL Associated with Improved Agronomic Traits C. Expression QTL Mapping D. Simulation and Modeling of MAS VIII. The Role of Computational Systems in Molecular Breeding Programs A. Germplasm Evaluation B. Managing Breeding Populations C. Genetic Map Construction D. Identifying Marker‐Trait Associations E. Marker‐Assisted Selection F. GEI Analysis G. Breeding Design and Simulation H. Information Management and Integrated Tools IX. Future Prospects for the Molecularization of Public Crop Improvement Acknowledgments References
Molecular markers and genetic maps are available for most important food crops. Marker‐trait associations have been established for a diverse array of traits in these crops, and research on marker/quantitative trait loci (QTL) validation and refinement is increasingly common. Researchers are now routinely using candidate gene‐based mapping and genome‐wide linkage disequilibrium and association analysis in addition to classical QTL mapping to identify markers broadly applicable to breeding programs. Marker‐assisted selection (MAS) is practiced for enhancing various host plant resistances, several quality traits, and a number of abiotic stress tolerances in many well‐ researched crops. Markers are also increasingly used to transfer yield or quality‐ enhancing QTL alleles from wild relatives to elite cultivars. Large‐ scale MAS‐based breeding programs for crops such as rice, maize, wheat, barley, pearl millet, and common bean have already been initiated worldwide. Advances in ‘‘omics’’ technologies are now assisting researchers to address complex biological issues of significant agricultural importance: modeling genotype‐by‐environment interaction; fine‐mapping, cloning, and pyramiding of QTL; gene expression analysis and gene function elucidation; dissecting the genetic structure of germplasm collections to mine novel alleles and develop genetically structured trait‐based core collections; and understanding the molecular basis of heterosis. The challenge now is to translate and integrate this knowledge into appropriate tools and methodologies for plant breeding programs. The role of computational tools in achieving this is becoming increasingly important. It is expected that harnessing the outputs of genomics research will be an important component in successfully addressing the challenge of doubling world food production by 2050. # 2007, Elsevier Inc.
APPLIED CROP GENOMICS
165
I. INTRODUCTION TO GLOBAL FOOD PRODUCTION AND MAJOR BREEDING CHALLENGES Worldwide cereal, legume, oilseed, root and tuber, and plantain and banana crops are grown annually on 1068 million ha with a total production of 3238 million metric tons (Mt) (http://faostat.fao.org/site/340/default.aspx, February 2006); of which cereals contribute 68.6%, roots and tubers 22.0%, legumes 1.9%, oilseeds 4.2%, and plantain and banana 3.3%. Asia is the largest contributor to cereal production (45.9%) followed by North and Central America (21.0%) and Europe (20.5%), while Africa and South America each contributes about 5%. North and Central America (37.3%) and South America (34.9%) dominate legume production, while Asia contributes only 18.2%. Both Africa and Europe contribute about 3% of legume production. For oilseeds, Asia is the largest producer (48.8%) followed by Europe (21.3%), Africa (16.0%), and North and Central America (9.0%), while South America contributes 3.3%. Asia, Africa, and Europe together contribute about 88% to the world production of root and tuber crops, while Africa predominates in plantain and banana production (71.8%) followed by South America (18.1%) and North and Central America (6.9%). Significant trends in production during the period from 1961 to 2005 were noted (Table I). For example, maize has overtaken both wheat and rice; soybean maintains its predominant position among legume crops, although peanut (groundnut) production doubled while beans production slowly but steadily increased by 58%; and substantial increases in cassava and banana production were noted. In contrast, worldwide oat production declined substantially. Millet production remained stagnated, while sorghum production declined by 21% since its peak production in the first half of the 1980s. Across regions, wide variation exists in productivity of these crop commodity groups: cereals from 1.24 t ha1 in Africa to 5.40 t ha1 in North and Central America; legumes from 0.55 t ha1 in Africa to 2.60 t ha1 in North and Central America; oilseeds from 0.78 t ha1 in Africa to 1.76 t ha1 in Europe; root and tuber crops from 8.23 t ha1 in Africa to 24.52 t ha1 in North and Central America; and plantain and banana from 5.61 t ha1 in Africa to 10.05 t ha1 in North and Central America. Many factors have contributed to increased productivity of these food crops: the development of higher yielding cultivars, increased application of fertilizers, herbicides for weed control, insecticides and fungicides for the control of pests, and increases in irrigation. Average increases in productivity vary considerably between crops: for example, maize (except for the period from 1986 to 1990), rice, and wheat productivity has increased steadily throughout the last 45 (1961–2005) years (Table II). In contrast, there were only marginal increases in barley and oat
166
Table I World‐Wide Average Production of the Major Cereal, Legume, Root and Tuber, and Banana and Plantain Cropsa Average production (million Mt) (1961–2005) Crop
1961–1965
1971–1975
1976–1980
1981–1985
1986–1990
1991–1995
1996–2000
2001–2005
28.8 17.3
31.8 21.5
34.6 23.3
38.0 23.1
44.3 25.3
52.8 27.9
61.2 29.8
70.1 32.3
111.8 214.3 2.5 46.8 241.3 4.5 247.7
110.9 261.8 2.9 50.5 287.9 5.5 308.9
139.1 317.7 2.8 49.3 329.8 6.1 354.9
161.9 386.6 2.6 45.5 374.9 6.4 421.8
162.5 435.7 2.8 44.9 442.6 7.0 485.6
171.5 458.9 2.8 40.0 489.8 6.3 532.9
161.5 518.2 2.7 33.3 532.4 6.0 549.2
141.8 597.9 2.8 28.1 587.2 6.2 593.0
143.0 650.8 2.8 26.1 595.7 5.8 594.5
11.8 5.5 7.0 1.0 0.9 10.7 15.5 1.8 28.6
12.0 4.4 6.3 1.1 1.0 9.0 16.8 1.8 40.3
12.7 4.3 6.2 1.1 1.1 8.9 18.1 2.0 53.8
12.9 4.3 6.8 1.1 1.3 9.2 17.6 2.1 75.3
15.0 4.2 6.4 1.1 1.7 10.5 19.8 2.5 90.4
15.6 4.3 6.9 1.6 2.5 14.8 23.1 2.7 100.7
16.2 3.3 7.6 2.3 2.4 13.3 26.5 2.7 119.3
16.6 3.6 8.5 3.2 2.9 11.4 32.4 2.9 150.8
18.7 4.3 8.0 3.7 3.4 10.9 35.4 3.1 192.6
78.3 269.8 100.6 9.4
92.1 291.7 123.8 14.4
103.3 282.4 136.0 13.5
119.9 276.4 140.8 12.0
130.5 273.9 129.8 11.8
144.3 275.4 124.5 15.9
162.4 278.4 128.0 30.6
166.6 308.9 136.7 35.9
193.3 319.5 131.4 39.2
Banana and plantain Banana 23.3 Plantain 14.0 Cereal Barley Maize Millet Oat Rice Sorghum Wheat Legume Beans Broad bean Chickpea Cowpea Lentil Pea Peanut Pigeon pea Soybean Root and tuber Cassava Potato Sweet potato Yam a
(http://faostat.fao.org/faostat/collections?version¼ext&hasbulk¼0&subset¼agriculture).
S. L. DWIVEDI ET AL.
1966–1970
Table II World‐Wide Average Productivity of the Major Cereal, Legume, Root and Tuber, and Banana and Plantain Cropsa Average production (t ha1) (1961–2005) Crop
1961–1965
Banana and plantain Banana 10.81 Plantain 5.42
1966–1970
1971–1975
1976–1980
1981–1985
1986–1990
1991–1995
1996–2000
2001–2005
11.33 5.98
11.49 6.26
12.68 5.91
13.08 5.67
13.34 5.93
14.16 5.97
15.36 6.23
15.74 6.27
1.75 2.34 0.66 1.67 2.22 1.10 1.42
1.87 2.69 0.66 1.67 2.41 1.27 1.62
2.00 3.10 0.68 1.70 2.63 1.38 1.82
2.05 3.46 0.76 1.76 3.08 1.50 2.08
2.26 3.50 0.76 1.79 3.36 1.39 2.37
2.21 3.82 0.73 1.75 3.61 1.36 2.50
2.41 4.29 0.77 1.98 3.84 1.41 2.69
2.54 4.56 0.80 2.13 3.93 1.33 2.78
Legume Beans Broad bean Chickpea Cowpea Lentil Pea Peanut Pigeon pea Soybean
0.49 1.04 0.59 0.31 0.56 0.99 0.85 0.65 1.16
0.51 0.93 0.61 0.21 0.59 1.09 0.87 0.63 1.42
0.54 1.05 0.62 0.25 0.60 1.10 0.90 0.68 1.53
0.54 1.14 0.65 0.34 0.60 1.24 0.95 0.70 1.65
0.59 1.25 0.66 0.32 0.68 1.25 1.06 0.73 1.75
0.60 1.42 0.70 0.35 0.77 1.57 1.17 0.74 1.83
0.65 1.47 0.72 0.34 0.81 1.76 1.24 0.67 2.01
0.66 1.53 0.76 0.36 0.82 1.82 1.40 0.70 2.18
0.71 1.61 0.78 0.38 0.88 1.67 1.42 0.70 2.28
7.68 12.34 7.94 7.50
8.22 13.82 10.62 8.39
8.34 14.03 11.35 7.97
9.00 14.51 11.94 8.58
9.41 14.70 13.53 6.56
9.85 15.35 13.70 8.25
9.81 15.37 14.03 10.21
10.13 16.12 14.85 9.82
10.83 16.81 14.51 9.14
Root and tuber Cassava Potato Sweet potato Yam a
(http://faostat.fao.org/faostat/collections?version¼ext&hasbulk¼0&subset¼agriculture).
167
1.48 2.01 0.58 1.45 1.99 0.96 1.18
APPLIED CROP GENOMICS
Cereal Barley Maize Millet Oat Rice Sorghum Wheat
168
S. L. DWIVEDI ET AL.
productivity during the same period, while millet productivity has stagnated and average sorghum productivity declined. For the legumes, cowpea remained the lowest yielder, while lentil, chickpea, pigeon pea, and beans productivity remained stagnated for most part but broad bean yields steadily increased. In contrast, peanut productivity increased by 67%, while soybean yields consistently increased and remained the top yielder among the legumes. Three distinct patterns have emerged in the productivity of root and tuber and plantain and banana: plantain yield remained stagnant while cassava and yam yield moderately increased. In contrast, substantial increases in productivity were observed for potato, sweet potato, and banana, with potato being the highest yielder among these vegetatively propagated crops. Both abiotic and biotic constraints limit the productivity of all food crops: for example, drought, salinity, temperature (both extreme high and low), phosphorous limitation, and aluminum toxicity in acidic soils among the abiotic stresses, and insect pests and fungal, bacterial, and virus diseases among the biotic stresses are the major constraints to sustainable production of these crops. The biotic constraints of greatest eVect worldwide include bacterial blight (BB) and blast and several virus diseases in rice; rust in wheat, barley, soybean, and common bean; powdery mildew and Fusarium head blight (FHB) in wheat and barley; Barley mild mosaic virus (BaMMV) complex, Barley yellow dwarf virus (BYDV), and Russian wheat aphid in barley; stem borer in rice, corn, and sorghum; Maize streak virus in corn; downy mildew in corn, pearl millet, and sorghum; nematodes in soybean; rust and leaf spots in groundnut; common bacterial blight (CBB) and several virus diseases in common bean; anthracnose in common bean, cassava, and yam; Ascochyta blight in pea and chickpea; Cassava mosaic virus and Cassava brown streak virus in cassava; Yam mosaic virus (YMV) in yam; late blight and several virus diseases in potato; and Black Sigatoka in banana and plantain. Additionally, parasitic weeds, for example Striga, Electra, and Orobanche, seriously limit the production of cereal and legume crops in Africa and Asia. There are many documented cases where these constraints alone or in combination have caused havoc to production and famine in many parts of the world. Some fungal diseases of crop plants also produce mycotoxins that are detrimental to human and animal health. For example, aflatoxin (caused by Aspergillus flavus) in corn and peanut, and deoxynivalenol (DON) (caused by FHB) in wheat and barley pose serious risk to the safety of human food and livestock feed. Conventional breeding is undoubtedly responsible for substantial gains in the productivity of the many food crops, for example, the introduction of dwarfing genes (Sd1 in Dee Geo Woo Gen rice and Rht1 and Rht2 in Norin 10 wheat) and hybrid maize tolerant to high crop density adapted these crops to intensive agriculture worldwide in what is collectively known as the Green
APPLIED CROP GENOMICS
169
Revolution. The Green Revolution helped many developing countries to produce the needed food for their growing population. However, environmentalists, economists, and social scientists criticized this technology for what they assessed as its shortcomings (e.g., use of fertilizers and pesticides as well as monoculture of a few crop cultivars), or who benefited (Swaminathan, 2006). Additionally, only limited progress has been achieved through conventional breeding to address the production constraints with genetically more complex traits such as tolerance to drought and salinity, resistance to pathotypes (in the case of diseases) and biotypes (in the case of pests) with complex inheritance, low heritability, and high genotype‐by‐environment interaction (GEI). From 5.66 billion in 1995, the world population will reach 7.5 billion in 2020, with developing and developed countries’ share accounting for 97.5% and 2.5%, respectively (Pinstrup‐Anderson et al., 1999). The global demand for cereals during the same period will increase by 39% to 2466 Mt; meat by 58% to 313 Mt; and root and tuber crops by 37% to 864 Mt. The large increases in food demand will result not only from population growth but also from urbanization, income growth, and changes in lifestyles and food preferences. The developing countries will account for about 85% of the increase in global demand for cereal and meat. A demand‐driven ‘‘livestock revolution’’ is under way in the developing world and the demand for meat in the developing world is projected to double between 1995 and 2020 (Pinstrup‐Anderson et al., 1999). In response to the strong demand for meat products, demands for cereals for feeding livestock will double in developing countries. Demand for maize in developing countries will increase much faster than for any other cereal and will overtake demand for rice and wheat by 2020. To meet this demand, the world’s farmers will have to produce 40% more grain in 2020. Increases in cultivated area are expected to contribute only about one‐fifth of the increase in global cereal production between 1995 and 2020, so substantial improvements in crop yields will be required to bring about the necessary production increases. This will need to be achieved through a combination of genetic improvements in cultivar and improved agronomic practices. However, without substantial and sustained additional investment in agricultural research and delivery mechanism, it will become more and more diYcult to maintain, let alone increase, yields of these crops in the longer term. As gains from conventional breeding are gradually exhausted, further yield growth will be generated as conventional breeding is combined with wide‐crossing, genomics, and transgenic technologies to tailor crop cultivars with multiple resistance to biotic and abiotic stresses and adapted to diverse agroecological niches (Rosegrant et al., 1995). Crop biomasses are potential raw materials for the production of agricultural biofuels (ethanol from sucrose or starch derived from vegetative
170
S. L. DWIVEDI ET AL.
biomass or grains) or bio‐diesel (from vegetable oils and animal fat). Preliminary work has already demonstrated that a great potential exists to develop cellulose‐based bioenergy systems. This could lead to more demand for cereals (in terms of biomass and grains) for biofuel and oilseeds for bio‐ diesel production that will compete with the demand of these crop commodity groups for food and feed purposes. Multipurpose crops combining food, feed, fiber, and biofuel traits are therefore needed to respond to these market changes (IFPRI 2020 vision for food, agriculture, and the environment). Since the development of DNA marker technology in the 1980s, it has undergone tremendous advances in terms of marker development, genetic maps, functional and comparative genomic linkages, utilization of genome sequencing, and scale and cost of application technologies. As new developments unfold, the power of genomics to facilitate a more genetic‐led approach to plant breeding will be one of the most important advances enabling crop improvement to solve some of the world’s most diYcult problems regarding sustainable agricultural production in many parts of the world. Molecular markers can now be routinely applied to assess and enhance diversity in germplasm collections, to identify genes that control key traits, and to introgress valuable traits from new sources. The ability to introgress beneficial genes under the control of specific promoters through transgenic approaches is another milestone on the path to targeted approaches to crop improvement for which genomic sciences have already identified a vast array of genes that have exciting potential for crop improvement (Delmer, 2005). There are several generic reviews on plant genomics with respect to genetic mapping, quantitative trait loci (QTL) analysis, molecular breeding, and modeling genetic variability of plant responses to environmental stresses (Ası´ns, 2002; Dekkers and Hospital, 2002; Dwivedi et al., 2005; Guo, 2000; Mohan et al., 1997; Stuber et al., 1999; Tardieu, 2003; Varshney et al., 2005a). Similarly, there are a number of crop‐specific reviews on applied genomics, including rice (Ashikari and Matsuoka, 2002; Mackill and McNally, 2004; Xu, 2003), wheat (Koebner et al., 2001), barley (Koebner et al., 2001; Thomas, 2003), common bean (Broughton et al., 2003; Miklas et al., 2006a), cowpea (Ortiz, 2003), peanut (Dwivedi et al., 2003), plantain and banana (Crouch et al., 1998b), yam (Mignouna et al., 2003a), and potato (Barone, 2004). However, in this chapter, we focus on how progress in plant genomics has oVered new opportunities for plant breeders and the extent to which these have been successfully applied in real breeding programs. We then go on to review the essential allied technologies that will be required for successful molecular breeding programs and synthesize the problems and prospects for a future technology‐assisted crop improvement paradigm.
APPLIED CROP GENOMICS
II.
171
DEVELOPMENT OF MARKERS FOR ASSISTING SELECTION A. GENETIC RESOURCES
Plant genetic resources (PGR) are the basic raw materials required to power current and future progress in crop improvement programs. The use of PGR in crop improvement is one of the most sustainable ways to conserve valuable genetic resources for the future, and simultaneously to increase agricultural production and food security. Key to successful crop improvement is a continued supply of genetic diversity including new or improved variability for target traits. The centers of the Consultative Group on International Agricultural Research (CGIAR) have the responsibility to collect, preserve, characterize, evaluate, and document the genetic resources of the cultivated and wild relatives of the cereals (barley, maize, millets, oat, rice, sorghum, and wheat), legumes (Bambara groundnuts, chickpea, common bean, cowpea, faba bean, grasspea, lentil, pea, peanut, pigeon pea, and soybean), roots and tubers (Andean root and tuber crops, cassava, potato, sweet potato, and yam), and Musa (both banana and plantain). Additionally, they have genetic improvement programs that integrate these genetic resources into elite breeding material for use in national cultivar development programs. These germplasm collections are under the aegis of FAO held in trust, and available to researchers globally for diverse use. Collectively, the CGIAR centers possess about 600,000 samples from about 370,000 cultivated accessions, 34,000 wild and weedy accessions, and nearly 177,000 accessions from an uncertain (unknown) category (Table III). The largest representation is of the cereals (64.65%) followed by legumes (30.28%), roots and tubers (4.82%), and Musa (0.25%). The CGIAR System‐wide Information Network for Genetic Resources (SINGER) links the genetic resources information systems of individual CGIAR centers around the world, allowing them to be accessed and searched collectively. SINGER contains key data of more than half a million individual accessions of crops, forage, and agroforestry genetic resources held in the center genebanks (http://www. singer.cgiar.org/). The remaining germplasm are stored in other international, regional, and national genebanks, many of which collaborate closely with CGIAR centers. Crop germplasm collections held in genebanks are the best genetic resources for detailed characterization of important traits such as tolerance to biotic and abiotic stresses, yield, nutrition, and grain quality. These existing diverse germplasm collections are ‘‘gold mines’’ for analysis of allelic diversity. The eYciency of crop improvement programs, whether conventional breeding alone or powered with marker‐assisted selection (MAS), depends on the
172
S. L. DWIVEDI ET AL.
Table III Wild and Cultivated Accessions of the Andean Root and Tubers, Banana, Barley, Bean, Cassava, Chickpea, Faba Bean, Grasspea, Lentil, Maize, Minor Millets, Musa, Oat, Pea, Peanut, Pearl millet, Pigeon pea, Potato, Rice, Sorghum, Soybean, Sweet potato, Wheat, and Yam Preserved in CGIAR Gene Banks No. of accessions stored in CGIAR’s gene bank Crop Andean root and tuber crops Bananaa Barley Barley (wild Hordeum) Barnyard millet Cassava Chickpea Common bean Cowpea Faba bean BPL Faba bean Finger millet Foxtail millet Grasspea Kodo millet Lablab bean Lentil Lima bean Little millet Maize Mung bean Oat Pea Peanut Pearl millet Pigeon pea Potato Proso millet Rice (indica and japonica) Rice (wild) Sorghum Soybean Sweet potato Wheat (bread and durum) Wheat (primitive) Wheat (Triticum and Aegilops) Yam Total
Cultivated
Wild and weedy
1042 979 17,759 15 743 3009 30,748 31,263 11,268
58 178 79 1817
2952 5844 1481 379 658
3025 105 54 1116
2646
498
466 21,993
177
7137 419 2272 1779
Unknown
283 6382
679
14,494 5285 6602
1815 42 6825 40
122 679 1658 14,966 20,844 13,077 4579 842 49,644 33 36,975 193 4717 85,152 525 29 2897 370,055
a Also contains accessions from INIBAP. (http://singer.grinfo.net/).
16 176 453 750 555 2108 644 3789 418
4271
67047 4020 16985
1403 1 5 5126 17 34,175
41,469 84 12 362 176,819
Total 1100 1440 24,220 1832 743 10,825 31,167 33,535 27,541 5285 12,579 5949 1535 3310 658 42 9969 40 466 22,170 122 695 6105 15,419 21,594 13,632 6688 842 11,7335 7842 37,393 17,178 6120 126,622 614 5167 3276 581,050
APPLIED CROP GENOMICS
173
accuracy and precisions of evaluation techniques used to generate appropriate phenotyping data. However, the size of most crop‐related global germplasm collections is simply too vast for systematic evaluation in replicated multilocational trials. Moreover, the diversity of adaptation and major phonological traits of such material highly confounds attempts to generate directly comparable agronomic performance data. Undoubtedly, the robustness of phenotyping is the single most important constraint for eVective selection of appropriate new genetic resources, particularly for abiotic stress tolerance and yield potential. Genomic analysis will have a major role to play in helping to identify subsets of germplasm that are small enough to allow precision phenotyping of replicated multilocational trials for groups of accessions with suYcient homogeneity of phenological and adaptation backgrounds, yet maximum diversity for the target trait: genetically structured trait‐based core collections. The development of core collections has been shown to be a particularly powerful strategy for providing crop breeding programs with a systematic yet manageable entry point into global germplasm resources. Core collections are a cost‐eVective means of identifying accessions with desirable agronomic traits as well new sources of disease and pest resistance or abiotic stress tolerance. Core collections are usually constituted from the 10% of the entire germplasm collection that represents at least 70% of the collections variability in that collection (Brown, 1989). These representative accessions in these core collections are identified based on all available information, including passport data plus botanical and agronomical descriptors. In this way, the development of a core collection has the advantage of displaying much of the phenotypic variability conserved in the genebank in a limited number of accessions. This allows researchers to identify trait‐based hot spots, for example, for new sources of resistance to new isolates or biotypes of diseases and pests at a substantially lower cost than systematically evaluating the entire collection. However, this approach can only be as good as the phenotypic data on which it is based, and thus may not be a more eVective route for identifying the best genetic variability for new traits. In this case, it is hoped that a new generation of core collections based on combined phenotypic and genotyping analysis may be more eVective. Conventional core collections are available in barley, cassava, cowpea, finger millet, maize, Musa, pearl millet, potato, quinova, rice, sorghum, sweet potato, West African yam, and wheat (Table IV), and for several legumes crops (Dwivedi et al., 2005 and reference therein). However, in crops, such as rice, wheat, and maize, or even in legumes, such as chickpea, peanut, and cowpea with large number of accessions stored in the genebank, even a core collection could be unmanageably large so a further reduction is warranted provided it is not associated with losing too much of the spectrum of diversity. Thus, Upadhyaya and Ortiz (2001) developed a two‐stage strategy for developing a mini‐core collection, again based on selecting 10% of the accessions from the core collection representing 90% of the variability of the entire
S. L. DWIVEDI ET AL.
174
Table IV Description of Core Collection in Banana, Barley, Cassava, Cowpea, Finger Millet, Maize, Pearl Millet, Potato, Rice, Sorghum, Sweet potato, West African Yam, and Wheat
Crop Banana
Description
No. of accessions
Caribbean maize Cassava
West African plantain core collection East Asian barley core collection European barley core collection USDA‐ARS barley core collection American barley core collection Core collection Core collection Core collection
380 79 2303 151 670 100 630
Cowpea Finger millet Maize Pearl millet
Core collection Core collection Chinese maize core collection Core collection
2062 622 1193 1600
Potato Rice
Core collection USDA core collection IRRI core collection Core collection Core collection Core collection Core collection
Barley
Sorghum Sweet potato Uruguayan maize West African yam Wheat
25
306 1801 11,200 3475 210 85 720
References Swennen and Vuylsteke, 1987 Liu et al., 1999 Liu et al., 2000a Bowman et al., 2001 Liu et al., 2001a Fu et al., 2005 Taba et al., 1998 Chavarriaga‐Aguirre et al., 1999 Mahalakshmi et al., 2007a Upadhyaya et al., 2006b Li et al., 2004b http://icrtest:8080/ Pearlmillet/Pearlmillet/ coreMillet.html Huama´n et al., 2000 Yan et al., 2004b Mackill and McNally, 2004 Rao and Rao, 1995 Deu et al., 2006 Huama´n et al., 1999 Malosetti and Abadie, 2001
Core collection
391
Mahalakshmi et al., 2007b
Novi Sad Core collection Chinese common wheat core collection
710 340
Kobiljski et al., 2002 Dong et al., 2003
collection. In this process, first a representative core collection is developed using all the available information on geographic origin, characterization, and evaluation data. In the second stage, the core collection is evaluated for various morphological, agronomic, and quality traits to select a subset of 10% accessions from this core subset (or 1% of the entire collection) that captures a large proportion (i.e., more than 80% of the entire collection) of the useful variation. At both stages in selection of core and mini‐core collections, standard clustering procedures are used to separate groups of similar accessions combined with various statistical tests to identify the best representatives. Mini‐core collections are reported for crops such as chickpea (Upadhyaya and Ortiz, 2001), peanut (Upadhyaya et al., 2002), pigeon pea (Upadhyaya et al., 2006c), and rice (1536 accessions, D. J. Mackill, IRRI, personal communication). Evaluation of core and mini‐core collections has been suggested as the most eYcient and reliable
APPLIED CROP GENOMICS
175
means of carrying out an initial search of germplasm collections for desirable traits. Such eVorts have led to the identification of diverse germplasm with beneficial traits in barley (Bowman et al., 2001), quinoa (Ortiz et al., 1999), and many legume crops of significant economic values (see Dwivedi et al., 2005 and references therein; Brick et al., 2006). It is appropriate to emphasize that the core or mini‐core collections do not replace the need for evaluating large parts of the entire collection but simply oVer a means of stratifying the process into more manageable batch sizes that can be evaluated more eVectively. There is no doubt that this approach may still miss some useful alleles that are present at a very low frequency. In this case, for well‐studied traits it may be possible to use genomics technologies to pursue allele mining and gene discovery approaches (Latha et al., 2004; Maccaferri et al., 2005). The genomic revolution, including dramatic advances in molecular biology, bioinformatics, and information technology, provides the scientific community with tremendous opportunities for improving the pace and scale of plant breeding progress and thereby helping to solve some of the world’s most serious agricultural and food security issues. For example, molecular markers can be used for (1) diVerentiating cultivars and constructing heterotic groups; (2) identifying germplasm redundancy, underrepresented alleles, and genetic gaps in current collections; (3) monitoring genetic shifts that occur during germplasm storage, regeneration, domestication, and breeding; (4) screening germplasm for novel genes or superior alleles; and (5) constructing a representative subset or core collection (Xu et al., 2003). This realization led to the formation of the Generation Challenged Program (GCP) (www.generationcp.org). The GCP aims to utilize molecular tools and comparative biology to explore and exploit genetic diversity housed in existing germplasm collections, with a particular focus on improving the drought tolerance of various cereals, legumes, and clonal food crops. A primary goal of the GCP is extensive genomic characterization of global crop‐related genetic resources (composite collections), initially using simple sequence repeat (SSR) markers to determine population structure and now moving onto whole‐genome scans [including single nucleotide polymorphism (SNP) arrays and diversity arrays technology (DArT)] and functional genomic analysis of subsets of germplasm (mini‐composite collections). Thus, the GCP has created composite collections to cover global diversity for most of the 20 CGIAR‐mandated crops. These consists of 3000 accessions or no more than 10% of the total number of available accessions for inbreeding crops and 1500 accessions for outbreeding species (where each accession must be treated as a population). It is expected that this analysis will also lead to the development of genetically broad‐based mapping and breeding populations. The results from these GCP‐supported projects are already starting to flow for the benefit of the scientific community. For example, a global composite collection of 3000 accessions has been developed in chickpea (Upadhyaya et al., 2006a), its genetic structure defined using
176
S. L. DWIVEDI ET AL.
50 polymorphic microsatellites, and a reference collection of 300 accessions identified (ICRISAT/ICARDA unpublished). Further, GCP is supporting a project on allele diversity at orthologous candidate (ADOC) genes that will produce and deliver a public dataset of allelic diversity at orthologous candidate genes across eight important GCP crops and assess whole sequence polymorphism in a DNA bank of 300 reference accessions for each crop. This reference germplasm, which has already undergone genome scan, will be evaluated for traits associated with drought tolerance to test for association between observed polymorphism and trait variability (http://www.intl‐pag. org/14/abstracts/PAG14_W264.html). The mini‐composite collections and the associated marker technologies developed under GCP will be freely available to all those interested in using these genetic and genomic resources. Eshed and Zamir (1994) proposed to exploit introgression lines (ILs), also known as chromosome segment substitution lines (CSSLs) or contig lines (CLs), which could be generated by systematic backcrossing and introgression of marker‐defined exotic segments in elite genetic background. ILs have a high percentage of the recurrent parent genome and a low percentage of the donor parent genome. ILs oVer several advantages over conventional populations: first, they provide useful stocks for highly eYcient QTL or gene identification and fine‐mapping of these; second, they can contribute to the detection of epistatic interactions between QTL; and third, they can be used to map new region‐specific DNA markers (Eshed and Zamir, 1995; Fridman et al., 2004). Several sets of ILs are now available in barley, maize, rice, soybean, and wheat (Table V) that contain beneficial alleles from wild relatives, thus enriching the genetic diversity in primary gene pools of these crops. These ILs when crossed produce progenies with enhanced trait values as demonstrated for increased yield in tomato and wheat (Gur and Zamir, 2004; Shubing et al., 2006). Other useful genetic resources being developed in many crops include recombinant inbred lines (RILs) (Burr et al., 1988), advanced backcross lines (Tanksley and Nelson, 1996), near isogenic lines (NILs) (Muehlbauer et al., 1988), and double‐haploid lines (DHL) (Kasha and Kao, 1970) that can be used to identify genes underlying traits by marker‐ phenotype correlations, dissecting the genetic structure of the complex traits, and for enhancing the trait performance. In addition to naturally available and conventionally bred genetic resources preserved in genebanks, researchers are also creating new genetic variation by using novel technique such as Targeting Induced Local Lesions IN Genome (TILLING), which is a powerful reverse genetics technique that employs a mismatch‐specific endonuclease to detect single base pair (bp) allelic variation in a target gene using high‐throughput assay. Its advantages over other reverse genetic techniques include its applicability to virtually any organism, its facility for high throughput, and its independence of genome size, reproductive system, or generation time (Gilchrist and Haughn, 2005).
Table V ILs (also known as Chromosome Substitution Lines, CSSLs) in Barley, Maize, Rice, Soybean, and Wheat Description of genetic resources Barley (H. vulgare) 146 recombinant chromosome substitution lines, derived from BC2F6 of the cross Harrington and Caesarea (H. vulgare ssp. spontaneum), covering average H. spontaneum genome of 12.5% Two sets of ILs, containing 49 and 43 ILs, derived from BC2DH populations of H. vulgare ssp. spontaneum (ISR42‐8) crossed with German spring barley cultivar Scarlett and Thuringia, covering at least 98.1% and 93.0% of the exotic genome in overlapping introgressions and containing on average 1.5–2.0% additional nontarget introgressions
Rice (O. sativa) 147 ILs from O. sativa (Taichung 65) and O. glumaepatula reciprocal crosses containing O. glumaepatula or Taichung 65 cytoplasm but with entire chromosome segments of O. glumaepatula developed 140 near isogenic ILs derived from a cross between japonica cultivar Nipponbare, and an elite indica line Zhenshan 97B 75 CSSLs, representing on average 97.6% background genome, carrying overlapping chromosome segments of Pai6S in a genetic background of elite cultivar 9311 20,000 ILs in three elite genetic backgrounds (IR64, Teqing, and IR68552‐55‐3‐2) containing a significant portion of loci aVecting complex phenotypes at which allelic diversity exists in the primary gene pool of rice 25 monosomic alien addition lines (MAALs) containing the complete genome of O. sativa and individual chromosomes of O. oYcinalis 159 ILs carrying variant introgressed segments from O. rufipogon GriV. in the background of indica cultivar, Guichao representing 67.5% of the O. rufipogon genome and recurrent parent genome ranging from 92.4% to 99.9%, with an average of 97.4%. The average proportion of donor genome was about 2.2% Soybean (G. max) 22 monosomic addition lines, containing an extra chromosome from G. tomentella to the 2n soybean complement, possess several modified plant characteristics such as flowering habit, plant height, degree of pubescence, seed fertility, number of seeds per pod and plant, pod and seed color, and seed yield
von KorV et al., 2004
Kynast et al., 2001
Sobrizal et al., 1999 Mu et al., 2004 Xiao et al., 2005 Li et al., 2005a Tan et al., 2005 Tian et al., 2006b
Singh et al., 1998
Pestsova et al., 2001 Pestsova et al., 2006
177
Wheat (T. aestivum) 36 homozygous lines carrying diVerent segments of individual chromosomes of Aegilops tauschii genome 84 ILs containing a single homozygous introgression from A. tauschii genome in ‘‘Chinese Spring’’ background
Matus et al., 2003
APPLIED CROP GENOMICS
Maize (Zea mays) Maize chromosome disomic (2n ¼ 6x þ 2 ¼ 44) addition lines for chromosomes 1–4, 6, 7, and 9 and monosomic (2n ¼ 6x þ 1 ¼ 43) addition line for chromosome 8; and for monosomic (n ¼ 3x þ 1 ¼ 23) addition lines for maize chromosome 5 and 10 to a haploid complement of oat isolated from oat maize cross
References
178
S. L. DWIVEDI ET AL.
As TILLING provides mutation in the target gene, it oVers much greater prevision than previous random mutation techniques (using chemical or radioactive mutagens), and it has been successfully used for the detection of both induced and natural variation in several plant and animal species (Perry et al., 2003; Smits et al., 2004; Stemple, 2004; Till et al., 2003, 2004; Wienholds et al., 2003). For example, Slade et al. (2005) generated 246 alleles in the granule‐bound starch synthase 1 (GBSS1) gene (waxy) in wheat using TILLING. Reduction or loss of GBSS1 function results in starch with a decreased or absent amylase fraction, desired for its improved freeze‐thaw stability and resistance to staling compared to conventional starch. Similarly in maize, Till et al. (2003, 2004) screened pools of DNA samples for mutations in 1‐kb segments from 11 diVerent genes, obtained 17 independently induced mutations from a population of 750 pollen‐mutagenized maize plants, and established the public TILLING service for maize modeled on Arabidopsis TILLING project (Till et al., 2003) at Purdue University (http:// genome.purdue.edu/maizetilling). More recently, an EcoTILLING facility has been established at IRRI to identify putative SNPs in both cultivated and wild rice germplasm. EcoTILLING a set of 900 of the Oryza sativa lines for 1800 bp of coding and regulatory region of ERF3 (a candidate gene associated with drought tolerance) identified 31 SNP and short indels that grouped into 9 haplotypes corresponding to the cultivar types (McNally et al., 2006). Powdery mildew is the devastating disease of barley. The genes mlo and Mla are involved in the host plant resistance of barley against the fungal pathogen causing powdery mildew. Mla has multiple alleles at its locus, while mlo is a single copy gene. Using EcoTILLING approach, Mejlhede et al. (2006) not only detected point mutations and deletions in each of the 11 mlo mutants tested but also identified most of the Mla alleles from 25 natural variants of Hordeum vulgare ssp. spontaneum, although the identification was complex due to the presence of highly similar paralogues of Mla. Among the legumes, TILLING is being used to develop soybeans with better seeds (improved oil and protein content and allergen‐free soybeans) (http://www.ars.usda.gov/is/pr/2005/050705.htm). TILLING has great potential to detect both induced and natural polymorphic variation, and as more DNA markers become available and the technological innovations advanced thus reducing the cost of high‐throughput analysis, this technique has great potential for application in crop improvement. These structured mutant populations are also a valuable resource for forward genetic screens. Natural biodiversity is an underexploited sustainable resource that can enrich the genetic basis of cultivated plants with novel alleles and genes to improve yield potential and stability adaptation and resilience. Wild relatives possess a high level of resistance to many biotic and abiotic stresses but are agronomically inferior to modern cultivars (albeit sometimes harboring masked genes of beneficial value for these traits). Tools developed for genetic
APPLIED CROP GENOMICS
179
dissection of traits in cultivated germplasm can also be used to identify and assist the transfer of useful genes from wild relatives (Tanksley and Nelson, 1996) that has been eVectively used for improving both yield and/or seed quality in barley, chickpea, common bean, oat, peanut, pearl millet, pigeon pea, rice, sorghum, soybean, and wheat (Dwivedi et al., 2007). For many crops, the level of genetic diversity in the primary gene pool is narrow. Expanding the genetic base of these crops is, therefore, important for continued crop improvement. Rapid developments in molecular genetic technologies have opened up the vast majority of plant genomes to investigation that in turn will enable the release of genetic variation not previously accessible through conventional crossing and selection.
B. GENOMIC RESOURCES 1.
Genetic Markers
Genetic markers were originally used in genetic mapping to determine the order of the genes along chromosomes, and evolved from morphological markers through isozyme markers to DNA markers which themselves evolved from hybridization‐based detection to polymerase chain reaction (PCR) amplification and now to new sequence‐based systems. Both morphological and isozyme markers are limited in number. Additionally, the morphological markers are aVected by the environment, and a given marker can aVect other morphological traits because of pleiotropic gene action. Consequently, genome‐wide analysis is not feasible using both morphological and isozyme markers. DNA markers are typically derived from a small region of DNA that shows sequence polymorphism between individuals within a species, and may be classified into random DNA markers (RDM) (also known as anonymous or neutral markers), gene‐targeted markers (GTM) (also known as candidate gene marker), and functional markers (FM) (Andersen and Lu¨bberstedt, 2003). RDM are derived at random from polymorphic sites across the genome, whereas GTM are derived from polymorphisms within the gene. FM are derived from polymorphic sites within genes causally associated with phenotypic trait variation and are superior to RDM owing to complete linkage with trait locus alleles (Andersen and Lu¨bberstedt, 2003). The major draw back of the RDM is that their predictive value depends on the known linkage phase between marker and target locus alleles (Lu¨bberstedt et al., 1998). In contrast, once genetic eVects have been assigned to functional sequence motifs, FM derived from such motifs can be used for fixation of gene alleles in a number of genetic backgrounds without additional calibration. FM are superior to GTM and RDM owing to their association with genes of known function.
180
S. L. DWIVEDI ET AL.
a. Random DNA Markers. Restriction fragment length polymorphisms (RFLPs) were the first DNA markers to be developed that have been widely and successfully used to construct linkage maps and detect QTL in many crop species. However, with the discovery of the polymerase chain reaction (PCR) (Saiki et al., 1988), attention shifted to developing a wide range of PCR‐based assays including random amplified polymorphic DNA (RAPD), amplified fragment length polymorphisms (AFLPs), and SSR (also known as microsatellites). RFLP, although providing high‐quality codominant information, is labor intensive, time consuming, requires large amount of DNA, and is dependent on radioisotope‐based protocols. While RAPD and AFLP only provide dominant information; the former suVers from reproducibility problems. However, it is possible to convert tightly linked RFLP markers into PCR‐based sequence‐tagged site (STS) markers (Olson et al., 1989) and both RAPD and AFLP bands can be converted into sequence‐characterized amplified region (SCAR) markers (Paran and Michelmore, 1993) or cleaved amplified polymorphic sequences (CAPs) markers (Konieczny and Ausubel, 1993). Microsatellite markers are ideal DNA markers for genetic mapping and population studies because of their abundance, high level of polymorphism, multiallelic nature, codominant inheritance and wide dispersion in genomes, ease of assay using PCR, and ease of dissemination among laboratories (Powell et al., 1996). Barley has the largest collection of SSR markers followed by rice, wheat, maize, and sorghum (Table VI). Soybean, chickpea, pea, and peanut also have large well‐assembled collections of SSR (Dwivedi et al., 2005; Moretzsohn et al., 2005; Sethy et al., 2006). Other legume crops, such as cowpea and common bean, which are also globally important, are lagging behind in terms of SSR development, as is the case for Musa and many other clonal crops (Table VI). DArT is microarray‐based technique that detects genetic polymorphism, which can be used to construct medium‐density genetic linkage maps in species with various genome sizes (Jaccoud et al., 2001). DArT markers are biallelic and behave in a dominant (present vs absent) or codominant (two doses vs one dose vs absent) manner. DArT is a good alternative to currently used techniques (such as RFLP, AFLP, SSR, and SNP), in terms of cost and speed of marker discovery and analysis, for whole‐genome fingerprinting. It is cost‐ eVective, sequence‐independent, nongel‐based technology that is amenable to high‐throughput automation, discover hundreds of high‐quality markers in a single assay, and integration of DArT markers in genetic map is straightforward. An open source software package, DArTsoft, is available for automatic data extraction and analysis. DArT technology has been successfully developed for barley, cassava, rice, and wheat, while work is in progress to establish DArT in chickpea, pigeon pea, and sorghum (http://www.diversityarrays.com/pub/ huttneretal2005.pdf). For example, a genetic map with 385 unique DArT markers spanning 1137‐cM barley genome (Wenzl et al., 2004) constructed, DArT markers with AFLP and SSR markers mapped on wheat genome
APPLIED CROP GENOMICS
181
Table VI SSR Markers Reported in Banana, Barley, Cassava, Maize, Oat, Pearl Millet, Potato, Rice, Sorghum, Sweet potato, Wheat, and Yam Summary of the marker information reported Banana 24 SSRs from M. acuminata ssp. malaccensis 44 B‐genome‐specific SSRs from enriched library of M. balbisiana cultivar Tani 9 B‐genome‐derived SSRs Barley 45 SSRs from genomic DNA library and from public databases 568 SSRs from database sequences and small‐insert genomic libraries 1856 SSRs from 24,595 ESTs 127 SSRs from genomic DNA of barley cultivar Franka 3530 SSRs from 170,746 ESTs Cassava 14 SSRs containing GA‐repeats from cassava genome 9 SSRs from genomic library of Ipomoea batatas 172 SSRs from 692 putative DNA clones from cassava Maize 6 SSRs from sequences 200 SSRs from maize sequences 655 indels from 8 maize inbreds 1051 SSRs from maize microsatellite‐enriched libraries and microsatellite‐containing sequences from public and private databases 200 SSRs from maize sequences
References
Crouch et al., 1998a Buhariwalla et al., 2005a Oriero et al., 2006 Liu et al., 1996 Ramsay et al., 2000 Thiel et al., 2003 Li et al., 2003b Nicot et al., 2004 Chavarriaga‐Aguirre et al., 1998 Buteler et al., 1999 Mba et al., 2001 Senior and Heun, 1993 Chin et al., 1996 Bhattramakki et al., 2002 Sharopova et al., 2002
http://www.maizegdb.org/ ssr.php
Oat 34 SSRs from three oat microsatellite‐enriched libraries
Li et al., 2000
Pearl millet 50 SSRs from pearl millet BAC clones 18 SSRs from small‐insert genomic library 44 SSRs from a (CA)n‐enriched small‐insert library
Qi et al., 2001 Budak et al., 2003 Qi et al., 2004
Potato 42 SSRs from potato genomic libraries and SSR‐containing sequences in the public databases Rice 2414 SSRs representing 2240 unique marker loci, with majority from regions flanking perfect repeats 24 bp, corresponding to (GA) (36%), (AT) (15%), and (CCG) (8%) motifs. These SSRs along with previously mapped 500 SSRs total 2740 SSRs, 1 SSR every 157 kb
Ashkenazi et al., 2001
McCouch et al., 2002
(continued)
182
S. L. DWIVEDI ET AL. Table VI (continued )
Summary of the marker information reported Sorghum 47 SSRs from sorghum genomic libraries and 2 SSRs from GenBank SSR‐containing sequences 10 SSRs from sorghum genomic libraries and 3 SSRs from database searches 313 SSRs from sorghum BAC and genomic‐DNA libraries 38 SSRs from sorghum genomic DNA libraries Sweet potato 5 SSRs from size‐fractionated genomic libraries 112 SSRs from EMBL database, cDNA, and selectively enriched small‐insert DNA libraries 102 SSRs from small‐insert genomic library, microsatellite‐enriched library, and mining EST‐databases 15 SSRs from Ipomoea trifida sequences, closely related to sweet potato Wheat 230 SSRs from A, B, and D genomes 22 EST‐SSRs and 20 genomic‐derived SSRs 897 EST‐derived SSRs 540 SSRs from A, B, and D genomes in addition to 570 previously reported SSRs Yam 20 SSRs identified from Gnidou parent
References
Brown et al., 1996 Taramino et al., 1997 Bhattramakki et al., 2000 Kong et al., 2000 Jarret and Bowen, 1994 Milbourne et al., 1998 Hu et al., 2004a Hu et al., 2004b
Ro¨der et al., 1998 Eujayl et al., 2002 Gupta et al., 2003 Song et al., 2005
Mignouna et al., 2003b; Scarcelli et al., 2005
(Semagn et al., 2006), and a cassava DArT genotyping array containing 1000 polymorphic clones (Xia et al., 2005) are available and display a high level of polymorphism that shows the genetic relationships among the samples consistent with the information available on them. b. Gene‐Targeted Markers. Expressed sequence tags (ESTs) are currently the most widely sequenced nucleotide element from the plant genomes with respect to the number of sequences and the total number of nucleotides available to researchers. EST provides a robust sequence resource that can be exploited for gene discovery, genome annotation, and comparative genomics. ESTs are typically unedited, automatically processed, single‐read sequences produced from cDNA. Over 38 million sequences have been deposited in the publicly available plant EST sequence databases (dbESTrelease 090806; http://www.ncbi.nlm.nih.gov/dbEST_summary.html). Many of these EST have been sequenced as an alternative to complete genome sequencing or as a substrate for cDNA array‐based expression analysis.
APPLIED CROP GENOMICS
183
Bioinformatics‐based sequence analysis tools have extended the scope of EST analysis into the field of proteomics, marker development, and genome annotation. Although ESTs are no substitute for a whole‐genome scaVold, this ‘‘poor man’s genome’’ resource forms the core foundations for various genome‐scale experiments for less well‐funded crops or species with very large genomes (Rudd, 2003). EST constitutes a novel source of markers that are physically associated with coding regions of the genome. Moreover, ESTs are also a source of SSR in many crops. Kumpatla and Mukhopadhyay (2005) used this approach to examine the abundance of SSR in more than 1.54 million EST belonging to 55 dicotyledonous species. The frequency of EST‐containing SSR among species ranged from 2.65% to 16.82%, with dinucleotide repeats most abundant followed by tri‐ or mononucleotide repeats, thus demonstrating the potential of in silico mining of EST for rapid development of SSR markers for genetic analysis and application in dicotyledonous crops. However, EST‐SSR (also known as genic SSR) produce high‐quality markers, but these are often less polymorphic than genomic SSR (Cho et al., 2000; Eujayl et al., 2002; Thiel et al., 2003). SSR markers may also be transferable to related species and are useful for assaying the functional diversity in natural populations or germplasm collections and also as anchor markers for comparative mapping and evolutionary studies (Varshney et al., 2005b). Tang et al. (2006) identified 428 UNI‐SSR‐EST from wheat genome homologous in rice, maize, and barley. They designed 243 SSR primers and when tested in each species 154 primers produced clear amplicons across the four species, demonstrating a high eYcient transferability of wheat EST‐SSR markers to the other cereal crops. Similarly, Choi et al. (2006) used 274 unigene sequences to develop PCR‐based genetic markers across 15 legume genomes, representing 6 crops or model legume species from the phaseoloid and inverted repeat loss clades. They found 129 of these unigene sequence‐amplified fragments representing single‐copy loci across most target diploid genomes that 70.5% of these markers are intron spanning and 85.3% linked to legume genetic maps. EST resources are also being used to mine SNP (Kota et al., 2003; Picoult‐Newberg et al., 1999). EST provides a quantitative method to measure specific transcripts within a cDNA library and represents a powerful tool for gene discovery, gene expression, gene mapping, and the generation of gene profiles. The National Center for Biotechnology Information (NCBI) database, dbEST 090806 (http://www. ncbi.nlm.nih.gov/dbEST_summary.html), contains the largest collection of EST in rice, wheat, barley, maize, soybean, sorghum, and potato (also see Table VII). Development of EST in cassava is catching up, while only a few hundred ESTs are reported in Musa and other clonal crops (Table VII) and legumes (except for soybean) (Dwivedi et al., 2005; also see Table VII). Clearly, there is an urgent need to develop SSR in the legumes and clonal crops.
184
S. L. DWIVEDI ET AL.
Table VII Expressed Sequenced Tags (ESTs) Reported in Banana, Barley, Cassava, Chickpea, Common Bean, Maize, Oat, Potato, Rice, Sorghum, Soybean, Sweet Potato, and Wheat Summary of the ESTs reported Banana 2286 ESTs from the leaves of M. acuminata ssp. burmannicoides variety Calcutta 4 Barley 13,109 ESTs from 3 cDNA libraries of barley cultivar, Barke, resulting 4,000 genes 271,630 ESTs from 23 barley varieties cDNA libraries resulting 56,302 tentative consensus sequences 110,981 ESTs from 22 cDNA libraries resulting 25,224 unique sequences 437,321 ESTs reported in dbEST release 090806 Cassava 4000 ESTs from cassava mosaic disease resistant genotype 23,000 ESTs from various cassava tissues and genotypes identified 6000–7000 unigenes 5700 unigenes from ESTs of root tissues of cassava varieties with high and low starch contents and those challenged by cassava BB (Xanthomonas axonopodis pv. manihotis) 17,954 ESTs reported in dbEST release 090806
References
Santos et al., 2005
Michalek et al., 2002 Kota et al., 2003 Zhang et al., 2004 http://www.ncbi.nlm.nih.gov/ dbEST_summary.html Fregene et al., 2004 Anderson et al., 2004 Lopez et al., 2004
http://www.ncbi.nlm.nih.gov/ dbEST_summary.html
Chickpea 477 ESTs from root tissue of two closely related genotypes resulted 106 EST‐based markers
Buhariwalla et al., 2005b
Common bean 5255 ESTs from 3 cDNA libraries resulting into 3126 unigenes
Melotto et al., 2005
Maize 73,000 ESTs from multiple organs and developmental stages resulting 22,000 tentative unique genes 1,143,737 ESTs reported in dbEST release 090806 Oat 9792 EST from oat cDNA library detected 2800 cold‐induced UniGene sets 7632 ESTs reported in dbEST release 090806 Potato 61,949 ESTs from aerial tissues, below ground tissues, and tissues challenged with late blight (Phytophthora infestans) identified 19,892 unique sequences 219,917 ESTs reported in dbEST release 090806
Fernandes et al., 2002 http://www.ncbi.nlm.nih.gov/ dbEST_summary.html Bra¨utigam et al., 2005 http://www.ncbi.nlm.nih.gov/ dbEST_summary.html Ronning et al., 2003
http://www.ncbi.nlm.nih.gov/ dbEST_summary.html (continued)
APPLIED CROP GENOMICS
185
Table VII (continued ) Summary of the ESTs reported Rice 1,188,881 ESTs reported in dbEST release 090806 Sorghum 204,208 ESTs reported in dbEST release 090806 Soybean 27,513 unigenes obtained from a variety of soybean cDNA libraries made from a wide array of source tissues and organ systems, developmental stages, and stress or pathogen‐challenged plants 2003 ESTs from full‐length cDNA library of wild soybean (50,109) leaf treated with 150‐mM NaCl 359,158 ESTs reported in dbEST release 090806 Sweet potato 7841 ESTs reported in dbEST release 090806 Wheat 855,066 ESTs reported in dbEST release 090806
References
http://www.ncbi.nlm.nih.gov/ dbEST_summary.html http://www.ncbi.nlm.nih.gov/ dbEST_summary.html Vodkin et al., 2004
Ji et al., 2006 http://www.ncbi.nlm.nih.gov/ dbEST_summary.html http://www.ncbi.nlm.nih.gov/ dbEST_summary.html http://www.ncbi.nlm.nih.gov/ dbEST_summary.html
Target region amplification polymorphisms (TRAP) are derived from a rapid and eYcient PCR‐based technique, which uses bioinformatic tools and EST database information to generate polymorphic markers around targeted candidate gene sequences (Hu and Vick, 2003). This TRAP technique uses two primers of 18 nucleotides to generate markers. TRAP are amplified by one fixed primer designed from a target EST sequence in the database and a second primer of arbitrary sequence except for AT‐ or GC‐rich cores that anneal with introns and exons, respectively. The TRAP technique should be useful in genotyping germplasm collection and in tagging genes with beneficial traits in crop plants. TRAP markers are reported in mapping QTL in wheat (Liu et al., 2005), mapping disease resistance genes in common bean (Miklas et al., 2006b), and for nutritional quality of straw or tolerance to salinity and terminal drought in pearl millet (Mukhopadhyay, Senthilvel, and Hash, ICRISAT, personal communication). SNPs are the most abundant sequence variations encountered in most genomes (Cho et al., 1999; Picoult‐Newberg et al., 1999). Their development costs are similar to those of SSR, but genotyping platforms are now available with very high‐throughput potential and very low unit cost (Kanazin et al., 2002). SNPs are especially useful for association studies because of their high
186
S. L. DWIVEDI ET AL.
frequency across the genome and because they are genetically more stable than SSR. Thus, SNPs are ideally suited for the generation of high‐density genetic maps (Cho et al., 1999). However, currently there are only a few crops with large SNP marker resources; rice, maize, barley, and oat having the largest collection of SNPs (Table VIII). There are also a few hundred SNPs in soybean and common bean, and very few in peanut (Dwivedi et al., 2006). For outbreeding crops, such as maize, polymorphic markers are highly abundant—1 SNP per 60.8 bp (Ching et al., 2002) as compared to inbreeding species such as rice—3.0 SNP per kb in coding regions to 27.6 SNP per kb in transposable elements (Yu et al., 2005)—or barley—1 SNP per 200 bp (Rostoks et al., 2005). More research is needed to fully develop the potential of this class of marker, but this will surely rapidly occur due to the cost eYciencies gained during large‐scale genotyping with SNPs. c. Functional Markers. FM are derived from polymorphic sites within the genes known to be causally involved in phenotypic trait variation. The development of FM requires allele‐specific sequences of functionally characterized genes from which polymorphic, functional motifs aVecting plant phenotype can be identified. Table VIII Single Nucleotide Polymorphisms (SNP) Marker Reported in Barley, Cassava, Common Bean, Maize, Oat, Rice, and Wheat Summary of the SNPs and indels reported
References
Barley 3069 intervarietal and 3377 intravarietal SNP
Kota et al., 2003
Cassava 80 intercultivar and 146 intracultivar SNP
Lopez et al., 2005
Common bean 318 SNP and 68 indel
Melotto et al., 2005
Maize 169 SNP and indel from 36 maize inbreds 14,832 SNP from 102,551 maize EST
Ching et al., 2002 Batley et al., 2003
Oat >2000 genome‐wide SNP Two SNP, SNP‐REMAP and SNP‐RAPD, linked with dwarfing gene, Dw6
Rostoks et al., 2005 Tanhuanpa¨a¨ et al., 2006
Rice 2800 SNP from 3 Oryza ssp. (japonica, indica, and wild rice) 384,431 SNP and 24,557 indels from two subspecies
Nasu et al., 2002 Feltus et al., 2004
Wheat 20 SNP from 12 wheat genotypes 40 SNP from 10 wheat cultivars
Somers et al., 2003 Ablett et al., 2006
APPLIED CROP GENOMICS
187
Dwarf8 in maize encodes a gibberellin response modulator from which FM can be developed for plant height and flowering time. For example, nine sequence motifs in the Dwarf8 gene of maize were shown to be associated with variation in flowering time, and one particular 6‐bp deletion accounted for 7–11 days diVerence in flowering time between inbreds (Thornsberry et al., 2001). However, Dwarf8 is a pleiotropic gene (also aVecting plant height) and thus needs to identify FM from ‘‘additional flowering time genes’’ in addition to using Dwarf8‐derived FM. Orthologues to Dwarf8 have been identified in wheat (Rht1) (Peng et al., 1999), rice (SLR1) (Ikeda et al., 2001a), and barley (sln1) (Chandler et al., 2002), and we know that such genes were bred into the high‐yielding wheat and rice cultivars of the Green Revolution (Hedden, 2003). Altered function of alleles in these orthologous genes can reduce the response to gibberellin and consequently lead to decreased plant height. Thus, biallelic (gibberellin sensitive and insensitive) FM can be derived for targeted and rapid cultivar breeding aiming at increased lodging tolerance. Brown midrib (bm) mutants in maize have an increased digestibility but inferior agronomic performance (Barriere and Argillier, 1993). Two of the four bm genes (bm1 and bm3) are involved in monolignol biosynthesis (Barrie`re et al., 2003). These two genes and additional lignin biosynthesis genes have been isolated based on sequence homology. Candidate genes putatively aVecting forage quality have been identified by expression profiling using isogenic bm lines, and detected association between a polymorphism at the caVeic acid O‐methyltransferase (COMT) locus and digestible neutral detergent fiber (DNDF) in a collection of maize inbred lines (Lu¨bberstedt et al., 2005). Silage maize is a major source of forage for dairy cattle due to its high‐energy content and good digestibility. Lignin structure and cross‐linking between cell wall components influence digestibility (Barrie`re et al., 2003). Analysis of allelic diversity in relation to cell wall digestibility revealed ZmPox3 peroxidase, a candidate gene for silage maize digestibility improvement (Guillet‐Claude et al., 2004), as it is colocalized with a cell wall digestibility and lignification QTL (Barrie`re et al., 2003). GBSS, starch branching enzymes 1 (SBE1) and 3 (SBE3), are major enzymes involved in starch biosynthesis in rice endosperm. Using variation in sequence diversity at Sbe1 and Sbe3 loci and Wx gene markers, Liu et al. (2004c) diVerentiated an indica allele from a japonica allele for both Sbe1 and Sbe3 loci. The same research team also showed that Wx and Sbe3 loci had significant eVects on the amylose content (AC) variation, and together account for 79% of the observed AC variation in a double‐haploid population. The flavor and fragrance of Basmati and Jasmine rice is associated with increased levels of 2‐acetyl‐1‐pyrroline (2AP) (Yoshihashi, 2002). Although various methods are employed to select for fragrant rice, such methods are diYcult, labor intensive, time consuming, require more sampling, and are often unreliable (Reinke et al., 1991). Fragrance in rice is a recessive trait and a deletion in
188
S. L. DWIVEDI ET AL.
the gene encoding BAD2 on chromosome 8 that disables the BAD2 enzyme is the most likely cause of fragrance (Bradbury et al., 2005). Bradbury et al. (2005) used a low‐cost robust technique, allele‐specific amplification (ASA), which allows discrimination between fragrant and nonfragrant rice cultivars and identifies homozygous fragrant, homozygous nonfragrant, and heterozygous nonfragrant individuals in populations segregating for fragrance. This test detects a 355‐bp fragment from a nonfragrant allele and a 257‐bp fragment from a fragrant allele, allowing simple analysis on agarose gels. In wheat, two candidate genes control a QTL for high‐molecular‐weight glutenin subunit (HMW‐GS) GluBx: Glu‐B1‐1 (structural gene coding for Glu1Bx) and spa‐B (the B homoeologous gene coding for SPA) located on the 1BL chromosome at a distance of 1.3 cM from each other within the confidence interval of a QTL for the quantity of GluBx (Guillaumie et al., 2004). In the absence of linkage disequilibrium (LD) between Glu‐B1‐1 and spa‐B, Ravel et al. (2006) conducted an association mapping (AM) study to identify the individual gene responsible for the QTL, and detected significant associations only between Glu‐B1‐1 polymorphism and most of the traits (protein content, the quantity of HMW‐ GS, and protein fractions for each HMW‐GS) evaluated. Malt from barley grains is the raw material for the production of beer. Genetic improvement of malting quality is impaired by the quantitative inheritance and the comparatively low heritability. By monitoring mRNA profiles during grain germination, Potokina et al. (2004) identified between 17 and 30 candidate genes for each of the 6 malting parameters, and 5 of the 8 mapped candidate genes display linkage to known QTL for malting‐quality traits. Genes determining growth habit are well known in diVerent species and all are recognized as CEN/ TFL1 homologous or CEN/TFL1‐like genes (Avila et al., 2006 and references therein). Avila et al. (2006) designed primers for conserved domains from sequences of TFL1/CEN‐like genes and used Hind1II enzyme to produce a clear polymorphism between determinate and indeterminate genotypes in faba bean. This cleaved amplified polymorphism (CAP) marker showed 100% eYciency in discriminating determinate and nondeterminate individuals in an F2 population segregating for growth habit. These examples demonstrate that gene‐based markers are more robust than anonymous markers linked to the trait loci of interest.
2. Genome Sequencing Plant genome sizes vary from the modest—54 million base pairs (Mb) in the bitter cress (Cardamine amara)—to the enormous—124,000 Mb in the lily Fritillaria assyriaca. Among the most important food crops, rice has the smallest genome (389 Mb) (IRGSP, 2005) and wheat has the largest genome (15,999 Mb). Other crops could be grouped into seven classes based on the
APPLIED CROP GENOMICS
189
progressive increase in genome size: Musa, cowpea, and yam (555–613 Mb); sorghum, bean, chickpea, and pigeon pea (709–818 Mb); soybean (1115 Mb); potato and sweet potato (1597–1862 Mb); maize, pearl millet, and peanut (2352–2813Mb); pea and barley (4397–5361 Mb); and oat (11,315 Mb) (Arumuganathan and Earle, 1991). Although plant genomes vary substantially in size, the larger genomes do not necessarily have proportionally more genes, but instead the extra genome size is due to repetitive elements that have proliferated in the genomes of plant species across the plant kingdom (Bennetzen, 1998; Bennetzen et al., 1994). Genome sequencing in most plants is diYcult because of the size and complexity of the genomes. Rice is the first cereal to be fully sequenced (Table IX) because of its importance as one of the major cereals in addition to its small genome size, small number of chromosomes (n ¼ 12), well‐ characterized genetic and genomic resources, and availability of a large number of DNA markers and high‐density genetic linkage map. The extremely large genome of other crops makes them diYcult to sequence. Sequencing hexaploid wheat could yield a considerable amount of important new information about cereals and crop plant biology. The International Wheat Genome Sequencing Consortium (IWGSC) has been formed to advance agricultural research for wheat production and utilization by developing DNA‐based tools and resources that result from the complete sequencing of the expressed genome of common (hexaploid) bread wheat and to ensure that these tools and the sequences are available for all to use without restriction and cost (Gill et al., 2004; http://www.wheatgenome.org/). Sorghum is an important bridge to closely related large‐genome crops in its own tribe such as maize and sugarcane and thus provides a better road map for study of these crops at the DNA level. Sorghum is also a C4 photosynthesis plant which uses a complex combination of biochemical and morphological specializations that result in more eYcient carbon assimilation at high temperature. The genus Sorghum also includes one of the world’s most noxious weeds, the Johnsongrass (Sorghum halepense). The rapid dispersal, high growth rate, and durability that make Johnsongrass such a troublesome weed are actually desirable in many forage, turf, and high‐biomass crops that are genetically complex. Therefore, sorghum oVers novel learning opportunities relevant to weed biology as well as to improvement of a wide range of forage crops. The extremely large size of many cereal genomes makes it diYcult to decode using the standard methods of genome sequencing such as clone‐ by‐clone (Lander et al., 2001) and whole‐genome shotgun (Venter et al., 2001). Determining their complete sequences is daunting and costly. In recent years, two genome filtration strategies, methylation filtration (MF) (Rabinowicz et al., 1999) and C0t‐based cloning and sequencing (CBCS; Peterson et al., 2002) or high C0t (HC; Yuan et al., 2003), have been
190
S. L. DWIVEDI ET AL. Table IX Status of Genome Sequencing in Banana, Maize, Rice, and Sorghum
Summary of sequencing information Banana Two BAC clones of M. acuminata sequenced: MuH9 is 82,723‐bp long with an overall G þ C content 38.2% and gene density of 1 per 6.9 kb while MuG9 73,268‐bp long with an overall G þ C content 38.5% and gene density of 1 per 10.5 kb Maize 100,000 maize sequences reported using methylation filtration method of genome sequencing One‐eighth of the genome of maize inbred B73 sequenced (307 Mb) that contain large percentage of the genes and transposable elements: repeat sequences 58% and genic regions 7.5%, with 59,000 predicted genes 66% of the maize genome consists of repetitive elements; retrotransposons far more frequent than DNA transposons; full‐length genes averaged 4 kb; 42,000–56,000 genes predicted
References
Aert et al., 2004
Palmer et al., 2003 Messing et al., 2004
Haberer et al., 2005
Rice A draft sequence of indica variety 93–11 contains 46,022–55,615 genes. 80% of A. thaliana genes had a homologue in rice but only 49.4% of rice genes had a homologue in A. thaliana A draft sequence of japonica variety Nipponbare consists of 32,000–50,000 predicted genes. 98% of the known maize, wheat, and barley proteins are homologues to proteins in rice. Extensive synteny and gene homology between rice and other cereals but limited synteny with Arabidopsis 95% of the 389‐Mb sequenced genome detected 37,544 nontransposable‐element‐related protein‐coding genes of which 71% had a putative homologue in Arabidopsis. 29% of the 37,544 genes appear in clustered gene families. 2859 genes unique to rice and other cereals, and some might diVerentiate monocot and dicot lineages Of the 38,000–40,000 genes, only 2–3% of these unique to the genomes of indica and japonica rice; 18 distinct pairs of duplicated segments cover 65.7% of the genome and 17 of these pairs date back to a common time before the divergence of the grasses
IRGSP, 2005
Sorghum 300 Mb of the 735‐Mb of sorghum genome sequenced, tagging 96% of the genes with an average coverage of 65% across their length
Bedell et al., 2005
Yu et al., 2002
GoV et al., 2002
Yu et al., 2005
suggested for selectively sequencing the gene space of large genomes. MF is based on the characteristics of plant genomes in which genes are largely hypomethylated but repeated sequences are highly methylated. Methylated DNA is cleaved when transferred into an Mcr þ Escherichia coli strain and only hypomethylated DNA is recovered. CBCS/HC separates single‐ and
APPLIED CROP GENOMICS
191
low‐copy sequences, including most genes, from the repeated sequences on the basis of their diVerential renaturation characteristics. Using the MF strategy, Bedell et al. (2005) sequenced 96% of the genes in sorghum with an average coverage of 65% across their length. This strategy filtered away repetitive elements when sequencing the genome of sorghum that reduced the amount of sorghum DNA to be sequenced by two‐third, from 735 Mb to 250 Mb. Both MF and HC have been used for eYcient characterization of maize gene space (Palmer et al., 2003; Whitelaw et al., 2003). Using HC and MF, Martienssen et al. (2004) generated up to twofold coverage of the gene space with less than 1 million sequencing reads and simulations using sequenced BAC clones predicted that 5 coverage of gene‐rich regions, accompanied by less than 1 coverage of subclones from BAC contigs, will generate high‐quality mapped sequence that meets the needs of geneticists while accommodating unusually high levels of structural polymorphism. Haberer et al. (2005) selected 100 random regions averaging 144 kb in size, representing about 0.6% of the genome, to define their content of genes and repeats for characterizing the structure and architecture of the maize genome (Table IX). Combining CBCS with genome filtration can greatly reduce the cost while retaining the high coverage of genic regions. An alternative approach is the identification of gene‐rich regions on a detailed physical map and sequencing large‐insert clones from these regions. The banana genome is relatively small, 500‐ to 600‐Mb (slightly bigger than rice) DNA across 11 chromosomes. A Global Musa Genomics Consortium (GMGC) is already in place to decode the Musa genome (http://www. newscientist.com/article.ns?id‐dn1037); already two BAC clones of Musa acuminata Calcutta 4 have been sequenced (Table IX). The Musa genome has unique characteristics that will provide researchers with a powerful model for investigating fundamental questions with potentially widespread applications to agriculture. For example, comparing the genome of wild bananas that reproduce sexually with those of asexual crop bananas will provide insights into how quickly plant genomes evolve or comparing the genomes of wild Asian cultivars with those of African cultivars will provide an uncommon look at the eVects of disease agents on genome evolution of the two species (M. acuminata and M. balbisiana), which gave rise to most cultivated bananas. A Global Cassava Partnership (GCP), an alliance of the world’s leading cassava researchers and developers, has proposed that sequencing the cassava genome should be a priority (Fauquet and Tohme, 2004). The US Department of Energy’s Joint Genome Institute (JGI) is providing fund and technical assistance to decode the cassava genome involving 10 institutes (http://www.ars.usda.gov/is/pr/2006/060830.htm). The benefits of deciphering cassava’s genetic code include not only high‐yielding pest‐ and disease‐resistant cultivars with high protein content but also boosting its potential for fuel ethanol in developing countries. Genomic information from cassava could
192
S. L. DWIVEDI ET AL.
also expedite research to reestablish castor bean, a close relative, as domestic source of industrial oil, minus the toxin ricin. Researchers from Purdue University and those from the JGI are sequencing the genome of soybean, Glycine max, the world’s most valuable legume crop, to locate genes on the soybean chromosomes in order to create a physical map. Integrating the physical map with parts of the genetic map already available will ultimately allow sequencing of the entire soybean genome (http://www.csrees.usda.gov/newsroom/news/ csrees_news/06news/soybean_dna.html). Completed genome sequences provide templates for the design of genome analysis tools in orphan species lacking sequence information. For example, Feltus et al. (2006) designed 384 PCR primers to conserve exonic regions flanking introns, using sorghum and millet EST alignment to the rice genome. These conserved‐intron scanning primers (CISPs) amplified single‐copy loci at 37 to 80% success rates; that is, sampling most of the 50 million years of divergence among grass species. When evaluating 124 CISPs across rice, sorghum, millet, Bermuda grass, tef, maize, wheat, and barley, about 18.5% of them seemed to be subject to rigid intron size constraints that were independent of per nucleotide DNA sequence variation. Likewise, about 487 conserved‐noncoding sequence motifs were identified in 129 CISP loci. As pointed out by Feltus et al. (2006), CISP provides the means to eVectively explore poorly characterized genomes for both polymorphism and noncoding sequence conservation on a genome‐wide or candidate gene basis, and also provides anchor points for comparative genomics across a diverse range of species. After sequencing the whole genome of the major food crops, plant breeders may access new gene tools that will facilitate their ability to select outstanding individuals with resistance to biotic and abiotic stresses, possessing good seed quality, and producing more than the existing available cultivars.
C. GENETIC LINKAGE MAP Genetic linkage mapping refers to determining the order and genetic distance between loci along chromosomes using recombination‐based information in segregating populations. In contrast, physical mapping determines the absolute distance between diVerent parts of the genome. Generally, researchers have started by producing a high‐resolution genetic map populated with markers; then produced, fingerprinted, and assembled a deep‐ coverage library of bacterial artificial chromosomes (BACs); and then through comparative analysis of molecular markers, integrated the genetic and physical maps. Marker‐dense meiotic linkage maps serve multiple purposes ranging from dissection of simple and complex phenotypes to the isolation of genes by
APPLIED CROP GENOMICS
193
map‐based cloning (Tanksley et al., 1995), facilitating for the construction of physical maps (Klein et al., 2000), and for developing MAS of desirable genes in breeding programs (Burr et al., 1983; Tanksley et al., 1989). Meiotic linkage mapping uses the frequency of recombination events that occur during meiosis as a basis for calculating genetic distances between loci. The observed recombination frequency is commonly converted into map units (Centimorgan) by applying a mapping function (Kosambi, 1944), and by following the segregation of genetic markers in a meiotic mapping population, recombination events are linearly ordered along each chromosome, thus defining intervening segments of chromosomes, which vary in both physical and genetic size. The size of the mapping population, the number of markers, and the number of recombination events that occur during meiosis greatly influence the quality of resultant map. The genetic map provides a framework for anchoring the physical map. Deep‐coverage large‐insert genomic libraries, such as yeast artificial chromosomes (YAC) or BACs, are used for constructing the physical map. BACs are most preferred over YAC in plants for the construction of large‐insert genomic libraries as they are easy to manipulate, produce low frequency of chimerism, and the clones are highly stable. By merging probe‐to‐BAC hybridization data with DNA fingerprint data, and using the BACRF method (Lin et al., 2000) to resolve the chromosomal origin of BAC clones detected by multiple‐ DNA probes, the robustness of a physical map is improved over other methods that use arbitrary primer PCR‐based fingerprinting of complex DNA populations resulting from pooling of low‐coverage BAC libraries (Klein et al., 2000). Cytogenetic stocks can also be used to generate a physical map by using genetically mapped DNA markers linked to specific chromosomal segments in cytogenetic stocks. However, isolation of a large number of cytogenetic stocks is a daunting task and not possible at all in some crops. For example, deletion stocks are generally not viable in diploid species. Additionally, the resolution of a physical map based on cytogenetic stocks is not only dependent on the number of stocks but also on the accuracy of their cytological characterization. A cytologically defined chromosomal fragment can include several megabases of DNA, which could significantly limit the power of such physical maps. The integrated genetic and physical genome maps are extremely valuable for map‐based gene isolation, comparative genome analysis, and as sources of sequence‐ready clones for genome sequencing. Genetic linkage maps are reported for most of the legumes (Dwivedi et al., 2005; Table X) and for cereals, and clonal crops (Table X), but with varying marker density and genomic coverage. For example, crops such as barley, maize, potato, rice, sorghum, and wheat have high‐density genetic maps, while cassava, Musa, oat, pearl millet, sweet potato, and yam have less saturated genetic linkage maps. Soybean and common bean are the only
Marker and mapping population Azuki bean 486 markers (SSR, RFLP, AFLP) and 187 BC1F1 (JP81481 Vigna nepalensis)
Barley 252 SSR and 86 DHL (Lina H. spontaneum) 1172 markers (AFLP, SSR, STS, and vrs1) and 95 RIL (Russia 6 H.E.S. 4) 1237 markers (SNP, SSR, RFLP, AFLP) and 3 DH populations Black gram 145 markers (RFLP, AFLP, SSR, and morphological) and 180 BC1F1 Cassava 168 markers (RFLP, RAPD, SSR, isozymes) and (TMS 30573 CM 2177‐2) F1
References
486 markers mapped into 11 LGs spanning 832.1 cM with an average marker distance of 1.85 cM, 95% genome coverage, LGs length ranging from 54 to 124 cM and marker loci from 28 to 75 per LG
Han et al., 2005
77 of the 90 loci mapped on 15 LGs (ranging from 4 to 80 cM) with a total map length of 606 cM while 13 segregated independently
Faure´ et al., 1993
242 markers on 7 LGs, with a total map length of 1173 cM that is comparable to those observed in DHLs using RFLPs (Heun et al., 1996) but showing strong segregation distortion around the centromeric region of chromosome 2 H The map consists of 7 LGs with a total distance of 1595.7 cM, and average marker density of 1.4 cM per locus. This map length longer than those of Ramsay et al. (2000) (1173 cM) or Costa et al. (2001) (1387 cM) The integrated map based on 3 mapping populations consisted of 1237 loci, grouped into 7 LGs, with a total map length of 1211 cM and an average marker density of 1 locus per centimorgan
Ramsay et al., 2000
The map consists of 11 LGs with a total distance of 783 cM, markers per LGs ranging from 6 to 23 and average distance between markers varying from 3.5 to 9.3 cM
Chaitieng et al., 2006
The map consists of 20 LGs spanning 931.6 cM, with an average marker density 7.9 cM and covering 60% of the cassava genome. The male gametes‐derived map contains 159 markers, 24 LGs, and 1220 cM map. Reduced recombination in gametes of the female parent resulted greater genetic distances on the male gamete‐derived map between markers common to both parents
Fregene et al., 1997
Hori et al., 2003
Rostoks et al., 2005
S. L. DWIVEDI ET AL.
Banana 90 markers (RFLP, RAPD, isozyme) on 92 F2 (SF265 Banksii)
Summary of the genetic and/or physical map
194
Table X Overview of the Genetic and/or Physical Maps Reported in Azuki Bean, Banana, Barley, Black Gram, Cassava, Maize, Oat, Peanut, Pearl Millet, Potato, Rice, Sorghum, Sweet Potato, Wheat, and Yam
472 SSR and 286 F2 (TMS 30572 CM 2177‐2) Maize 1736 markers (EST and STS, 90 core marker, and 237 from other grass species) and 54 F2 (Tx303 Co159)
Oat 441 markers (RFLP, AFLP, RAPD, STS, SSR, isozyme, morphological) and 136 F6:7 RIL (Ogle TAM O‐301)
510 markers (RFLP, AFLP, and SSR) and 152 F2:6 RIL (Ogle MAM17‐5) (OM)
Okogbenin et al., 2006
The 1736 loci mapped on 10 LGs, with a total map length of 1727.4 cM and marker density of 0. 9 cM. 90 core markers with even spacing along chromosome delineate the 100 bins on the map with an average bin size of 17 cM. This map provides a more than fivefold increase in number of loci compared to previous map published in this population (Chao et al., 1994) but slightly smaller than that of Matz et al., 1995 (1883.6 cM) and Causse et al., 1996 (1765 cM) The 803 loci mapped on 10 LGs, with a total map length 4906 cM (347.7–714.5 cM per chromosome) of IBM map, with an average marker density of 6.6 cM Framework maps consists of 237 and 271 loci in IBM and LHRF populations, that both maps contain 1454 loci (1056 on IBM_Gnp2004 and 398 on LHRF_Gnp2004) corresponding to 954 cDNA probes, and map size of 1825 cM for IBM_Gnp2004 and 1862 cM for LHRF_Gnp2004
Davis et al., 1999
426 loci produced 34 LGs (with 2–43 loci each) spanning 2049 cM of the oat genome (from 4.2 to 174.0 cM per LG). Comparisons with other Avena maps revealed 35 genome regions syntenic between hexaploid maps and 16–34 regions conserved between diploid and hexaploid maps. 89–95% conservation of diploid genome regions on the hexaploid map; however, much lower colinearity at whole chromosome level 28 LGs, containing from 3 to 33 markers and varying in size from 5.2 to 123.0 cM, with a total distance of 1396.7 cM. Comparison with previously published hexaploid map from Kanota Ogle (KO) (O’Donoughue et al., 1995) revealed 9 OM LGs homologous to the LGs in the KO map
Portyanko et al., 2001
Sharopova et al., 2002 Falque et al., 2005
APPLIED CROP GENOMICS
184 RFLP and 748 SSR and 277 RIL (B73 Mo17) 954 cDNA probes and two RIL populations: IBM (B37 Mo17) and LHRF (F2 F252)
The map has 100 markers spanning 1236.7 cM, distributed on 22 LGs with an average marker density of 12.36 cM, and markers uniformly distributed across cassava genome
Zhu and Kaeppler, 2003
(continued )
195
Table X (continued )
Peanut 204 SSR and 93 F2 (Arachis duranensis Arachis stenosperma)
Pearl millet 418 (RFLP and SSR) markers and four populations
RFLP (potato and tomato) and BC1 [(Solanum tuberosum Solanum berthaultii) S. berthaultii] >10,000 AFLP markers and heterozygous diploid potato Rice 726 markers and 113 BC1 (BS125 WL02) BS125 2275 markers and 186 (Nipponbare Kasalath) F2 703 markers and japonica cultivar Nipponbare
References
SSR‐ and AA‐genome‐based map consists of 11 LGs covering 1230.89 cM, with an average marker density of 7.24 cM. This map is comparable to the 1063 cM in previously reported map from two AA‐genome diploid species (Halward et al., 1993) and to half of the 2210 cM reported for tetraploid map (Burow et al., 2001)
Moretzsohn et al., 2005
A consensus genetic map of 353 RFLP and 65 SSR markers mapped on 7 LGs, with 85% of the markers clustered and occupy less than a third of the total map length; marker density in four maps ranged from 1.49 to 5.8 cM.
Qi et al., 2004
304 RFLP loci mapped on the 12 LGs with a total map length of 1034 cM and marker density of 3.4 cM. Comparisons between potato RFLP maps revealed conservation of marker order but diVerences in chromosome and total map length High‐density map contains more than 1000 markers with an average marker density of 1.2 cM, diVerentiating the tomato and potato genomes by 5 chromosomal inversions An ultradense genetic linkage map with >10,000 AFLP loci, with marker density proportional to physical distance and independent of recombination frequency
Gebhardt et al., 1991
The map consists of 12 LGs with a total distance of 1491 cM and average marker density of 4.0 cM on the framework map, and 2.0 cM overall The map consists of 12 LGs with a total distance of 1521.6 cM, and average marker density of 0. 67 cM per locus Physical map of rice chromosome 10 developed using FISH mapping of BAC clones on meiotic pachytene chromosomes that fully integrate with a genetic linkage map of rice chromosome 10 with uniform distribution of genetic recombination but with suppression in centromeric region BAC‐based physical map of chromosome 4 consists of 11 contigs with a total length of 34.5 Mb, 94% of the chromosome size (36.8 Mb), long and short arm length 5.13 and 2.9 Mb, respectively
Tanksley et al., 1992 van Os et al., 2006 Causse et al., 1994 Harushima et al., 1998 Cheng et al., 2001
Zhao et al., 2002
S. L. DWIVEDI ET AL.
Potato 230 RFLP probes and two mapping populations
Summary of the genetic and/or physical map
196
Marker and mapping population
6713 EST from 19 Nipponobare cDNA libraries screened on 4387 YAC clones
2050 RFLP probes and 65 F2 (Sorghum bicolor S. propinquum) Sweet potato AFLP markers and (Tanzania Bikilamaliya) F2 population
Chen et al., 2002
The map consists of 470 loci that mapped into 10 LGs, with a total map distance of 1406 cM and average marker density of 2.99 cM The 1713 cM map encompassed 2926 loci distributed on 10 LGs, and markers mapped between 121 and 243 on these LGs The RIP 1 map consisted of 187 markers (AFLPs, SSRs, RFLPs, and RAPDs) distributed over 10 LGs with a total map length of 1265 cM while RIP 2 map had 228 markers spread into 12 LGs with a total map length of 1410 cM. The combined map contained 339 markers on 11 LGs with a map length of 1424 cM, comparing well with other maps except for few inversion, deletions, and additional distal regions The S. bicolor S. propinquum map is composed of 2512 loci on 10 LGs that collectively span 1059.2 cM, with an average marker density of 0.4 cM
Bhattramakki et al., 2000 Menz et al., 2002
632 (Tanzania) and 435 (Bikilamaliya) AFLP markers placed in 90 and 80 LGs, respectively. Total map lengths were 3655.6 and 3011.5 cM, respectively, with an average distance of 5.8 and 6.9 cM, respectively, between adjacent markers
Wu et al., 2002b
Haussmann et al., 2002
Bowers et al., 2003 Kriegner et al., 2003
APPLIED CROP GENOMICS
Sorghum 470 loci (147 SSR, 323 RFLP) and 137 RIL (BTx623 IS3620C) 2590 PCR‐based markers and 137 RIL (BTx623 IS3620C) 187 markers on 225 RIP 1 (IS9830 E 36‐1) and 228 markers on 226 RIP2 (N13 E36‐1)
BAC‐based physical map of rice developed that represents 90.6% of the rice genome, and its comparison with genetic map reveals that recombination is suppressed severely in centromeric regions as well on short arms of chromosomes 4 and 10 YAC‐based transcript map consists of 6591 ESTs covering 80.8% of the genome, with chromosomes 1, 2, and 3 have relatively high EST densities, approximately twice those of chromosomes 11 and 12, and contain 41% of the total EST sites on the map. Most EST dense regions distributed on the distal regions of each chromosome arm
(continued )
197
198
Table X (continued ) Marker and mapping population Wheat 230 SSR and ITMI population (Opata 85 W7984)
478 SSR and 96 DHL (Kitamoe Mu¨nstertaler) Yam 341 AFLP markers and intraspecific F1 population
References
279 loci amplified by 230 primers placed on to a genetic framework map composed of RFLPs previously mapped in ITMI population. 93 loci mapped to the A genome, 115 to the B genome, and 71 to the D genome. The markers randomly distributed along the linkage map, with clustering in several centromeric regions The genetic map consists of 567 markers assigned to 21 LGs, with a total map length of 3521.7 cM. Approximately similar map length for the A (1148.0 cM), B (1204.8 cM), and D (1168.9 cM) genomes but the D genome had only half the markers (115) of the other two genomes (A, 224; B, 228). This map is very similar in length to those reported for the ITMI map (3551 cM), CS Synthetic map (2,830 cM), Arina Forno map (3086 cM), and other 3 maps of 3164–4110 cM The first SSR‐based linkage map from intraspecific cross of common wheat consisted of 464 loci spread into 23 LGs, with a total map length of 3441 cM covering 86% wheat genome
Ro¨der et al., 1998
The maternal map consists of 155 markers, 12 LGs, 891 cM map distance and 7.4 cM marker density while the paternal map contains 157 markers, 13 LGs, 852 cM map distance and 6.5 cM marker density
Mignouna et al., 2002a
Quarrie et al., 2005 and references therein
Torada et al., 2006
S. L. DWIVEDI ET AL.
567 markers (RFLP, AFLP, SSR, and morphological and biochemical) and 96 DHL (CS SQ1)
Summary of the genetic and/or physical map
APPLIED CROP GENOMICS
199
legume crops that have saturated maps (Dwivedi et al., 2005). The large variation in map length results from diVerences in number of chromosomes and total size of the genomes as well as the use of diVerent numbers of markers (increasing the number of markers will generally, until a certain threshold is reached, give a larger total map length), inclusion of skewed markers (that tend to exaggerate map distances), and use of diVerent mapping software (which vary in estimates of genetic distances). In addition, many published maps report more linkage groups (LGs) than the basic chromosome number of that species. This is frequently the result of insuYcient marker density, as most saturate maps can be directly aligned with the basic chromosome complement (Tekeoglu et al., 2002). The generation of integrated genetic and physical maps in many crops of significant economic importance is a daunting task because of large genome size, large amount of repetitive DNA, and polyploidy nature. However, genome‐wide physical maps are reported in rice (Chen et al., 2002; Cheng et al., 2001), sorghum (Klein et al., 2000), and maize (Coe et al., 2002; Cone et al., 2002; Yim et al., 2002), which will be useful in genome sequencing, targeted marker development, eYcient positional cloning, and high‐ throughput EST mapping in these and also closely related lesser studied crops wherein the genomic resources are not as developed as in these crops. For example, the sorghum genetic and physical map has been aligned to varying degrees with the genetic maps of wheat, rice, sugarcane, maize, and Arabidopsis and with the QTL mapped in these taxa. There is a growing awareness that levels and patterns of allelic diversity are related to the chromosomal context of a locus. ‘‘Diversity maps’’ showing the distribution(s) of allelic diversity across the chromosomes and genomes of a variety of organisms are also related to structural features of chromosomes such as centromeres and telomeres and with the unique selection pressure specific to certain gene pools (Dvorak et al., 1998; Gaut et al., 2000; Hamblin and Aquadro, 1999). Diversity analysis of individual genes promises to shed new light on crop productivity and evolutionary processes underlying plant domestication (Wang et al., 1999). When Draye et al. (2001) constructed diversity maps with genome‐wide resolution based on neutral DNA markers for three gene pools in sorghum (Sorghum propinquum, S. halepense, and S. almum), they found a number of common features and also some key diVerences. Each gene pool showed low levels of variation near the central region of the LG ‘‘G’’ and both termini of the LG. The cultivated sorghum showed by far the lowest level of diversity of the three gene pools, the exotic diploid sorghum showed intermediate diversity, and the polyploids showed remarkably high levels of diversity. Similarly in one region near the marker Psb347, the tetraploid gene pool showed unusually high level of diversity, whereas the two diploid gene pools each showed unusually low levels of diversity. Crops with high resolution of genetic maps, such as rice,
200
S. L. DWIVEDI ET AL.
maize, and sorghum, are ideal for developing diversity maps that promise new information about the consequences of natural selection, domestication, and polyploidy formation. Clearly, the approach of relating molecular level variation to phenotypic diversity is an essential precursor for diversity analysis studies using large populations of candidate genes. In this way, QTL information can be used together with association approaches to select a small number of candidates most likely to be directly related to a specific phenotype.
D. MARKER‐TRAIT ASSOCIATIONS FROM ANALYSIS OF DIVERSE GERMPLASM Conventional genetic linkage mapping approaches for polygenic traits are confounded by epistasis (adaptation and phenology traits influencing the target trait) and GEI (reducing the accuracy of phenotype data) that erodes the precision and power of QTL detection. In addition, linkage mapping has two other major constraints, particularly aVecting practical applications: (1) marker‐trait associations determined in genetic populations must be validated in target breeding populations before routine application can be considered which is time consuming and often introduces a major level of redundancy into the process, and (2) marker‐trait associations identified in this way are based on genetic distance in the mapping population and tight linkage (and thus power of selection) may be eroded or lost entirely when the marker is applied to breeding populations with very diVerent recombination patterns between the target loci and marker. Association mapping (AM), also known as linkage disequilibrium (LD) mapping, is a method that relies on LD to study the relationship between phenotypic variation and genetic polymorphism (Flint‐Garcia et al., 2003). LD refers to nonrandom association between two markers, or two genes, or between a gene and a marker locus. Mutation, population structure, epistasis, population perturbations like migration, inbreeding, and selection all influence LD, and some of these can lead to spurious associations (Jannink and Walsh, 2002). AM deals with unrelated individuals or members of a family with varying levels of phenotypic expression that are evaluated to detect and measure the degree of association between molecular markers and traits of interest. The principal advantage of this procedure lies in its ability to capture informative data stored in unrelated individuals who have undergone several rounds of gene shuZing over multiple generations. Significantly, it can be used on material oVering better overall relevance to breeding programs and thus reduce the level of redundancy between marker identification and marker validation steps. AM can be investigated using candidate genes as well from randomly chosen molecular markers that are evenly distributed across genome. Indeed,
APPLIED CROP GENOMICS
201
for outbreeding crops such as maize, the use of this type of marker in AM is highly desired. There are many reviews describing the fundamentals of LD mapping (Boreck and Suarez, 2001; Flint‐Garcia et al., 2003; Gupta et al., 2005a; Rafalski and Morgante, 2004). Both gene‐based and genome‐wide or chromosome‐wide LD‐based AM detected association of DNA markers with ecology, geography, disease resistance, and agronomic and seed quality traits in higher plants, thus being a viable alternative to classical QTL analyses (Dwivedi et al., 2005 and references therein; Breseghello and Sorrells, 2006a; Gupta et al., 2005a; Kraakman et al., 2006; Maccaferri et al., 2005; Malysheva‐Otto et al., 2006; Morrell et al., 2005; Roy et al., 2006; Stich et al., 2006; Szalma et al., 2005). In addition, many of the associated markers were located in chromosome regions previously identified as harboring QTL for yield and yield components, providing good validation that AM of diverse germplasm is a viable alternative to classical QTL analyses based on crosses between inbred lines (genetic populations), especially for complex traits (Breseghello and Sorrells, 2006a; Kraakman et al., 2006; Szalma et al., 2005). Large variation in LD estimates in diVerent plant genomes or in diVerent parts of the genome of an individual species is reported: 10–20 cM in barley and wheat, 100 kb in rice, <4 to 10 kb in sorghum, <50 kb in soybean (all self‐pollinated species). The LD estimates in cross‐pollinated crops ranged from 0.4 to 1.0 kb in maize, <3 cM in sugar beet, 0.3–1.0 cM in potato, and 10 cM in sugarcane (Gupta et al., 2005a and references therein). Inbreeding drives lineages to homozygosity rendering recombinations ineVective in breaking down LD, while rapid decay of LD in outbreeding is probably because of increased crossover eVects. Population‐wide associations between loci due to LD can be used to map QTL with high resolution. However, spurious associations between markers and QTL can also arise as a consequence of population stratification and statistical methods that cannot diVerentiate between loci associations due to linkage disequilibria from those caused in other ways can render false‐positive results (Deng et al., 2001). The transmission‐disequilibrium test (TDT) is a robust test for detecting QTL. TDT exploits within‐family associations that are not aVected by population stratification (Spielman et al., 1993). It is used to check jointly for linkage and LD by testing whether alleles at a particular marker locus segregate randomly from parents to a specific subset of their oVspring. TDTs have been developed for dichotomous and quantitative traits (Allison, 1997; Martin et al., 2000; Rabinowitz, 1997; Zhao et al., 2000). However, some TDTs are formulated in a rigid form, with reduced potential applications. Herna´ndez‐Sa´nchez et al. (2003) developed TDT that uses mixed linear models to allow greater statistical flexibility. In this test, allelic eVects are estimated with two independent parameters: one exploiting the robust within‐family information and the other the potentially biased
202
S. L. DWIVEDI ET AL.
between‐family information. Using this approach, they confirmed previous observations on eVects of the fourth melanocortin receptor (MC4R) on production traits in pig that polymorphism is either causal or in very strong LD with the causal mutation, and provided no evidence for spurious associations. Breseghello and Sorrells (2006b) compared the potentials and limitations of germplasm bank collections, synthetic populations, and elite germplasm as experimental materials for association analysis integrated with plant breeding practices and the application of AM diVers among those populations in several aspects. They found that synthetics oVer the most favorable balance of power and precision for association analysis and would allow mapping of quantitative traits with increasing resolution through cycles of intermating. Hence, Breseghello and Sorrells (2006b) proposed a model to describe the association between markers and genes as conditional probabilities in synthetic populations under recurrent selection, which can be computed on the basis of assumptions related to the history of the population. This model is useful for predicting the potential of diVerent populations for association analysis and forecasting the response to MAS. For eYcient integration of AM with other methods currently in use, materials that are routinely generated and evaluated should be used for both purposes. For example, in case of germplasm, core collections (see Section II.A) are expected to represent a large proportion of the total genetic variability with a manageable number of accessions, and thus are suitable for genetic studies. Core collections representing the genetic diversity of a species are attractive for AM because of the wide allele diversity encompassed within a relatively small number of genotypes for which replicated multilocational precision phenotyping is feasible. The level of LD in a crop germplasm collection determines the scale at which AM will resolve the localization of favorable variation in the genome. The use of genome‐wide survey for selecting a less‐structured subsample of accession improves the significance of results and thus opens the door to genome‐wide association studies and supports the identification of reference collection to integrate phenotypic and molecular characterization eVorts (Deu and Glaszmann, 2004). The process of selection of a minimum sample with maximum variation has a normalizing eVect that is expected to reduce population structure and LD between unlinked loci, thus creating a situation favorable for AM (Breseghello and Sorrells, 2006a). A diYculty likely to occur in this type of material is related to genetic heterogeneity within samples. Thus, it is not recommended at this time to use primary landraces and natural populations or any other mixture of genotypes, which will confound the genotyping and erode the precision of phenotyping. For elite materials, the sample could be composed of lines and checks evaluated in regional trials, whereas for synthetic populations, the evaluation unit should be largely homogeneous, whether it is an individual or
APPLIED CROP GENOMICS
203
a family. Core collections are useful materials for AM for quality traits such as disease resistance, seed quality, and domestication‐related traits. Conversely, the broad genetic variability of those collections normally makes them unsuitable for analysis of quantitative traits because part of accessions would be unadapted to the growing conditions and prevalent diseases of the test environment, resulting in poor precision of trait measurement. Similarly, phenological traits are likely to be highly variable in core collections which will highly confound attempts to measure traits such as abiotic stress tolerances. Elite lines are the most desirable materials for AM when attempting to analyze low heritability traits, including yield, yield components, and tolerance to abiotic stresses because elite lines are genetically stable and are well adapted to specific known growing conditions (Breseghello and Sorrells, 2006b). Synthetic populations are normally designed and maintained by random mating, and therefore population structure is expected to be mild or absent, which is an important advantage of synthetics for AM. The level of LD in synthetic populations is expected to be high in the initial generations, such that a genome scan could detect large chromosome segments associated with traits, and trace them back to parental haplotypes. In subsequent generations, the decay of LD by recombination would favor refined mapping. However, synthetic populations are often subjected to intense recurrent selection which could build up LD by favoring allelic combinations or by promoting genetic drift (Palaisa et al., 2003). For this reason, populations subjected to mild or no selection would be preferred for AM. Alternatively, marker analysis of a large number of available genotypes can be used to define a subset of lines that represent the desired population structure for AM. AM in synthetic populations under selection will require intensive genotyping because in each cycle, new progenies have to be tested to reflect the current state of the population and for implementation of MAS. On the other hand, information about the population is cumulative over years, allowing a progressively refined genetic analysis of traits of interest to the breeding programs. Both linkage analyses (LA) and LD mapping have their own limitations when used alone. Therefore, a joint linkage and LD mapping strategy has been devised for genetic mapping (Wu and Zeng, 2001; Wu et al., 2002a) that has power to simultaneously capture the information about the linkage of the markers (as measured by recombination fraction) and the degree of LD created at historic time. This approach is based on the principle that during the transmission of genes from parents to progeny, linkage between marker and QTL is broken due to meiotic recombination. Thus, by combining the information about the linkage and LD, the joint mapping method displays increased power to detect LD compared to traditional methods of LD analyses. The use of this approach has also been suggested for multitrait fine‐mapping of QTL (Lund et al., 2003; Meuwissen and Goddard, 2004).
S. L. DWIVEDI ET AL.
204
Like the genetic and physical maps developed in many plant genomes, LD maps can also be constructed in plants as is being done in humans using ALLASS and LDMAP VERSION 0.1 (University of Southampton, United Kingdom) softwares. These LD maps will make use of molecular markers that flank marker intervals delimited on the basis of estimations of LD, the distance being represented as LD units (Zhang et al., 2002).
III.
MARKER VALIDATION AND REFINEMENT
It is clear from Section II that there have been major advances that have occurred in the development of DNA markers, construction of genetic linkage maps, and the mapping of economic traits controlled by major genes and QTL. While the number of reports of mapped genes continues to grow rapidly, the literature on practical validation application of those markers in breeding populations remains relatively limited. One reason for this is that there are several scientific and logistical issues that must be resolved before a practical MAS strategy can flow from a mapping study, and at each step there will be a certain level of redundancy. Moreover, in some cases, researchers are more interested in understanding the genetic control of the trait and subsequent gene discovery, thus leave the validation and application to plant breeders who may be less interested in publishing their findings. Furthermore, once the mapping study is published, it may be diYcult to publish the results of activities associated with validation, refinement, and application of those markers, particularly if the selective power of the marker lessened or lost when applied in breeding programs. This generally involves validation of the QTL or gene marker in a diVerent set of germplasm or populations and development of markers assays suitable for high throughput, low cost, and MAS (Collard and Mackill, 2007; Langridge et al., 2001). Marker validation step usually has some level of redundancy leading to the need to develop new markers or marker types around the target locus in order to find more polymorphic markers or develop gene‐ based markers for marker‐trait associations that are shared across diVerent breeding populations. The availability of thousands of SNP markers rather than several hundreds of SSR markers in some crops (Table VIII) that are currently being used makes it practical to validate marker‐trait association through high‐precision genotyping using the same set of markers for diVerent parental lines and breeding populations. Thus, it is much more likely that the parents of any breeding population will be polymorphic for at least one of them, allowing breeders to track the alleles donated from each parent throughout the breeding process, speeding MAS and marker‐assisted back crossing (MABC) in any cross. Marker validation can be also done through
APPLIED CROP GENOMICS
205
selective genotyping and pooled DNA analysis, and development of gene‐ based markers and closely linked markers, as additions to testing marker‐ trait association in alternative or target populations. However, validation requirements can be minimized by MAS using large‐eVect QTL, precision phenotyping, identification of context independent QTL, mapping as we go, AM using large numbers of inbreds, genome‐wide association scan, using breeding materials for mapping, and utilization of haplotype‐based selection rather than single‐marker based selection.
A. MARKERS FOR SIMPLY INHERITED TRAITS For major gene traits such as many disease resistances, gene validation is fairly straightforward. In these cases, the eVect of genetic background is usually minimal, and the ease of phenotyping makes fine‐mapping of the gene simpler. In mapping studies, a gene for simply inherited trait can be mapped with adequate accuracy in a mapping population of 100–200 individuals. This can then be followed by fine‐mapping involving larger populations of over 500 individuals. The fine‐mapping will allow identification of tightly linked markers that will not suVer recombination between marker and target gene in segregating breeding populations. An alternative to use a tightly linked gene in MAS is to use flanking markers on either side of the gene. Use of both flanking markers ensures that the gene is accurately detected in segregating populations, but it can also result in the transfer of large chromosomal fragments along with the target gene (linkage drag) if the interval between the two markers is large. If the donor of the gene contains deleterious alleles that are linked to the target gene, it will be necessary to identify more tightly linked flanking markers (Frisch et al., 1999a; Tanksley et al., 1989). The process of fine‐mapping can be carried forward to positional cloning of the target gene. Plant populations of several thousands are commonly used even in species with small genomes where recombination rates might be around 250 kb cM1 (Durrett et al., 2002). The marker or markers identified during the process of fine‐mapping may be suitable for direct application in breeding programs following some level of validation. However, in many cases, these markers may not be polymorphic in all breeding populations of interest, thus requiring the identification of alternative markers for those populations. For well‐characterized genomes, this is straightforward. In rice, for example, any one of the 2414 SSR markers can be quickly identified from the dense public maps or located using the genome sequence in online databases. In addition to identifying markers tightly linked to the gene of interest, it is also useful to identify a similar set of around 10 markers 3–10 cM either side of the target gene (Langridge et al., 2001). These markers can then be used to reduce the eVects
206
S. L. DWIVEDI ET AL.
of linkage drag if recombinant selection is practiced (Collard and Mackill, 2007). An ideal marker for selection of the target gene would be one that provides 100% accurate prediction of the phenotype. Except for the traits in alien gene introgression regions, this usually requires a marker associated with the sequence change in the gene associated with the favorable allele. These are so‐called ‘‘FM’’ (Andersen and Lu¨bberstedt, 2003) or ‘‘perfect markers’’ (see Section II.B). These markers provide suYcient benefits for MAS application to justify cloning of important economic genes and QTL aside from the other benefits that gene discovery can bring (see Section VII.B).
B. QTL MARKER FOR COMPLEX TRAITS The diYculty for phenotypic selection of many quantitative traits in plant breeding gave rise to an optimistic view of the prospects of MAS for QTL. However, to date very few studies have demonstrated the usefulness of marker‐ QTL information vis‐a`‐vis conventional phenotypic selection for the development of genetically enhanced breeding populations. Many studies reported that no substantial genetic progress was achieved or only a fraction of putative QTL actually contributed to the improvement of the trait when validated through MAS (Bohn et al., 2001; Bouchez et al., 2002; Flint‐Garcia et al., 2003; Schneider et al., 1997; Stromberg et al., 1994; Yousef and Juvik, 2001a). Several factors contribute to false positive (Type I errors) in QTL mapping studies, including population structure and size, parental selection and genetic background eVects, epistasis and inaccurate phenotyping, QTL environment interaction and inappropriate evaluation conditions, and finally inappropriate logarithm of odds (LOD) thresholds or low statistical rigor (Beavis, 1998; Moreau et al., 1998). Additionally, inaccurate phenotyping data in the mapping populations further reduce the capacity to detect real QTL. In a literature search conducted for the crops under review from 1991 to 2005 in journals with high‐impact factor, over 500 articles reported QTL contributing to phenotypic variance for several agronomic and seed quality traits as well resistance to biotic and abiotic stresses, predominantly in cereal crops such as barley, maize, rice, and wheat. In contrast, during the same period, there were only 80 articles that dealt with validation of the reported QTL (Tables XI and XII), concentrating mostly in wheat, barley, rice, maize, and few in common bean, soybean, pea, yam, and potato. However, the community has become more concerned about reporting false QTL discovery, with a resultant increase in the number of reports regarding validation of QTL. The low resolution of most QTL mapping studies reduces the likelihood of successful QTL marker validate (Holland, 2004). In a milestone publication by Beavis (1998), the power, precision, and accuracy of QTL mapping was
Table XI Validation of Marker/QTL Associated with Resistance to Biotic and Abiotic Stresses in Barley, Common Bean, Maize, Pea, Potato. Rice, Soybean, Wheat, and Yam Trait
Gene
Validated marker/QTL
References
Biotic stresses
BaMMV and BaYMV Barley stripe rust (BSR) (Puccinia striiformis Westend. f. sp. hordei)
ym4 QTL4, QTL5, and QTL7
BYDV BaYMV FHB (Fusarium graminearum Schwabe) and Kernel discoloration (KD) Leaf rust (Puccinia hordei) Leaf stripe (Pyrenophora graminea) Net form of net blotch (NFNB) [Drechslera teres (Sacc.) Shoem. f. teres Smedeg] Net type net blotch (NTNB) (Pyrenophora teres f. teres) Powdery mildew (Erysiphe graminis f. sp. hordei) Russian wheat aphid (RWA) [Diuraphis noxia (Mordvilko)] Spot blotch (SB), NTNB, Septoria speckled leaf blotch (SSLB), and leaf scald (LS)
Yd2 rym1 and ryn5
13 QTL Rdg2a 7–12 QTL
1–6 genes
Closest marker 2.2 cM from the resistance locus Rrs.B87 OP‐ZO4H660 QTL4 and QTL5 linked with BSR resistance at seedling stage; three QTL linked with BSR resistance at adult plant stage YLM A CAPS marker from an RFLP probe MWG2134 Two major QTL (near HVBKasi and the Vrs1 locus); a major QTL for KD and a QTL for FHB Six QTL (Rphq1–6) MWG2018 EBmac0906 and Bmac0181
Ml(La)
M61P12K116, M55P13T311, Bmag0173, and Ebmac0874 l MWG097‐R,L and MWG097
Two genes
ABG8 and KV1/KV2
2 QTL each for SB, NTNB, and SSLB and one QTL for LS
Rcs‐qtl‐7H‐2‐4 and Rcs‐qtl‐4H‐4‐6 for SB; Rpt‐qtl‐3H‐4 and Rpt‐qtl‐4H‐5‐7 for NTNB; Rsp‐qtl‐2H‐7‐11 and Rsp‐qtl‐6H‐10‐14 for SSLB, and Rrs‐qtl‐1H‐1‐4 for LS
Williams et al., 2001 Ordon et al., 1995 Castro et al., 2003a,b
Paltridge et al., 1998 Okada et al., 2003a Canci et al., 2004; Mesfin et al., 2003 van Berloo et al., 2001 Arru et al., 2003 Raman et al., 2003
Cakir et al., 2003 Mohler and Jahoor, 1996 Raman and Read, 2000 Yun et al., 2006
207
Rrs.B87
APPLIED CROP GENOMICS
Barley Barley leaf scald (Rhynchosporium secalis)
(continued )
208
Table XI (continued ) Trait
Gene
Validated marker/QTL
References
Common bean BCMV CBB (Xanthomonas campestris pv. Phaseoli)
Pea Ascochyta blight (Mycosphaerella pinnodes, Phoma medicaginis variety pinodella, Ascochyta pisi)
Melotto et al., 1996
R7313 and R4865
Tar’an et al., 1998
SBD51300
Miklas et al., 2000
5 QTL
Three QTL for SDM and two QTL for RDM, a major QTL confers resistance to SDM and RDM
Nair et al., 2005
Many QTL
Six QTL on LG II, III, IV, V, and VI (two QTL)
Timmerman‐Vaughan et al., 2004
Ny and Ry
CD17, GP125, CT168, and TG508 linked with Ryadg
Ha¨ma¨la¨inen et al., 1997
Pi44(t) Pi‐z
AFLP348 MRG5836
Pi‐ta2, Pi‐kh, Pi‐ks, and Pi‐b Pi‐z Gm2
SSRs Pi‐b (RM138, RM166, RM208), Pi‐kh (RM144, RM224), and Pi‐ta2 (OSM89, RM155, RM7102) SSRs AP5659‐1, AP5659‐3 and AP5659‐5 F10600 and F81700
Chen et al., 1999 Conaway‐Bormans et al., 2003 Fjellstrom et al., 2004
–
Six QTL
Potato PVY Rice Blast [Pyricularia grisea (Cooke) Sacc.]
Gall midge (Orseolia oryzae Wood‐Mason) Sheath blight (Rhizoctonia solani Kuhn)
Fjellstrom et al., 2006 Nair et al., 1995 Pinson et al., 2005
S. L. DWIVEDI ET AL.
Maize Sorghum downy mildew (SDM) (Peronosclerospora sorghi) and Rajasthan downy mildew (RDM) (Peronosclerospora heteropogoni)
SW13690
A dominant gene, I, and six recessive genes A major and few minor genes bc‐I2
Soybean Brown stem rot (BSR) (P. gregata)
Root knot nematode [Meloidogyne incognita (Kofoid and White) Chitwood] Soybean cyst nematode (SCN) (H. glycines Ichinohe)
Hessian fly [Mayetiola destructor (Say)] Leaf rust (Puccinia recondita f. sp. tritici)
Klos et al., 2000
Few genes
BSR3.sp1, K375.sp1, 14H13.sp1, 21E22.sp1, 21E22.sp2, 30L19.sp1, 35E22.sp1, 98P22.sp2, and Satt244 Satt012, Satt358, Satt492, and Satt505
rhg1, rhg2, rhg3, rhg4 and rhg5
Two major QTL against resistance to SCN race 3 QTL containing rhg1 on LG G and QTL rhg4 on LG A2
Wang et al., 2001b Concibido et al., 2004
Bt‐10
UBC196590
Demeke et al., 1996
H1 to H25 Lr9
Sumai 3‐derived QTL on 3BS and 6BS gwm389, gwm493, gwm533, and gwm644 SSRs linked to the major QTL on chromosome 3BS OpA01 and OpA17 OPA‐071500, OPR15950, and J13/1 þ 2
Lr19 Lr10
Ep‐D1c Lrk10–6
Lr28 Lr28, Lr35, and Lr39 Lr19 and Lr24 Lr9, Lr10, Lr19, Lr24, Lr28, Lr29, Lr35, and Lr39
OPJ01378 Puc19/HpaII900 STS RFLP and AFLP markers 1100 bp, 310 bp, 130 bp, 310 bp, 850/900 bp, 900 bp, and 100 bp
Anderson et al., 2001 Yang et al., 2003 Zhou et al., 2003b Dweikat et al., 1994 Schachermayr et al., 1994 Winzeler et al., 1995 Schachermayr et al., 1997 Naik et al., 1998 Sharp et al., 2001 Singh et al., 2004 Blaszczyk et al., 2004
SCS5550
Li et al., 2001a
APPLIED CROP GENOMICS
Wheat Common bunt [Tilletia tritici (Bjrk.) Wint. and T. laevis Kuhn] FHB (F. graminearum)
Rbs1, Rbs2, and Rbs3
Gupta et al., 2005b
209
(continued )
Trait
Gene
Leaf rust and leaf tip necrosis (LTN)
Lr34 and Ltn
Powdery mildew [E. graminis DM f. sp. tritici (Em. Marchal)]
Pm1 and Pm2 Pm1 to Pm25
Yam (Dioscorea spp.) YMV in white yam (Dioscorea rotundata) Anthracnose (Colletotrichum gloeosporioides) in water yam (Dioscorea alata)
Pm1 Dn4 Sr2 Sr39 and Lr35 Yr17, Lr37, and Sr38 Ymv‐1 More than one dominant gene
Validated marker/QTL Major QTL for leaf rust (QLr.sfr‐7DS) and QLtn.sfr‐7DS for LTN located within the Xgwm1220‐Xgwm130 interval Whs350–1,2
References Schnurbusch et al., 2004
Xgwm337 QPm.vt‐1B, QPm.vt‐2A, and QPm.vt‐2B Xsts638‐7A, XE39M58‐77‐7A, and Xgwm344‐7A Xgwm106 and Xgwm337 gwm533120 Sr39F2/R3900 VPM1383, scar15550, and Xgwm636104
Mohler and Jahoor, 1996 Huang et al., 2000 Liu et al., 2001c Stepien et al., 2004 Arzani et al., 2004 Spielmeyer et al., 2003 Gold et al., 1999 Sharp et al., 2001
OPW850 and OPX850 OP171700 and OPE6950
Mignouna et al., 2002b Mignouna et al., 2002c
Abiotic stresses Barley Aluminum (Al) toxicity Frost tolerance
Alp Fr1
Maize Abscisic acid (ABA)
Bmag353 OPA17 and Psr637
Raman et al., 2001 Toth et al., 2004
Major QTL for leaf ABA
Landi et al., 2005
Rice Submergence tolerance
Sub1
RM219 and RM464A linked to Sub1
Xu et al., 2004b
Soybean Salt tolerance
Ncl
Sat_091 and Sat237
Lee et al., 2004
Wheat Boron (B) toxicity
Bo1, Bo2, and Bo3
Xpsr680‐7B and Xpsr160‐7D
JeVeries et al., 2000
S. L. DWIVEDI ET AL.
RWA [(D. noxia (Mordvilko)] Stem rust (Puccinia graminis) Stem rust and leaf rust Stem rust, leaf rust, and yellow rust
210
Table XI (continued )
Table XII Validation of Marker/QTL Associated with Agronomic and/or Seed Quality Traits in Barley, Pea, Rice, Soybean, and Wheat Trait
Two QTL on chromosome 3 Many QTL with small to large eVects
References
aABG396 and aCDO113 loci on chromosome 3 flanked by aABG057 and aABG37 A QTL on ‘‘plus’’ arm of chromosome 7(5H)
Larson et al., 1996
QTL1 and QTL6 on chromosome 3 and 6, respectively Xabg057, Bmy1, and XEBmac501
Romagosa et al., 1999
Spaner et al., 1999
Diastatic power (DP) aVecting malt quality Malt extract
25 chromosome regions
Two alleles each from chromosome 2H and 2 regions chromosome 5H
Collins et al., 2003
Pea Lodging
Two genes
A001 and A004
Warkentin et al., 2004
fgr
SCU015RM and RSP04 RZ474 and RZ575 sd‐1 linked with RG220 and RG109
Christopher et al., 2004 Kwon et al., 2001 Cho et al., 1994
cqProt‐001 and cqProt‐002 for seed protein; cqOil‐001, cqOil‐002, and cqOil‐003 for oil content; cqSd wt‐001 and cqSd wt‐002 for seed weight
Fasoula et al., 2004
Rice Fragrance Regeneration ability Semidwarf stature Soybean Seed weight, protein, and oil content
Nine QTL
Validated marker/QTL
sd‐1 to sd‐60 Many QTL
Coventry et al., 2003
211
(continued )
APPLIED CROP GENOMICS
Barley Agronomic traits (grain yield, plant height, maturity, and lodging severity)
Gene
212
Table XII (continued ) Trait Wheat Bread‐making quality (HMW glutenins)
Flour color Grain protein content (GPC) HMW glutenins Noodle quality
Seed dormancy Semidwarf Storage protein (Gliadines and glutenins)
Validated marker/QTL
Six genes at Glu‐1 loci on 1A, 1B, and 1D
A 15 bp in‐frame insertion in Glu‐B1–1d(B‐x6) discriminate genotypes with good or bad bread‐making quality Oligonucleotide primers: P1 and P2 (Dx2 and Dx5 alleles), P3 and P4 (Dy10 and Dy12 alleles), and P5 and P6 (Bx7 allele)
Schwarz et al., 2004
Xcdo34752 WMC41 and WMC415 Ax2 F2543, Ax2 R3605, Bx7F‐428, Bx7R693, Bx7F‐572, Bx7R693, Dx5F384, and DxR655 GBSS‐4A null mutation 440 bp from GBSS4A 42 SSRs Xhbe03 Rht‐B1b and Rht‐D1b
Sharp et al., 2001 Singh et al., 2001b Radovanovic and Cloutier, 2003 Zhao et al., 1998 Briney et al., 1998 Prasad et al., 2003 Torada et al., 2005 Ellis et al., 2002
PCR product of genotypes with LMW‐2 glutenin has 50‐bp longer fragment than those with LMW‐1 glutenin
D’Ovidio, 1993
HMW glutenin subunits 1Dx5 þ 1Dy10 linked with high dough strength/good bread while 1Dx2 þ 1Dy12 with poor bread quality QTL on chromosome 7A Six QTL Glu‐1 and Glu‐B1 locus GBSS locus null GBSS 4A allele 13 QTL Major QTL Rht‐B1b (Rht1) and Rht‐D1b (Rht2) Alleles in Gli‐B1 and Glu‐B3 locus associated with variation in HMW and LMW, respectively
References
Ahmad, 2000
S. L. DWIVEDI ET AL.
Doughs
Gene
APPLIED CROP GENOMICS
213
clearly shown to be highly dependent on sample sizes (n). When populations of less than 500 individuals are used for QTL mapping (irrespective of marker density), the power to detect true QTL is low and the estimated proportion of the genetic variance explained by mapped QTL is overestimated (see below), and it is very unlikely that QTL with small eVects will be identified.
Number of true QTL 10 10 40 40 10 10 40 40
h2
Sample size
0.30 0.30 0.30 0.30 0.95 0.95 0.95 0.95
100 500 100 500 100 500 100 500
Power (%) 9 57 3 11 39 94 6 46
Bias (s2g ) (%) þ559 þ144 þ2104 þ423 þ197 þ106 þ690 þ165
Bias in the estimated genetic variance occurs mainly due to sampling of small populations, where the true QTL that are not detected (most of them in small sample sizes) tend to enhance the apparent eVects of those QTL that are detected, through what is often referred to as the ‘‘Beavis eVect’’ (Beavis, 1998; Melchinger et al., 1998). Using a large population composed of 976 F5 maize testcross progenies evaluated in 19 environments, Scho¨n et al. (2004) also detected large eVect of sample size on the power of QTL detection as well as on the accuracy and precision of QTL detection. The number of detected QTL and the proportion of genotypic variance explained by QTL generally increased more with increasing population size than with increasing the number of test environments, although the average bias of QTL estimates and their range are reduced by increasing population size and by increasing the number of test environments. Cross‐validation performed well with respect to yielding asymptotically unbiased estimates of the genotypic variance explained by the QTL. However, by increasing the population size from 478 to 976, the increase in the proportion of genetic variance explained by QTL per additionally tested genotype is smaller as compared to increasing the population size from 244 to 488. This diminishing returns relationship (as the population size is increased) is expected due to the nonlinear relationship between sample size and power of QTL detection beyond a certain threshold (Lynch and Walsh, 1998). Genetic factors, such as enzyme variation in metabolic pathways, lead to an L‐shaped distribution of QTL eVects for a given quantitative trait (Bost et al., 2001). For example, this trend was reported for grain moisture in maize with the result that the distribution was skewed toward smaller values (L‐shaped) (Scho¨n et al., 2004).
214
S. L. DWIVEDI ET AL.
Care should also be taken to report QTL‐trait associations only at higher significance thresholds to avoid false identification of QTL when in fact a QTL is not present (Type I error). For example, Bernardo (2004) suggested that to prevent false QTL from confusing the literature and databases, a detected QTL should, in general, be reported as a QTL only if it is identified at a stringent significance level (Type I error probability or ac ¼ 0.0001). Increasing the size of the mapping population leads to both increased power (Beavis, 1998) and a lower rate of false‐positive QTL. However, the breeders in general like to work on many populations with small sample size rather than concentrating on few populations with large sample size. This trend needs to be reversed in order to exploit the QTL information in crop breeding programs or otherwise to deploy statistical methods for combining QTL analysis from related populations. Also more eVorts should be directed toward accurate evaluation of progenies (both at the genotypic and phenotypic level) in order to avoid application failures. Benjamini and Yekutieli (2005) suggested using a false discovery rate (FDR) estimate in QTL analysis. The FDR is the expected proportion of Type I errors. FDR‐controlling procedures ensure reproducible results with few false positives and oVering increased power of QTL discovery. The two advantages of the FDR approach, which make it particularly suitable for QTL analysis, are its flexibility regarding the amount of information in the data and its scalability. Controlling the FDR for multiple traits may result in no loss of power to detect QTL. However, a renewed optimism regarding QTL mapping has emerged based on analysis of cloned QTL, which indicates that the original low‐density map positions are relatively accurate (Price, 2006). Clearly, marker validation should be carried out after initial QTL mapping in order to determine whether fine‐mapping is required. When traits are controlled by multiple QTL of small eVect, the confidence intervals for their location are wide (Visscher et al., 1996). For these QTL, flanking markers may be widely spaced (>20 cM) and a large chromosomal fragment will be transferred during MAS. Thus, QTL of relatively large eVect are the most appropriate targets for MAS. These QTL are easier to validate, are more likely to be eVective in diVerent genetic backgrounds, and less likely to suVer from confounding linkage drag problems during MAS (Holland, 2004; Mackill, 2006). They are also easier to fine‐map, a process which requires accurate diVerentiation between the phenotypes resulting from the two alleles of the QTL. QTL of large eVect may also be readily detected even in populations of smaller size (Vales et al., 2005). The genetic background of parental genotypes of the mapping population has a profound eVect on the number, location, and eVects of the identified QTL. For example, if a QTL allele with beneficial eVect is identified in population A, its introgression by means of MAS in population B will not necessarily lead to tangible benefits. This is because population B may
APPLIED CROP GENOMICS
215
already have alleles of similar or even greater value at this QTL and/or because of diVerent interactions between the QTL and the two genetic backgrounds. Campos et al. (2004) estimated that most drought‐tolerant QTL detected in maize would have limited utility for applied breeding, partially due to the prevalence of genetic background and environment eVects. Use of MAS for transferring QTL is more suitable when a trait is being introduced from an exotic source into elite germplasm, thus ensuring higher levels of polymorphism and higher probability of expression of the gene/QTL in the new genetic background (i.e., more likely that the allele is diVerent to the recipient). A mapping study involving an exotic donor crossed with an elite line lacking the trait will increase the chance that the identified markers will be useful in the targeted cultivars. Large‐eVect QTL are also more likely to be expressed in diVerent genetic backgrounds. For traits controlled by smaller QTL, the eVect of the background can be extreme. However, it is currently impossible to predict these interaction eVects in most crops, thus field evaluation must be used to validate the expression of introgressed QTL. Epistasis, as detected by identification of diVerent QTL when the same donor is crossed to diVerent parents, is often observed. In Arabidopsis, significant eVects of epistasis were observed for two QTL found in a 210‐kb interval controlling growth rate, with gene eVects depending on genetic background (Kroymann and Mitchell‐Olds, 2005). Li et al. (2006a) provided an example of complex interactions among QTL for partial resistance to bacterial blight (BB) in rice, and it is suggested that this results from genetic networks of the underlying genes. Clearly, even for QTL that are observed in multiple populations, their robustness for applications in breeding must still be validated in relevant populations. Development of reciprocal introgression lines is useful for estimating the eVects of the genetic background. For many traits, the overlap of QTL detected in reciprocal genetic backgrounds is low, showing the large eVect of background on trait expression. QTL E eVects are another factor that must be considered during validation studies. There are many reports of the lack of consistency between QTL detected in diVerent environments. For examples, when Paterson et al. (1991) evaluated F2 and F2:3 progenies in 3 environments, they detected 29 putative QTL distributed over 11 of the 12 chromosomes, accounting for 4.7–42% of the phenotypic variation for fruit size, soluble solids concentration, and pH in tomato. Of these, 4 were detected in all the 3 environments, 10 in 2 environments, and 15 only in a single environment. QTL mapping using the same rice population for analysis of seedling vigor revealed major diVerences for QTL detected at diVerent temperature regimes (Redon˜a and Mackill, 1996). Experiments conducted with the same mapping population in nine environments showed that rice QTL detection for plant height and heading date was markedly aVected by environment (Li et al., 2003a). Drought stress at flowering adversely aVects grain yield in maize that causes a delay in silking, an increase in anthesis silking interval (ASI), thus decrease in grain yield. Vargas
216
S. L. DWIVEDI ET AL.
et al. (2006) identified QTL for ASI that are stable across the eight environments and corresponded well with those reported by Ribaut et al. (1996). For grain yield, Vargas et al. (2006) detected a much larger GEI than for ASI; however, a couple of QTL consistent across environments identified, thus confirming the previous report of the QTL for grain yield and yield components on chromosomes 1 and 10 (Ribaut et al., 1997a). Cross‐validation of QTL in independent samples and in diVerent genetic backgrounds and environments is necessary to obtain unbiased estimates of QTL eVects and the proportion of the genetic variance explained by the detected marker‐QTL association before using them in MAS breeding programs. In general, QTL detected in multiple mapping studies using diVerent populations would be considered as the most important targets for MAS application. For example, a grain length and weight QTL near the centromere of rice chromosome 3 was identified in at least eight independent mapping studies and has been identified as a putative transmembrane protein (Fan et al., 2006). In some cases, mapping the QTL in multiple generations from the same cross can be used to confirm the presence of QTL, as was observed for sheath blight in a rice RIL population (Pinson et al., 2005). Similarly, an advanced backcross population (BC2F6:8) validated all QTL for resistance to Septoria speckled leaf blotch of barley that had been identified in an RIL population with the same parents (Yun et al., 2006). QTL detected in a rice RIL population were validated in NIL developed for the two major plant‐type QTL (Kobayashi et al., 2006). However, usually it is only the successful validations that are reported in the literature. A rare exception to this is Steele et al. (2006) who attempted to validate four root QTL during the three backcrosses aimed at transferring root QTL from the upland rice cultivar Azucena into the variety Kalinga III. While all four root QTL were successfully introduced, only one showed a significant eVect when transferred into the Kalinga III background. Where recurrent selection is used in breeding programs, QTL eVects can change over time in subsequent selection cycles. This led to the development of the ‘‘Mapping As You Go’’ (MAYG) approach (Podlich et al., 2004), where QTL eVects are estimated in each cycle before selection and intermating are performed. Fine‐mapping of QTL is very useful for identifying tightly linked markers that will not suVer from loss of linkage due to recombination between marker and QTL during applications in diVerent breeding populations. This will also serve to minimize the size of the introgressed fragment during backcrossing. Few QTL with major eVects on traits of agricultural importance have been fine‐mapped and successfully delimited their position on the chromosome in tomato, rice, wheat, and maize (see Section VII.B).
APPLIED CROP GENOMICS
217
IV. SUCCESSFUL APPLICATIONS OF MARKER‐ASSISTED GENETIC ENHANCEMENT IN PUBLIC SECTOR BREEDING PROGRAMS MAS is most useful for traits where phenotypic evaluation is expensive or diYcult, particularly for those polygenic traits with low heritability that are highly aVected by the environment. It is also useful to break linkages between the target traits and undesirable genes in so‐called marker‐accelerated backcross breeding. MAS may also oVer the opportunity to address goals not possible through conventional breeding, such as pyramiding diVerent sources of disease resistance that have similar phenotypes. Indirect selection based on marker genotype rather than phenotype can be used to accelerate the speed and increase the precision of genetic progress, reduce the number of generations, and when integrated into optimized molecular breeding strategies, it can also lower the costs of selection. The eYciency of MAS depends on many factors associated with how the underlying marker‐trait associations were identified, including the size of the mapping population, the nature of the phenotyping, the design and analysis of the experiment, the number of markers used, the distance between marker loci, the genomic region containing the desired QTL, and the proportion of additive genetic variance explained by the marker, the selection method, and the experimental design (Dwivedi et al., 2005 and references therein). The eYciency of MAS also depends on many factors associated with its application, including the crop and breeding system, the molecular breeding process, and the nature of the genotyping pipeline. In this section, we briefly summarize the cases where MAS has been used to incorporate beneficial traits into improved genetic backgrounds of major food crops.
A. RESISTANCE TO BIOTIC STRESSES 1.
Single Gene Introgression
a. Cereals. MAS coupled with backcross and pedigree breeding methods and field evaluation has led reports in the literature of genetic enhancement for resistance to bacterial blight (BB) (Xa21), gall midge (Gm‐6t), and brown plant hopper (BPH) (Bph1 and Bph2) in rice; to leaf rust (Lr19, Lr51, and Yr15 ) in wheat; to yellow dwarf virus (Yd2), stripe rust (Yr4 ), and powdery mildew (mlo‐9) in barley; and to downy mildew (major QTL) in pearl millet (Table XIII). The progenies showed same resistance level as the donor parental lines both in greenhouse and field evaluations.
Gene
Breeding scheme
Marker
Marker‐assisted product
218
Table XIII Examples of Single Gene Transfer for Resistance to Biotic Stresses Using Marker‐Assisted Selection in Barley, Common Bean, Maize, Pearl Millet, Potato, Rice, Soybean, and Wheat References
Barley Barley yellow dwarf virus Yd2 Two backcrosses
YLM
Stripe rust (P. striiformis f. sp. hordei) Yr4 Double‐haploids from BC1F1
RFLPs
JeVeries et al., 2003
DHLs carrying mlo9mlo9 completely resistant to powdery mildew
Paris et al., 2003
DHLs carrying Yr4 less susceptible to stripe rust
Toojinda et al., 1998
Marker‐based selected RILs resistant to CBB
Yu et al., 2000
Progenies with improved resistance to SWCB leaf feeding damage selected
Willcox et al., 2002
HHB 67‐2 with improved downy mildew resistance
Hash, 2005
Common bean Comman bacterial blight (CBB) [Xanthomonas campestris pv. phaseoli (Xcp)] Quantitative Pedigree breeding BC420900 and C7900 Maize Southwestern corn borer (SWCB) (Diatraea grandiosella Dyar) 6–9 QTL Two backcrosses 89 RFLPs and a morphological marker, grain color (y1) Pearl millet Downey mildew (Sclerospora graminicola) Major gene Backcross breeding
Xpsm464, Xpsm716, Xpsm265, and Xpsm416
S. L. DWIVEDI ET AL.
Powdery mildew [Blumeria graminis f. sp. hordei (Bgh.)] mlo9 Double‐haploid breeding SNPs
Lines with Yd2 had few leaf symptoms but no adverse eVect on agronomic traits
Potato Late blight [P. infestans (Mont.) de Bary] RB Two backcrosses
RGA1/rga1, RGA2/rga2, RGA3/rga3, and RGA4/rga4
Several marker‐positive breeding lines showed resistance to late blight
Colton et al., 2006
Lines with high yield and BB resistance selected 6078(Xa21) performed well under heavy disease pressure
Chen et al., 2000
Gm‐6t successfully transferred to hybrid rice parents
Katiyar and Bennett, 2001
MAS‐selected lines comparable to phenotypic selection
Concibido et al., 1996
Families with Ep‐D1c allele resistant to leaf rust Lr51 transferred into three cultivars Yr15 transferred into Zak
Slikova et al., 2003
Rice
Gall midge (Orseolia oryzae) Gm‐6t Pedigree breeding
RAPD and STS
Chen et al., 2001
Soybean Soybean cyst nematode (H. glycines Ichinohe) Quantitative Pedigree breeding
98 RFLPs Wheat
Leaf rust (Puccinia triticina) Lr19 Pedigree breeding
Ep‐D1c
Lr51
Six backcrosses
XAga7 and Xmwg710
Yr15
Two backcrosses
1000 SSRs
APPLIED CROP GENOMICS
Bacterial blight (BB) [Xanthomonas oryzae pv. Oryzae (Xoo)] Xa21 Three backcrosses PCR‐based markers close to Xa21, and 128 RFLPs Xa21 Three backcrosses 21, C189, and AB9 for foreground and AFLPs for background selections
Helguera et al., 2005 http://wheatlifemagzine. com/0105/pg68_0105.pdf
219
220
S. L. DWIVEDI ET AL.
b. Legumes. In contrast to the cereals, there are very few reports in the literature of success stories for single gene transfer by MAS in legumes, and only in two crops: soybean and common bean. However, this is proportional to the relative stage of development of genomics in these crops and the number of trait mapping studies that has been completed. Loci for resistance to common bacterial blight in common bean and cyst nematode in soybean have been transferred into improved breeding lines using MAS (Table XIII). c. Roots and Tubers. Late blight is the most devastating disease in potato and has received much research attention across the world (Ojiambo et al., 2000). However, resistance breeding has been a challenge because of the short period during which race‐specific resistance genes remain eVective, while breeding for ‘‘horizontal’’ or race‐nonspecific resistance has achieved only moderate successes. Solanum bulbocastanum (2n ¼ 24), a diploid species native to Mexico, has been characterized as possessing durable resistance to all known races of late blight (van Soest et al., 1984), and mapped to a single locus on chromosome 8 (Naess et al., 2000). Using PCR‐ based DNA markers for tracking the RB gene in breeding populations, several marker‐positive selected lines showed resistance to late blight (Table XIII). RB has also been cloned and transformed into Katahdin, a highly susceptible potato cultivar. The Katahdin‐transformed plants with RB showed broad‐spectrum resistance against a wide range of late blight isolates (Lozoya‐Saldana et al., 2005; Song et al., 2003). Clearly, by having the full sequence of the target gene, it should be possible to develop a highly eYcient low‐cost assay system for this trait. 2.
Gene Pyramiding
Gene pyramiding is a useful approach to the durability or level of pest and disease resistances, or to increase the level of abiotic stress tolerance. Genes controlling resistance to diVerent races or biotypes of a pest or pathogen and genes contributing to agronomic or seed quality traits can be pyramided together to maximize the benefit of MAS through simultaneous improvement of several traits in an improved genetic background. a. Cereals. Many major genes (recessive or dominant) and QTL conferring resistance to pests and diseases have reported in major cereals. Using MAS coupled with field evaluation, researchers were able to combine multiple resistances to these pests and diseases in many cereal crops. Successful examples include improved pyramided lines and cultivars containing gene combinations for bacterial blight (BB) (xa3, xa4, xa5, xa7, Xa10, xa13, Xa21, and Om); blast (Bl) (Pi1, Piz‐5, and Pita); brown plant hopper (BPH) (Bph1 and Bph2);
APPLIED CROP GENOMICS
221
Bl (Piz‐5) and BB (Xa21); BB (Xa21) and yellow stem borer (YSB) (Bt); BB (Xa21), YSB (Bt), and sheath blight (ShB) (RC7 chitinase); and BB (Xa21 and Xa7), YSB (Bt), Bl (Pi1, Pi2, and Pi3), and BPH (Qbph1 and QBph2) in rice (Table XIV). In wheat, powdery mildew (Pm2, Pm4a, Pm6, Pm8, and Pm21) pyramided lines and those with resistance to Fusarium head blight (FHB) (six QTL), orange blossom midge (Sm1), and leaf rust (Lr21) were bred through MAS. Resistance to Barley mild mosaic virus (BaMMV) and Barley yellow mosaic virus (BaYMV and BaYMV‐2) complex (rym4, rym5, rym9, and rym11) and stripe rust (QTL: 1H, 4H, and 5H or their combination: 1H and 4H, 1H and 5H, 4H and 5H, or 1H, 4H, and 5H) has been separately incorporated through MAS in barley. Many of these pyramided lines showed enhanced resistance to pests and diseases, some even outyielded the controls under high disease or pest pressure in field conditions (Table XIV). b. Legumes. Reports of gene pyramiding in legumes include combining QTL for resistance to corn earworm and Pseudoplusia includens (soybean looper) with cry1Ac resistance in soybean; while resistances to rust and anthracnose (QTL) or to CBB, Bean common mosaic virus (BCMV), and anthracnose have been combined in common bean (Table XIV). The pyramided lines in soybean showed improved resistance to defoliators, while common bean lines showed multiple resistances to these diseases. c. Roots and Tubers. A single dominant gene for extreme resistance to Potato virus Y (PVY, genus Potyvirus), Ryadg, was mapped to a distal position on potato chromosome 11 (Ha¨ma¨la¨inen et al., 1997). For Potato virus X (PVX, genus Potexvirus), dominant genes, Rx1 and Rx2, were mapped to potato chromosomes 12 and 5, respectively (Ritter et al., 1991). The dominant gene Gro1 for resistance to all known pathotypes of the root cyst nematode (Globodera rostochiensis) was mapped to potato chromosome 7 (Barone et al., 1990). A single dominant gene Sen1 for resistance to potato wart (Synchytrium endobioticum) pathotype 1 was mapped to a similar position on potato chromosome 11 as the Ryadg (Hehl et al., 1999). Using four PCR‐based diagnostic assays, tetraploid progeny from tetraploid–diploid crosses combining the Ryadg for extreme resistance to PVY with Gro1 for nematode resistance and with Rx1 for extreme resistance to PVX, or with Sen1 for wart resistance were selected (Table XIV).
B. TOLERANCE TO ABIOTIC STRESSES 1.
Drought Tolerance
Rice: selection for a well‐developed root system with long thick roots should improve the drought tolerance of upland rice because the plant would avoid water stress by absorbing water stored in the deep soil layers
222
Table XIV Examples of Gene Pyramiding for Resistance to Biotic Stresses Using MAS in Barley, Common Bean, Potato, Rice, Soybean, and Wheat Gene
Breeding scheme
Marker
Marker‐assisted product
References
Barley BaYMV‐I, BaYMV‐II, and BaYMV‐III; BaMMV‐Ka1 and Na1 rym1 One backcross RFLPs
Barley stripe rust QTL (1H, 4H, and 5H)
Simple and complex crosses using double‐haploids
RAPDs and SSRs
Backcross‐derived ILs
SSRs
Okada et al., 2003b
Werner et al., 2005
ILs in susceptible genetic background carrying 1H, 4H, or 5H individually or in combinations were resistant to barley stripe rust
Richardson et al., 2006
Marker‐based selected progenies resistant to CBB, BCM, and anthracnose
http://www. Ontariobeans.on.ca/ liu5thcapsulem sapaperfinal.pdf
Lines combining resistance to rust and anthracnose developed
Faleiro et al., 2004
Common bean Common bacterial blight (CBB), BCMV, and anthracnose Several loci for BCMV and Complex crossing and UBC420, BC73, SW13‐I, anthracnose pedigree breeding and Co‐42
Rust (Uromyces appendiculatus) and anthracnose (Colletotrichum lindemuthianum) Nine major genes each for Three backcrosses RAPDs rust and anthracnose
S. L. DWIVEDI ET AL.
rym4, rym5, rym9, and rym11
Mokkei 01530 with rym1 resistant to BaYMV‐1 and BaYMV‐II, and similar in malt quality as of Haruna Nijo DHLs carrying rym4, rym9, and rym11 and those with rym5, rym9, and rym11 selected
Potato Potato virus Y (PVY), Potato virus X (PVX), nematode, and wart (S. endobioticum) F1 hybrids (2 4 cross) RYSC3 (Ryadg), Gro1–4 Ryadg(PVY), Rx1 (PVX), Gro1 (nematode), and (Gro1), CP60 (Rx1), and Sen1 (wart) N125 (Sen1)
Marker‐based selection of tetraploid potato clones showed multiple resistance to four diseases, all with monogenic resistance
Gebhardt et al., 2006
Lines carrying multiple genes provided broader spectra of resistance to BB Pyramided lines showed broader spectrum of resistance to BB
Yoshimura et al., 1995
Lines with Xa21 had increased resistance than xa5, xa13, or both Lines with gene combinations provided broader spectrum of resistance to BB Lines carrying multiple genes showed greater resistance than those with single gene(s) Angke (xa5) and Conder (xa7) released, and few other lines combining yield and resistance in advance trials in Indonesia
Sanchez et al., 2000
Rice Pedigree breeding
RZ390, RG556, RG207, XNpb181, and Oo72000
Xa4, xa5, xa13, and Xa21
Pedigree breeding
xa5, xa13, and Xa21
Three backcrosses
Npb181, Npb78, RG103, RG136, RG556, RZ28, RZ207, pTA248, and pTA818 RG556, RG207, RG136, and pTA248 RG556, RG136, and pTA248
Two backcrosses
xa5, xa13, and Xa21
Pedigree breeding
pTA248, RG136, and RM122
xa5, xa7, Xa21, and Om
Three backcrosses
RG556a (xa5), OPL13 (Om), pTAta258 (Xa21), and 10 RAPD markers
Huang et al., 1997
Singh et al., 2001a
Swamy et al., 2004
APPLIED CROP GENOMICS
Bacterial blight (BB) Xa3, Xa4, xa5, and Xa10
http://www.isuagcenter. com/inst/research/ stations/rice/ proceedings.pdf (continued )
223
224
Table XIV (continued ) Gene
Breeding scheme
Marker
BB, leaf folders, yellow stem borer (YSB) (Scirpophaga incertulas) Xa21 and Bt Pedigree breeding 21, 248, C189, AB9 for Xa21 and pFHBT1(1.8 kb) for Bt
Marker‐assisted product
References
Minghui63 containing Bt and Xa21 and its hybrids showed multiple resistance and produced two to three times more grain yield under natural infestation
Jiang et al., 2004
AFLP 1415, STS P3, M5, 248, RM144, RM224, and Pi2
Minghui 63(Xa21 and Xa7) showed broader resistance to BB; Minghui 63(Xa21 and Bt) showed combined resistance to BB and SB; Zhenshan97(Qbph1 and Qbph2) showed better resistance to BPH
Yuqing et al., 2004
BB, YSB, sheath blight (ShB) (R. solani) Xa21, Bt, and RC7 chitinase Pedigree breeding (Shb)
Pc822 (Xa21), Bt, and RC7 chitinase
Lines carrying three genes were resistant to BB, YSB, and ShB
Datta et al., 2002
The pyramided lines showed better resistance to blast
Hittalmani et al., 2000
Lines showed combined resistance to Bl and BB The pyramids showed enhanced resistance to blast and BB
Narayanan et al., 2002
Blast (Bl) [Magnaporthae grisea (Herbert) Borr. (ananmorphe Pyricularia oryza Cav.) Pi1, Piz‐5, and Pita Pedigree breeding Npb181, RZ536, RZ64, RZ612, RG456, RG64‐ SAP, RG869, RZ397, and RG241 Bl and BB Piz‐5 and Xa21 Piz‐1and Piz‐5 (blast) and Xa21 (BB)
Four backcrosses (Piz‐5) and transgenic (Xa21) Pedigree breeding
RG64750 (Piz‐5) and 1.4‐kb fragment (Xa21) RZ536 and r10 (blast) and Xa21 (1.4‐kb fragment of pC822)
Narayanan et al., 2004
S. L. DWIVEDI ET AL.
BB, stem borer (SB), blast, and BPH Xa21 and Xa7 (BB); Bt (SB); Pedigree breeding Pi1, Pi2, Pi3 (blast); and Qbph1 and Qbph2 (BPH)
Brown plant hopper (BPH) (Nilaparvata lugens Stal) (Bph1 and Bph2) Several major genes and Pedigree breeding em24G, EM5814N, em32G, QTL KPM1, KPM2, KPM3, KPM4, KPM5, and KPM8 Rice yellow mottle virus (RYMV) Many QTL Three backcrosses
RG869 and BNL 16–06 for foreground and RFLPs and SSRs for background selections
Pyramided lines showed similar resistance as to those with single gene
Sharma et al., 2004
Lines containing QTL 12 and QTL 7 alleles showed partial resistance to RYMV
Ahmadi et al., 2001
Nine SSRs
The pyramid lines had a detrimental eVect on larval weights and on defoliation by CEB
Walker et al., 2002
CEW and soybean looper (SBL) (P. includens) cry1Ac and QTL (PI Two backcrosses 229358)
Six SSRs and sequence‐ specific primers cry1Ac
Lines carrying cry1Ac and QTL alleles resistant to three lepidopteran pests
Walker et al., 2004
Wheat Fusarium head blight (FHB) (F. graminearum), orange blossom midge (Sitodiplosis mosellana), and leaf rust (Lr21) Two backcrosses gwm533, gwm493, and Resistant progenies containing Six FHB QTL, Sm1 for wmc808 chromosome segments FHB, Sm1 midge and Lr21 for leaf and Lr21 identified rust Powdery mildew (E. graminis DC. F. tritici Em. Marchal) Pm2, Pm4a, and Pm21 Pedigree breeding Xbcd1871‐5D‐EcoRV, Xwhs350‐5D‐EcoRV, Xbcd1231‐2A‐EcoRI, pHv62, and psr113 Pedigree breeding
RAPD and SCAR markers
Liu et al., 2000b
Wang et al., 2001a
225
Pm2, Pm4a, Pm6, Pm8, and Pm21
Gene combinations (Pm2 þ Pm4a, Pm2 þ Pm21, and Pm4a þ Pm21) integrated into Yang158 that showed resistance to powdery mildew Lines with Pm2 and Pm4a immune to powdery mildew
Somers et al., 2005
APPLIED CROP GENOMICS
Soybean Corn earworm (CEW) (Helicoverpa zea Boddie) QTL and Bt (cry1Ac) Three backcrosses
226
S. L. DWIVEDI ET AL.
(Yoshida and Hasegawa, 1982). However, phenotypic selection for root morphological traits in conventional breeding is not feasible. The tropical japonica rice cultivars are reported to have thicker and deeper roots than indica cultivars (Courtois et al., 1996). Using four QTL (QTL2, QTL7, QTL9, and QTL11) from Azucena (a japonica cultivar), which each contributing between 5% and 30% phenotypic variance for root traits (root length and thickness), Steele et al. (2006) initiated marker‐assisted backcrossing (MABC) to improve drought tolerance into Kalinga III, an upland indica cultivar. After five backcrosses and conducting over 3000 marker assays (2548 RFLPs and 700 SSRs) on 323 plants, the NILs were developed and evaluated for root traits. The target segment on chromosome 9 (RM242‐RM201) significantly increased root length under both irrigated and drought stress environments. Azucena alleles at the locus RM248 (below the target root QTL on chromosome 7) delayed flowering. However, selection for the recurrent parent allele at this locus produced early flowering NILs that are suited to upland environments in eastern India. Other target regions had no significant eVects on root length in Kalinga III genetic background. In a similar study, Shen et al. (2001) also demonstrated the eVectiveness of MAS to transfer QTL from three of the four target regions (chromosomes 1, 2, 7, and 9) associated with root traits (root length and root mass) from Azucena to NIL in IR64 genetic background. NIL carrying the QTL from chromosomes 1, 7, and 9 had shown significantly improved root traits over IR64, while none of the NIL containing QTL from chromosome 2 had root phenotype significantly diVerent from that of IR64. In both the studies, progenies containing QTL from chromosome 7 confer improved root characteristics that are now being tested under field conditions to assess their performance under water‐limited conditions. Maize: Anthesis silking interval (ASI) is an important trait associated with drought tolerance in maize. Ribaut et al. (1996, 1997b) initiated a major marker‐assisted breeding program to transfer five genomic regions involved in the expression of a short ASI from Ac7643 (a drought‐tolerant line) to CML247 (an elite tropical breeding line). Five genomic regions were transferred using flanking PCR‐based markers. Seventy of the best BC2F3 (i.e., S2 lines) lines were crossed with two testers, CML254 and CML 274. These hybrids and the BC2F4 families derived from selected BC2F3 plants were evaluated for 3 years under drought stress conditions. Results show that stress conditions induced a yield reduction of at least 80%, but the mean of the 70 selected genotypes performed better than the control (all evaluated as testcross products). In addition, the best genotypes among 70 selected (BC2F3 testers) performed two to four times better than the control. However, this diVerence became less marked when the intensity of stress decreased: for a stress inducing less than 40% yield reduction, performance of testcross hybrids resulting from MAS was no better than the ‘‘original’’ version of CML274.
APPLIED CROP GENOMICS
227
Pearl millet: a major QTL on LG2 is associated with increased grain yield and harvest index under terminal stress in PRLT 2/89‐33 (Yadav et al., 2002). PRLT 2/89‐33 is a drought‐tolerant, low‐tillering, and large‐panicle landrace from West Africa (Andrews and Anand Kumar, 1996). In contrast, H77/833‐2 is a drought‐sensitive, high‐tillering, and small‐panicle landrace from India (Kapoor et al., 1989). The performance of QTL MAS‐derived topcross hybrids (TCH) was compared with that of field‐based TCH. Progenies with the best overall ability to maintain under terminal stress environments were used to generate the TCH, and these were compared with randomly mated TCH made from randomly selected progenies from the entire population (irrespective of performance under terminal drought stress). In both the cases, progenies were selected irrespecitve of the presence or absence of favorable alleles at the putative drought‐tolerant QTL and evaluated across 21 environments (nonstress, terminal stress, and gradient stress). The QTL MAS‐derived hybrids were significantly, but only modestly, higher yielding both in full and partial terminal stress environments. However, this advantage under stress was at the cost of lower yield of the same hybrids under nonstressed environments. The QTL MAS‐derived hybrids flowered earlier and had limited eVective basal tillers, low biomass, and high harvest index. All these traits are similar to that of the drought‐tolerant parent PRLT‐2/ 89‐33, thus confirming the eVectiveness of the putative drought‐tolerant QTL on LG2 (Bidinger et al., 2005). A number of marker‐assisted backcross progenies have been generated from the cross between H77/833‐2 (drought sensitive) and PRLT 2/89‐33 (LG2 drought‐tolerant QTL). Initial results indicate that it has been possible to improve grain yield under terminal stress in these lines without a biomass penalty under stress conditions or a grain yield penalty under well‐watered conditions (Hash et al., 2004). Common bean: Schneider et al. (1997) identified four to five RAPD markers in two mapping populations that were consistently and significantly associated with yield under stress, yield under optimum irrigation, and geometric mean yield across a broad range of environments. To examine the eVectiveness of these markers, they selected genotypes from either extremes and evaluated them in three locations. MAS in the Sierra/AC1028 population was eVective in Michigan under severe stress but ineVective in Mexico under moderate stress. The Sierra/Lef‐2RB population showed improved performance by 11% in stress and 8% in nonstress environments.
2.
Submergence Tolerance
In many parts of the lowlands of south, southeast, and eastern Asia, rice cropping during the rainy season is completely submerged for varying periods of time, resulting in substantial losses to rice production in these regions.
228
S. L. DWIVEDI ET AL.
Genetic variation for submergence tolerance has been reported in rice, for example, FR13A, a landrace from India, can survive up to 2 weeks of complete submergence owing to a major QTL, submergence 1 (Sub1) on chromosome 9 (Xu and Mackill, 1996; Xu et al., 2000). Further, Xu et al. (2006) identified a cluster of three genes related to the ethylene‐response‐factor (ERF) at the Sub1 locus. A variant of Sub1A‐1 is found only in submergence‐tolerant rice, FR13A. Overexpression of Sub1A‐1 in submergence‐intolerant O. sativa ssp. japonica (cultivar Liaogeng) conferred enhanced tolerance. The same research group used marker‐assisted backcross breeding to introgress the Sub1A‐1 gene into a widely grown Indian cultivar, Swarna. The introgressed progenies showed strong submergence tolerance and maintained high yield and other agronomic properties of the recurrent parent, Swarna. Submergence tolerance has also been introduced into a Thai Jasmine rice, KDML105 following marker‐assisted breeding (Siangliw et al., 2003).
C. AGRONOMIC AND SEED QUALITY TRAITS Many agronomic or seed quality traits are conferred by QTL each with varying contributions and diVerent interaction with each other (epistasis) and the environment thus greatly complicating cultivar development. Unlike many success stories of pests and disease‐resistance transfer by MAS in many crops, there are few reports of successful transfer of beneficial alleles associated with improved yield or seed quality traits into improved genetic background. The foremost among them include yield‐enhancing QTL alleles from wild relatives of rice and soybean and grain quality in rice, wheat, and maize, and malt quality in barley. Rice: Using marker‐assisted backcross breeding, the two yield‐enhancing QTL alleles, yld1.1 and yld2.1 from wild rice Oryza rufipogon, have been successfully transferred into an improved agronomic background, whose progenies out‐yield the controls by 24–42%. Most of this improvement was accounted for by increases in two yield components: grains per panicle and 1000‐grain weight (Liang et al., 2004). In another marker‐assisted backcross breeding program, Yue‐guang et al. (2004) selected progenies in BC3 generation that produced more than 30% greater grain yield over Minghui 63, a restorer line of the many commercially grown hybrids in China. Grain quality represents a major problem, particularly in hybrid rice which are now commercially grown in substantial acreage worldwide. The most serious grain quality problems in hybrid rice are eating and cooking qualities, and to some extent milling quality. Both eating and cooking qualities are largely determined by three characters, specific to the physical and chemical properties of the starch in the endosperm, that is,
APPLIED CROP GENOMICS
229
amylose content (AC) (Juliano, 1985; Webb, 1980), gel consistency (GC) (Cagampang et al., 1973), and gelatinization temperature (GT) (Little et al., 1958). The chalkiness, or opacity, of the endosperm of the grains is another important grain quality trait that not only aVects the appearance of the grains but also the resistance to grain breakage during milling. Medium AC/soft GC/high GT together with a translucent endosperm represent good grain quality, while high AC/hard GC/low GT together with chalky endosperm represent poor grain quality (Tan et al., 1999, 2000). Shanyou 63, a hybrid between the male‐sterile line Zhenshan 97A and the restorer line Minghui 63, was the most widely grown hybrid rice in the 1990s, accounting for 25% of the rice production in China (Lin and Min, 1991). However, in recent years, the area declined as this hybrid became susceptible to bacterial blight and because of greater consumer awareness about its relatively poor cooking and eating qualities. AC, GC, and GT cosegregate and are controlled by the waxy locus and other genes tightly linked to this locus (Tan et al., 1999). It should be, therefore, possible to simultaneously improve all three traits. Chalkiness, or opacity, of the grains is controlled by 6 QTL located on 5 of the 12 rice chromosomes (Tan et al., 2000). Using MAS in three generations of backcrossing followed by one generation of selfing, Zhou et al. (2003a) successfully introduced the wx‐MH fragment from Minghui 63 into Zhenshan 97B, which was subsequently transferred to Zhenshan 97A. The improved version of the male‐sterile and maintainer lines, Zhenshan 97A (wx‐MH) and Zhenshan 97B (wx‐MH), contained a fragment less than 6.1 cM in length around the waxy gene region from the donor parent, with the rest of the genome being from the original Zhenshan 97. The introduction of this fragment has greatly improved the cooking and eating quality of inbred lines and their resultant hybrids, with the agronomic performance essentially the same as the original maintainer line and resultant hybrid. Additionally, the selected lines and their hybrids showed reduction in opacity (a change that is highly preferred from consumer’s view point) and grain weight. However, the hybrids yielded at a similar level to the original hybrid (Shanyou 63), presumably because of phenotypic plasticity as a result of strong heterosis (Zhang et al., 1994). Long‐te‐fu (LTF) and Zhan‐shan 97 (ZS) are the two key female parents widely used for the generation of indica hybrid rice in China. However, both have poor cooking and eating qualities because of high AC. Liu et al. (2006) used MAS to introgress Wx‐T allele (conferring intermediate AC and thus good quality) into the maintainer (LTF‐B and ZS‐B) and their relevant male‐sterile lines (LTF‐A and ZS‐A) to generate improved indica hybrids. The resulting maintainer lines (LTF(tt)‐B and ZS(tt)‐B) and hybrids showed improved cooking and eating qualities with no significant alterations in their agronomic traits. Rice with low glutelin content is suitable for patients aVected by diabetes and kidney failure. The Lgc‐1 locus confers low glutelin in the rice grain,
230
S. L. DWIVEDI ET AL.
located on chromosome 2 between flanking markers (Miyahara, 1999). This trait has been successfully incorporated into japonica rice with 93–97% selection eYciency using SSR2‐004 and RM358 markers (Wang et al., 2005a). Additionally, grain quality traits such as 1000‐seed weight, kernel length/breadth ratio, basmati type aroma, and high AC have been combined with resistance to bacterial blight using marker‐assisted backcross breeding (Joseph et al., 2003; Ramalingam et al., 2002). Wheat: the major grain quality traits in wheat are protein content and composition and grain color that influence bread‐ and noodle‐making qualities. Gliadins and glutenins determine physical quality of wheat flour dough (Payne, 1987). Dough with high elasticity and reasonable extensibility is ideal for bread making, while highly extensible dough is good for making biscuits, and dough with intermediate properties is good for flat bread or noodles. Most of these quality traits are genetically highly complex, conferred by many genes showing considerable GEI. Moreover, evaluation of these traits requires well‐developed laboratory procedures and equipments and a large sample size for evaluation. These factors force most wheat breeders to only evaluate quality traits in advanced generations of their breeding programs. Thus, it is surprising that although for many of these traits markers have been identified and validated (see Section III), their use in breeding has been limited. Exploiting allelic variation at the Glu‐1 (endosperm storage protein subunit) locus to improve bread‐making quality has been one of the early examples in which markers were used to improve wheat quality traits (de Bustos et al., 2001; Koebner, 2003). Sun et al. (2005) used a novel STS marker for improving polyphenol oxidase (PPO) activity in bread wheat. Breeding wheat cultivars with low PPO activity is the best approach to reduce undesirable darkening of bread wheat‐based end‐products, particularly for Asian noodles. Based on the sequences of genes conditioning PPO activity during kernel development, 28 pairs of primers were developed. One of these markers designated as PPO18, mapped to chromosome 2AL, can amplify a 685 and an 876‐bp fragment in the cultivars with high‐ and low‐PPO activity, respectively. QTL analysis indicated that the PPO gene cosegregated with the STS marker PPO18 and is closely linked to Xgwm312 and Xgwm294 on chromosome 2AL, explaining 28–43% of phenotypic variance for PPO activity across three environments. A total of 233 Chinese wheat cultivars and advanced lines were used to validate the correlation between the polymorphic fragments of PPO18 and grain PPO activity. The results showed that PPO18 is a codominant, eYcient, and reliable molecular marker for PPO activity and can be used in wheat breeding programs targeting noodle quality improvement. Maize: maize plays a very important role in human and animal nutrition. The endosperm of the maize seed has several distinct regions that have diVerent physical properties. The aleurone is the outer layer of the
APPLIED CROP GENOMICS
231
endosperm, composed of specialized cells that secrete hydrolytic enzymes during germination. Beneath the aleurone are starchy endosperm cells filled with starch and storage proteins, thus creating two distinct regions—the ‘‘vitreous’’ or glassy endosperm and the ‘‘starchy’’ endosperm. The vitreous endosperm transmits light, whereas the starchy endosperm does not. Typically, the endosperm is 90% starch and 10% protein (Gibbon and Larkins, 2005). Normal maize protein is deficient in two essential amino acids (lysine and tryptophan) and has a high leucine:isoleucine ratio and biological value (Babu et al., 2004). A naturally occurring recessive mutant gene opaque‐2, observed first in a Peruvian maize landrace, gives a chalky appearance to the kernels and has improved protein quality due to increased levels of lysine and tryptophan in the endosperm (Mertz et al., 1964). However, this trait appears to be associated with inferior agronomic traits such as brittleness and increased susceptibility to insect pests. With the discovery of ‘‘modifier genes’’ (mo2) that alter the soft, starchy texture of the endosperm, maize breeders developed hard endosperm o2 mutants designated as ‘‘quality protein maize’’ (QPM) (Nelson, 2001; Prasanna et al., 2001), which have the phenotypes and yield potential of normal maize but maintain the increased lysine content of o2. Opaque 2 is a recessive trait but due to the eVect of the modifiers, QPM behaves as a quantitative trait. Using SSRs and backcross breeding, Babu et al. (2004) developed maize lines that had twice the amount of lysine and tryptophan as compared to local cultivars and recovered up to 95% of the recurrent parent genome. Sweet corn is another class of edible‐grade maize, which is highly preferred as roasted/or boiled cobs. In sweet corn, breeding for improved seedling emergence and eating quality is complicated because of the inverse relationship between these traits. High kernel sugar content is one of the reasons for poor seedling emergence (Douglass et al., 1993), influenced by many kernel characteristics that are under the control of many genes (Azanza et al., 1996a,b). Evaluation of these traits requires diYcult and expensive characterization in the laboratory. However, using marker‐assisted backcross or population breeding, it has been possible to select progenies with improved seedling emergence that also has high sucrose content (Yousef and Juvik, 2001a, 2002). Barley: malt is a major raw material for the production of beer. Characters that aVect malting quality include malt extract content, a‐ and b‐amylase activity, diastatic power, malt b‐glucan content, malt b‐glucanase activity, grain protein content, kernel plumpness, and dormancy, all are quantitatively inherited variously influenced by the environment (Zale et al., 2000). There are few barley cultivars with good malt quality that brewers are reluctant to change from due to their concerns about the resultant changes in flavor and brewing procedures. For example, the goal of US Pacific Northwest barley breeding program is to produce high‐yielding NILs that maintain traditional
232
S. L. DWIVEDI ET AL.
malting quality characteristics but transfer QTL associated with yield, via marker‐assisted backcrossing, from the high‐yielding cultivar Baronesse to the North American two‐row malting barley industry standard cultivar Harrington. Schmierer et al. (2004) targeted Baronesse chromosome 2HL and 3HL fragments presumed to contain QTL that aVect yield. Using backcross breeding and QTL/marker information, they identified a NIL (00–170) that when evaluated for yield over 22 environments and for malt quality over 6 environments produced yield equal to Baronesse while maintaining a Harrington‐like malt quality profile. Other studies have also reported the development of lines with improved malt quality: white aleurone color and high a‐amylase content (Ayoub et al., 2003), and high in b‐glucan and fine‐coarse diVerence (Igartua et al., 2000). Soybean: Concibido et al. (2003) introgressed yield‐enhancing QTL from exotic soybean germplasm Glycine soja (PI 407305). They detected yield‐ enhancing QTL located on LG B2 (U26). In a 2‐year multilocation trial, individuals carrying the PI407305 haplotype at the QTL locus demonstrated 8–9% yield advantage over individuals that did not contain the exotic haplotype. When assessing the QTL eVect in various elite genetic backgrounds, they found that this QTL conferred enhanced yield in only two of the six genetic backgrounds, although individuals carrying the PI407305 haplotype at the QTL locus always had an average 9% yield advantage in yield trials across locations. Common bean: Tar’an et al. (2003) used an index based on QTL‐linked markers and ultrametric genetic distances between progeny lines and a target parent to select for increased yield in their breeding program. Lines with a combination of phenotypic performance and high QTL‐based index produced greater yield over those developed by using high QTL‐based index, conventional phenotypic selection, and a low QTL‐based index. They also demonstrated that the use of the QTL‐based index in conjunction with the ultrametric genetic distance to the target parent would enable a plant breeder to select lines that retain important QTL in a desirable genetic background. Pea: Resistance to lodging, a key objective in many pea breeding programs, is controlled by two genes that markers A001 (in coupling phase) and A004 (in repulsion phase) are associated with resistance to lodging (Warkentin et al., 2004). Zhang et al. (2006a) evaluated the eVectiveness of these markers in F2 population of eight crosses. The lowest lodging score for each population was obtained from plants with the combination of A001 (presence) and A004 (absence). They detected a higher proportion of lodging resistant F3 families from this marker combination as compared with phenotypic selection in F3 generation. Thus, A001 and A004 are useful for MAS for lodging resistance in early generation pea breeding populations. The preceding examples demonstrate that marker‐assisted breeding is a viable option to supplement conventional breeding programs for certain
APPLIED CROP GENOMICS
233
traits and where robust markers are available. To date, MAS has been frequently used to transfer simply inherited traits or to pyramiding genes with major eVects but much less for improving polygenic traits. However, a good knowledge of the trait genetics, interaction eVects (epistasis, genetic background, and environment), population size limitations, accurate phenotyping, user‐friendly PCR‐based marker assays, marker‐trait association, and genetic recombination (closer the distance between marker and the gene/QTL, lesser the chance of recombination and loss of selective power), and the ability to timely manage and interpret the voluminous marker data largely influence our ability to successfully integrate MAS into crop breeding programs. In addition, many breeders still consider the use of marker technology as prohibitorily expensive for routine use in breeding programs. However, it is encouraging to note that high‐throughput genotyping platforms for large‐scale, low‐cost applications are rapidly advancing, largely driven by the human diagnostics community. In turn, this is encouraging the development of a genotyping service industry, thus disconnecting breeding programs from the need to establish and maintain capital‐intensive in‐house facilities, although many of these companies struggle to provide a speed of service in‐line with the often very short breeders’ decision window. Hence, the cost for MAS genotyping will become more aVordable to breeding programs but probably only for those who can embrace SNP markers.
D.
SPECIFIC CHALLENGES FOR ALIEN GENE INTROGRESSION
Wild crop relatives are traditionally looked on as potential sources of gene(s) for resistance to many pests and diseases that are not available in cultigens, thus making them a valuable resource for gene transfer in cultivated species. Both conventional crossing and selection, and molecular breeding (MAS and transgenics) have been used to transfer pest and disease resistances from wild relatives to cultivated crop species (Dwivedi et al., 2007 and references therein). Resistance gene(s) from wild relatives have facilitated large‐scale cultivation of crops in disease or pest endemic regions of the world, that is, bacterial blight (BB) and grassy stunt virus in rice, BB in maize and potato, and nematodes in many crops. Wild relatives are usually inferior to modern cultivars with respect to yield and seed quality. However, the successful transfer of improved fruit yield and processing quality in tomato (Bernacchi et al., 1998a,b; de Vicente and Tanksley, 1993; Fridman et al., 2000; Fulton et al., 1997; Rick, 1974; Yousef and Juvik, 2001b) led to the realization that wild relatives can contain beneficial genes (in addition to resistance to biotic stresses) associated with yield and seed quality, although these are often phenotypically masked by deleterious genes and are thus diYcult to identify and transfer through conventional selection and breeding.
234
S. L. DWIVEDI ET AL.
Using advanced backcross and QTL analysis (Tanksley and Nelson, 1996), yield and grain quality enhancing alleles from wild relatives have been successfully introgressed in rice, wheat, barley, sorghum, common bean, and soybean (Dwivedi et al., 2007 and references therein). Dramatic yield advantages have been reported in rice, for example, through the introduction of two yield‐enhancing QTL alleles (yld1.1 and yld2.1) from O. rufipogon (AA genome) into 9311 (one of the top performing parental lines used in the production of super hybrid rice in China) contributed in excess of 20% yield increases in rice; that is, about 1 t ha1 gain in yield in some of the newly bred cultivars, largely because of increases in panicle length, panicles per plant, grains per plant, and grain weight. These improved lines with 9311‐type genetic backgrounds are being used to raise the existing yield potential of super hybrid rice in China (Liang et al., 2004). Oryza grandiglumis (allotetraploid, CCDD genome species) is another wild relative contributing positive alleles for increased grain yield in rice. In contrast, only 6–8% increase in grain yield was reported when positive alleles from Hordeum spontaneum were introgressed into barley. Wild relatives also contributed positive alleles for improved grain characteristics in rice (long, slender, and translucent grains, and grain weight), wheat (grain weight and hardness), and barley (grain weight, protein content, and some malt quality traits). Of particular interest is a locus for grain weight, tgw2, which contributed positive alleles from O. grandiglumis that are independent from undesirable eVects of height and maturity (Yoon et al., 2006). In a similar study, Ishimaru (2003) identified a grain weight QTL, tgw6, responsible for increased yield potential without any adverse eVects on plant type, or grain quality in the Nipponbare genetic background. Similarly, alleles from G. soja conveyed 8–9% increased in grain yield and improved the protein content in soybean (Concibido et al., 2003). Development of exotic genetic libraries (also known as CSSL, IL, or CL) is another approach to enhance utilization of wild relatives to expand crop gene pools (see Section II.A). These genetic stocks provide a well‐ characterized potential resource for uplifting the yield barriers through pyramiding beneficial loci and fixing of positive heterosis. For example, when tomato ILs carrying three independent yield‐promoting genomic regions were pyramided, the progenies produced more than 50% greater yield compared to controls (Gur and Zamir, 2004). In a report (Yoon et al., 2006), several rice lines outperformed Hwaseongbyeo (1 t ha1 increase in grain yield). Several grain characteristics, including grain weight, were improved after crossing an advanced IL containing O. grandiglumis segments, HG101 (very similar to Hwaseongbyeo) with Hwaseongbyeo. The above examples demonstrate that wild relatives contain desirable alleles for agronomic traits, even though their eVect is phenotypically not evident in wild relatives. It is important that more emphasis should be given to exploit
APPLIED CROP GENOMICS
235
wild relatives to identify yield enhancing alleles to further raise the yield potential of crop cultivars. This is now an achievable goal as we progress toward saturating the genetic linkage maps of many crops with user‐friendly markers, and the technological cost of applying marker technology is substantially reduced.
V. SUCCESSFUL APPLICATION OF MARKER‐ASSISTED GENETIC ENHANCEMENT IN PRIVATE SECTOR BREEDING PROGRAMS During the 1990s, MAS was often presented as holding the potential to replace phenotypic selection and dramatically reduce the time required to breed new cultivars (Mazur, 1995). Multinational seed companies have made large investments in genomic technologies and are now routinely using applied genomic tools to (1) dissect the genetic structure of the germplasm to understand gene pools and germplasm (heterotic) groups, (2) provide insights into allelic content of potential germplasm for use in breeding, (3) screen early generation breeding populations in order to select segregants with desired combinations of marker alleles associated with beneficial traits (especially where this avoids the costly phenotypic evaluations), (4) for accelerating the introgression and backcrossing of transgenes into diverse elite breeding lines, and (5) establish genetic identity (through DNA fingerprinting) of their products (Cooper et al., 2004; Crosbie et al., 2006; Fu and Dooner, 2002; Niebur et al., 2004). MAS has been successfully applied in cultivar development for maize (Crosbie et al., 2006; Eathington, 2005; Johnson, 2004; Niebur et al., 2004). Private sector soybean breeders have also made extensive use of MAS to select for resistance to soybean cyst nematode (SCN, Heterodera glycines), phytophthora root rot (Phytophthora sojae), and brown stem rot (Phialophora gregata). Using MAS breeders have been able to fix these resistance traits in their breeding materials before proceeding to yield trials (Cahill and Schmidt, 2004; Cregan et al., 1999; Crosbie et al., 2006). It is reported that MAS has allowed Pioneer to double their rate of genetic improvement for yield among SCN‐resistant cultivars (https://www.pioneer.com/ pioneer_news/press_releases/products/marker_assisted_selection). More recently, Monsanto breeders used MAS in the development of soybean cultivar Vistive that has low levels of linolenic fatty acid, thus reducing the need for postharvest processing to lower or eliminate the presence of unhealthy trans fats from foods. Vistive soybeans meet processor’s growing demand for low‐ linolenic oils, which attract premiums for growers. Other upcoming products from Monsanto are Vistive mid‐oleic (increase shelf life and flavor), Vistive low
236
S. L. DWIVEDI ET AL.
saturates (combining lower saturated fats, lower trans fats, and improved stability), and Vistive omega‐3 (providing consumers new options for omega‐ rich foods) products (http://www.monsanto.com/monsanto/layout/products/ seeds_genomics/oilseeds.asp). Despite these successes, many private sector breeding programs still rely heavily or solely on phenotypic selection and most agree that MAS will never entirely replace phenotypic evaluation. Introgression breeding, also referred to as MABC, has been one of the most, if not the most, successful form of MAS in private breeding programs to date. The use of MABC to introgress transgenes into elite maize or soybean inbred lines (Crosbie et al., 2006; Ragot et al., 1995) has permitted the rapid deployment of transgenic insect and herbicide resistance traits across regions, creating tremendous value for seed companies, farmers, and other downstream actors. MABC is also very eVective for introgressing specific genes or QTL from donor genotypes (nonadapted materials or related species) into elite breeding lines reducing both the time needed to produce commercial cultivars and the risk of undesirable linkage drag with deleterious donor attributes. Reports of successful use of MABC in private breeding programs are scarce in spite of positive outcomes from a variety of public programs on tomato, rice, barley, and soybean (Dwivedi et al., 2007). Financial cost‐benefit considerations will usually determine whether introgression breeding should be conducted with or without the assistance of molecular markers. In public breeding programs, marker‐assisted recurrent selection (MARS) has often been used in the context of population improvement (Gallais et al., 1997; Hospital et al., 1997; Knapp, 1998; Moreau et al., 1998; Xie and Xu, 1998), based on breeding schemes where selected individuals are random‐ mated. In contrast, private breeding programs, in particular for maize, have often implemented MARS schemes focused more on directed recombination (Crosbie et al., 2006; Eathington, 2005; Ragot et al., 2000) in order to recover an ideal genotype through the creation of a mosaic of favorable chromosomal segments from the parental genotypes. This approach is referred to as genotype construction and is based on simultaneous selection for multiple traits (often using marker information only) such as yield, biotic and abiotic stress resistance, and quality attributes (Eathington, 2005; Ragot et al., 2000). Although several of these target traits have complex inheritance, the commercial breeding programs report dramatic increases in the rate of genetic gain over phenotypic selection in maize (Crosbie et al., 2006; Eathington, 2005). The specific molecular breeding systems used by commercial breeding programs are often trade secrets, but it is likely that there are several critical factors in their success including: (1) simultaneous marker‐ only selection for several traits involving probably 10 to more than 50 QTL or genes, (2) multiple cycles of MARS per year using markers flanking
APPLIED CROP GENOMICS
237
QTL, (3) use of oV‐season nursery facilities for generation advance, and (5) genotyping large populations and use marker information to select plants prior to flowering to enable directed recombination. In these breeding systems, phenotypic selection is not applied at every generation. For example, the cycle length in MARS can be as short as 3 months, while that of phenotypic recurrent selection can span from 1 to several years. Such substantial diVerences in cycle length are expected to have significant impacts on the rate of genetic gain over the entire breeding system. Commercial breeding programs have also put great eVorts into reducing costs, not only for genotyping data but also for phenotypic data. It is likely that cost ratio between marker data points to experimental field plot data points is lower in large private breeding programs than in most public research laboratories or small private programs. These are important factors for the economic eYciency of MARS applications. Successful application of MAS in the private sector has been featured by its crops. For example, rice, as an autogamous crop, is very hard to make its hybrid vigor utilized compared to open‐pollinated crops such as maize. Hybrid rice breeding has been depending on using either male sterility and its fertility restoration or environment‐induced genic male sterility for hybrid seed production. The former needs a large number of testcrosses and progeny tests to identify the genes for male sterility and fertility restoration during the breeding process, while the latter depends on specific environments and multiple location or season trials to select for the related genes, both of which are extremely time consuming and labor intensive. MAS in hybrid rice breeding for the traits requiring testcrossing or progeny testing and for environment‐dependent traits has been intensively discussed elsewhere (Xu, 2003), and now has become routine in hybrid breeding using both cytoplasmic male sterility and environment‐induced genic male sterility. In addition, MAS has been widely used in the private sector for seed quality assurance. One of the examples is to identify and remove the false hybrids produced because the temperature during flowering time goes abnormal and down below the critical level that is required for conversion of environment‐ induced male sterility lines from sterility to fertility, which would not happen under normal temperature conditions. The international seed companies have invested heavily in the assembly, modification, and integration of new methods and tools for the detection of DNA polymorphisms, the continuous operation of nurseries, and the optimization of data management, analysis, and interpretation. The development of PCR technology and the large‐scale identification of SNPs (Lindblad‐Toh et al., 2000) have facilitated the development of molecular marker systems amenable to the levels of miniaturization and automation. This has in turn allowed the development of genotyping pipelines capable of rapidly and cost eVectively generating millions of data points a year. It is only
238
S. L. DWIVEDI ET AL.
at this level and timeliness of throughput that large breeding programs can realize true benefits of MAS. The allelic diversity at SNP loci is low (usually limited to two alleles, although generally providing codominant information), and the level of polymorphism at any given SNP loci may also be low in breeding populations. However, this is generally considered to be more than oVset by the very high abundance and random distribution of SNP loci which can be combined and analyzed as haplotypes (Ching et al., 2002). Thus, highly dense genetic maps can be developed with thousands of SNP markers, and marker‐trait associations can be readily identified that are very close or inside the target gene. For these reasons, SNP‐based genotyping is becoming the assay of choice for private MAS programs for well‐studied crops. The ability to select plants without their being phenotypically characterized is one of the main advantages of MAS. Many private breeding programs have upgraded or are upgrading their continuous nurseries (greenhouses, screenhouses, or open fields) so that they can be managed, equipped, and staVed in such a way that the plants complete their life cycle as quickly as possible and that tissue samples be collected eYciently at each generation for genotyping. EYcient MAS programs require access to and synthesis of very large amounts of data of diVerent types (phenotypes, genotypes, pedigrees, environmental characteristics) and from various sources into useful genetic information. The rapidly increasing amounts of data generated in crop research and breeding programs driving dramatic advances in supporting computational sciences. Modern molecular breeding requires a range of complex large‐scale data analyses to be carried out very rapidly. In particular, the development of computer software to track, manipulate, and comparatively analyze data for major genes, QTL, background haplotypes, and phenotypes across germplasm, pedigrees and cycles of the breeding process. Most of the computational tools used in private sector molecular breeding programs have been developed internally and remained under proprietary protection. Some large private breeding programs had established large research and support groups of dedicated data managers prior to the advent of MAS and genomics. Today, there is a fundamental dependence on dedicated specialists, systematically integrated into breeding programs, genotyping pipelines, and repositories of internal and external genetic information. Many private breeding programs have invested heavily in the implementation of MAS. While there are no public reports of the cost‐benefit ratio of the commercialization of MAS‐derived cultivars in private sector, the growing portfolio of patent applications associated with MAS technologies (e.g., US5,492,547 1996; US5,746,023P 1998; US6,368,806B1 2002; US6,399,855B1 2002; US6,455,758B1 2002; US2005/0144664A1 2005; WO2005/000006A2 2005; WO2005/014858A2 2005) clearly suggests that commercial breeding programs see significant comparative advantage from the use of such approaches. Moreover, the likely scale of the investment
APPLIED CROP GENOMICS
239
suggests that commercial seed companies are much more convinced of the benefits of MAS than most public breeding programs. Small‐ to medium‐sized seed companies without access to technology and with limited resources are forming alliances with multinational companies, universities, and CGIAR institutions to enable access to the necessary infrastructure, core competencies, and marker technologies without the prohibitively high‐capital investment normally associated with such endeavors, for example, the ‘‘Agribiotech Park’’ at ICRISAT in India (http://www.agri‐ sciencepark.icrisat.org/amenities.htm), the BecA at ILRI in Kenya (http:// www.biosciencesafrica.org/BecA%20home.htm/), the Agronatura at CIAT in Colombia (http://www.ciat.cgiar.org/agronatura/index.htm), and CRIL of the IRRI‐CIMMYT alliance (http://www:iita.org/cms/articlefiles/490Genonics%20Taskforce%20Report%20March%202006.doc).
VI.
IMPACT OF MARKER‐ASSISTED GENETIC ENHANCEMENT A. ENHANCED SELECTION POWER
The enhanced selection power of DNA markers resides in their ability to precisely identify a plant’s genotype for a specific target trait without the confounding eVects of the environment (Ribaut and Hoisington, 1998). The selection of genotypes based on genetic values predicted by molecular marker data can increase the rate of genetic gain by enhancing the precision of selection and by shortening selection cycles (Meuwissen et al., 2001). MAS may also be valuable for pyramiding genes of similar phenotypic eVect or selecting for resistance to pests and diseases not present in the breeding location. The high heritability of genetic markers (in theory being 1.0, although in practice rarely achieving this absolute level) compared to the trait for which they have been developed make them useful for MAS. Improvements in marker techniques have increasingly added to the selection power of MAS, both by providing more reliable types of markers and a rapidly increasing list of trait‐associated loci. A critical improvement was the move from time‐consuming hybridization‐based assay (RFLP) to PCR‐ based assays (initially RAPD) for which amplification is dependent on DNA concentration and quality, annealing temperature and thermocycling conditions, Taq polymerase concentrations, and the relative proportion of all components in the PCR cocktail. Unfortunately, RAPD suVers many reproducibility and transferability problems, thus considerable eVorts have been made to develop more robust PCR‐based marker systems such as SCAR markers and other single‐copy markers which have proven more reliable and repeatable and therefore of higher heritability. However, most recently two
240
S. L. DWIVEDI ET AL.
new classes of PCR‐based marker have emerged that have the added advantage of being highlight polymorphic in most breeding populations (SSR markers) or highly abundant across most plant species genes (SNP markers). SSR and SNP markers oVer greater precision, power of selection, and perhaps most importantly, ease of scale‐up, and thus, have become the markers of choice for molecular breeding programs of most crops. Thus, the type of marker has become an important determinant of the power of MAS to enhance selection. The selection power of molecular markers also resides in their good genome coverage and capacity to provide complete genome information, a characteristic that has also improved with newer marker technologies. The enhanced selection power of MAS in addition to being related to the reliability and ease of applying a given type of marker also depends on proximity of linkage between markers and the gene(s) of interest (Ribaut et al., 1997b). In addition, the level of phenotypic variance explained by the marker compared to the total genetic variance for the trait is also a critically important criterion (Bearzoti and Vencovsky, 1998). Greater distance between a marker and the gene(s) of interest underlying the target trait reduces the power of selection. In terms of linkage, the nature of the cross, particularly in terms of how closely are the parents related to each other and to the pedigree of target breeding populations, aVects the frequency of recombination around target genes within the mapping populations versus the target breeding populations. The choice of parental genotypes for mapping populations also determines the level of polymorphism and whether the marker will facilitate the positive selection for the desirable or undesirable alleles. The potential risk that recombination will decouple the linkage between marker locus and gene of interest can be addressed by using flanking markers, which have greater power to counteract the eVects of recombination around loci of interest by providing a diagnostic for the introgression of an entire genomic segment. MAS is most eVective when there is a high level of polymorphism in the crosses being screened, and this is also the breeding situation in which gene introgression is most diYcult, time consuming, and plagued by linkage drag. Not surprisingly, therefore, marker‐assisted introgression and marker‐accelerated backcross breeding are the areas where genomic applications have had their widest application and greatest success. Thus, there is a range of successful reports of using flanking markers for introgression of new traits through interspecific crosses with wild relatives or crosses between gene pools within the cultivated species, where markers are often more eVective. In the case of markers linked to the QTL, the proportion of the total phenotypic variance conveyed by each QTL is a key to the value of that marker in enhancing the breeding gain for the target trait. Similarly, there should be a high level of confidence in the existence of a QTL associated with
APPLIED CROP GENOMICS
241
the target trait, as determined by the use of high LOD likelihood threshold during the identification of QTL markers (Tanksley, 1993). Simulation studies have shown that when a moderate‐to‐large number of QTL are influencing the target trait, a whole‐genome scanning approach is often necessary and that the eYciency of MAS is substantially aVected by population size and heritability of the target trait (Bearzoti and Vencovsky, 2002; Lande and Thompson, 1990). Enhanced power of selection through MAS can come not only from the power to make positive selection for a single gene but also from its power to assert negative and positive selection for a suite of genes or QTL across the entire genome (Hospital and Charcosset, 1997). It is in this transition from single point interventions of MAS to holistic molecular breeding strategies that we expect to see an exponential gain from the application of genomics in plant breeding programs. In this case, marker genotypes at various loci (associated with several mono‐, oligo, and/or polygenic traits) are used within the context of an index for eliminating part of a breeding population, thus reducing nursery growout space and costs (Bearzoti and Vencovsky, 1998; Gimelfarb and Lande, 1994, 1995). The most common application of MAS is in marker‐assisted/accelerated backcross breeding. Optimally, this is based on positive foreground selection for donor trait, positive background selection for the recurrent parent genome, and negative background selection against undesirable donor parent alleles (Frisch et al., 1999b; Ribaut et al., 2002). Marker‐assisted introgression can dramatically reduce the number of generations of backcrossing required to recover the elite parent background (Hospital et al., 1992), although the number of generations saved depends on the size of the genome, level of recombination in the cross, size of the progeny population, and number of available markers. Genomic map length, population size, and duration of backcrossing also influence on the attainable rate of donor genome substitution. For example, larger genome requires larger population as well as more markers to attain a given rate of donor genome substitution (Stam, 2003). Meanwhile, partial or whole chromosome selection can be used when introgressing from an exotic genome where recombination with the cultivated genome is very low or nonexistent (Wittaker et al., 1995). MAS can also be a great assistance in the selection of favorable recombinants during inbreeding and/or crossbreeding cycles using backcross products, thus increasing the speed with which advanced lines are generated (Frisch et al., 2000). Furthermore, MABC can reduce the eVects of linkage drag by selecting for fewer and smaller donor genome fragments. In this case, increasing selection power and breeding gain is obtained by use of a greater number of background markers combined with closer flanking markers for the target trait gene(s). Using computer simulations and additive, dominance, and epistasis genetic model, Liu et al. (2004a) demonstrated that combining MAS in
242
S. L. DWIVEDI ET AL.
early generations with phenotypic selection in later generations is the most eYcient breeding strategy for self‐fertilizing crops. Investigation on diVerent crossing strategies and consideration of when to screen, what proportion to retain, and the impacts of dominant versus codominant marker expression revealed important choices in the design of MAS programs that can produce large eYciency gains. F2 enrichment, increasing homozygosity through inbreeding or DH, and backcrossing to increase the frequency of recurrent parent alleles are eVective strategies for improving the eYciency of MAS that will allow either smaller populations to be screened or selection at more loci. However, fixation of alleles in early generation requires larger populations and is undesirable in most instances (Bonnett et al., 2005).
B. REDUCED COST, INCREASED FEASIBILITY, TIME SAVINGS, AND PARENTAL SELECTION MAS can be useful for the selection of traits that are diYcult or impossible to breed through phenotypic selection due to logistical, biological, or quantitative‐based constraints. In terms of genetic associations, codominant markers for recessive genes are especially valuable since phenotypic selection will be highly ineYcient as it is likely to discard all heterozygous progeny during early generations of the breeding cycle. While recessive genes can be selected with progeny testing or testcrosses, this clearly adds substantial time and eVort to the breeding process. Thus, MAS has the advantages of obviating these time‐consuming steps and facilitating precise and eYcient early generation selection. Dominant markers in coupling phase with target trait can also be of value in such breeding systems. However, if only a dominant marker in repulsion is available, then early generation MAS would be limited to negative selection against homozygous dominant and heterozygous plants, which would be ineYcient since potentially useful allele‐carrying genotypes would be eliminated. This type of marker is most useful in advanced generations of self‐pollinated crops when a recessive gene has already been fixed by inbreeding. However, MAS with this type of marker is impossible in generations where no homozygous recessive plants exist at all such as the BC1F1 to the dominant allele‐containing parent. MAS scenarios with the greatest cost‐benefit ratio include traits that would otherwise require highly expensive phenotypic or biochemical evaluation procedures (Ribaut and Hoisington, 1998). This is the case for traits that require extensive field testing at specific locations or times of the year. Likewise, many phytochemical traits analyzed in reproductive or vegetative tissues at various growth stages are expensive to carry out. For example, the analysis of seed quality, secondary metabolites, and micronutrients remains expensive and time‐consuming and MAS can replace more costly and
APPLIED CROP GENOMICS
243
diYcult assays with more standardized DNA‐based technologies. Molecular markers are proving more eYcient, rapid, and simple to implement on a large scale for seed protein traits since they are based on DNA extracted at any growth stage from a small amount of expendable tissue. For example, in the selection of quality protein maize, MAS is cost‐eVective when a visual marker is not available (Dreher et al., 2002, 2003; Morris et al., 2003). Similarly, for the evaluation of mineral content in seed tissue, MAS might be less expensive than traditional quality evaluations, a process that sometimes requires dissected seed organs or collecting several grams of seed tissue. The advantage of MAS resides in the small amount of template DNA required for carrying out a large number of assays. Thus, MAS eYciency can be dramatically increased by using a single DNA extraction for the evaluation of several to many markers. After the development of molecular markers and validation of their power of indirect selection for the trait (see Section III), it is then often necessary to optimize the assay for scale‐up to large‐scale application (Young, 1999). Sometimes this involves changes in breeding program logistics, PCR protocols, marker detection technique, or even complete redesign of the markers themselves. In all cases, the driving criteria being to reduce unit costs and turn around times while increasing throughput and minimizing errors, and ultimately optimizing the cost‐benefit advantage of MAS over phenotypic selection. Marker redesign has been a common element of scaling‐up exercises and can involve something as simple as optimizing the size or genomic position of the PCR amplification fragment. Technologies that speed up the implementation process, reduce laboratory requirements or errors, and lower the costs associated with scaling‐up are crucial to the success of MAS (Gu et al., 1995). For example, techniques have been developed which reduce the cost of DNA extraction and result in large time‐savings (Dellaporta et al., 1983; Ikeda et al., 2001b; Klimyuk et al., 1993). Kuchel et al. (2005) designed a genetically eVective and economically eYcient marker‐assisted breeding strategy aimed at selecting for favorable alleles in wheat breeding. Although incorporating MAS for allele enrichment in the BC1F1 population, gene selection at the haploid stage, and the selection of recurrent parent background of DH prior to field testing was eVective to select for a high frequency of desired alleles, the incorporation of marker selection at the BC1F1 and haploid stage was the most eVective as it not only increased genetic gain over the phenotypic selection but also reduced cost by 40%. Furthermore, MAS can be used in conditions that are not favorable for phenotypic screening, for example selection of resistance genes in regions where quarantine restriction prevents introduction of an exotic pathogen or pathogen strain or where a pathogen does not occur at a suYciently high level to perform eVective field screening and selection (Ribaut and Hoisington, 1998). Markers for disease resistance have the advantage of
244
S. L. DWIVEDI ET AL.
obviating the need for field or greenhouse inoculations that sometimes are ineVective or unreliable if environmental conditions are not propitious and can result in savings in time and cost compared to phenotypic selection. A further advantage of MAS is that it can be implemented in any generation of the breeding process and under both field or greenhouse conditions, while phenotypic selection often requires planting a separate trial and provision of specialized labor for inoculation, agronomic management, and evaluations or scoring. In addition, phenotypic screening of fixed lines or segregating populations often requires replicated testing to minimize the eVect of GEI, whereas MAS can be evaluated on a single plant basis as long as the marker is associated with a locus which contributes a large percentage of the genetic variance of the target trait. A potential disadvantage of relying on MAS over phenotypic selection is that it commits a breeder to a unique gene or set of genes for a given trait. Thus, where a breeder relies solely on MAS for selection, this can exclude other possible genes and the use of other potentially useful parents that do not possess the allele(s) being targeted by the MAS. Of course, this is rarely the recommended approach, and most molecular breeding programs will involve at least one or two cycles of phenotypic evaluation during the overall breeding process. In this way, the results of the MAS can be validated, while other alleles and genes positively contributing to the target trait can be selected. A refined model for this approach has been proposed by Ribaut and Betra´n (2000) in maize for fixing valuable genes in a population improvement breeding program (that includes a large number of parents) through the application of single large‐scale (SLS) MAS and then intercrossing to recreate diverse populations for further selection. MAS can also help in situations where timeliness is a major constraint since DNA can be obtained at the seedling stage or depending on the crop, even from the seed itself. Timeliness is an especially important issue in the case of perennial crops where many economically important traits are only expressed at the reproductive stage which may take one or more years. Therefore, MAS for late cycle traits in long‐duration crops provides a much greater cost‐benefit ratio than in annual crops (Morris et al., 2003). When breeding complex traits with low heritability and high GEI, selection based on phenotypic evaluation can become very diYcult. In these situations, the dissection of complex traits into component traits can increase the chances of eVective selection as each component can be selected separately. Then, in turn, MAS for major QTL underlying each component trait may provide the best breeding gains. Selection of just the QTL that account for the largest proportion of phenotypic variance is advisable under these conditions (Ası´ns, 2002; Tanksley, 1993). In the case of polygenic traits, MAS has the potential for pyramiding diVerent sources of genes for a given trait, whether it be to create durable disease resistance through simultaneous deployment of multiple R gene combinations or to create superior cultivars
APPLIED CROP GENOMICS
245
through the accumulation of positive alleles for diVerent components of a given trait such as drought or low soil fertility tolerance. There are a number of good examples of successful pyramiding of pest or disease‐resistance genes (see Section IV.A). However, there are very few reports of successful applications of MAS for complex abiotic stress tolerance traits (see Section IV.B). Thus, the long held belief that MAS would have its greatest impact on trait with low heritability and high GEI interaction still awaits widespread practical demonstration. However, experience has shown that the ability to manipulate even one important component trait with confidence can make a breeding program more eYcient if that gene is highly desirable and valuable for advanced materials. MAS can also be useful in the selection of parental genotypes, especially in the breeding of crops where heterosis is expressed. In this case, parental selection can benefit from marker assessments of genetic distance between individuals in crops where genetic distance has been shown to be predictive of heterotic pools or combining ability. Finally, MAS can also be used to determine heterozygosity during the creation of inbred lines for allogamous crops.
C. OVERVIEW OF PRODUCTS FROM MOLECULAR BREEDING To date, polymorphic DNA markers and genetic maps are available for virtually all crops, albeit in varying numbers and levels of genomic saturation (see Sections II.B and C). Similarly, the genetics of many agronomic traits is well understood in many crops, and the marker‐trait linkages have been reported for many traits in a large number of crops, although reports of validation in diVerent genetic backgrounds and environments are naturally only beginning to emerge (see Section III). MAS is now being practiced in most well‐studied crops (see Section IV.A–C), yet in the private sector MAS applications are dominated by transgene introgression and backcross programs with only limited reports of their use for complex traits. In this section, we provide an overview of the products (cultivars and breeding lines) developed using MAS in combination with conventional breeding. Eighteen MAS‐derived cultivars and several advanced lines combining resistance to biotic and abiotic stresses or improved grain quality have been reported in rice, wheat, barley, pearl millet, common bean, and soybean (Table XV). To date, MAS has been most successful in the selection of resistance to diseases and for improving grain quality. For example, rice cultivars resistant to blast in United States and to bacterial blight in Indonesia, wheat cultivars resistant to rust in Canada, and common bean cultivars resistant to anthracnose and Bean golden yellow mosaic virus in United States, and those with resistance to Sclerotinia white mold in Canada have been developed using MAS and
246
S. L. DWIVEDI ET AL.
Table XV List of Cultivars and Hybrids, Advanced Lines and Improved Germplasm Developed by MAS in Barley, Common Bean, Pearl millet, Rice, Soybean, and Wheat Advanced lines and cultivars developed by marker‐aided breeding
References Barley
Aluminum Advanced lines including WB259, possessing good malt quality and aluminum tolerance developed in Australia Grain yield and malt quality An isogenic line 00‐170 consistently produced high yield and good malt quality in Australia
http://www.cdesign.com.au/ bts2005/pages/papers_2003/ papers/134venkatanagappaS.pdf Schmierer et al., 2004
Common bean Angular leaf spot Resistance to angular leaf spot transferred into Carioca type bean, Ruda´ in Brazil Anthracnose Co‐42 allele transferred into pinto beans (highly susceptible to Durango race) grown in North America Resistance to anthracnose incorporated in Pinto bean cultivar, USPT‐ANT‐1 containing Co‐42 gene that confers resistance to all known North American races of anthracnose in United States Resistance to anthracnose transferred in cultivar Perola in Brazil
(M. Blair, CIAT, personal communication) Miklas and Kelly, 2002 Miklas et al., 2003
Ragagnin et al., 2003
Bean common mosaic necrosis virus (BCMV) Red bean with resistance to BCMV, containing I and bc‐3, developed for central America
Beaver et al., 1998
BCMV and anthracnose 1800 breeding lines of climbing beans, containing bc‐3, I, C0‐4, and Co‐5, with combined resistance to BCMV and anthracnose selected in Colombia
http://www.african crops.net/ abstracts2/bean/blair.htm
Bean golden yellow mosaic virus (BGYMV) A pole bean cultivar, Genuine, resistant to BGYMV developed in Central America A pole garden bean cultivar, Genuine, with moderate resistance to BGYMV developed in United States Comman bacterial blight (CBB) Pinto bean germplasm, ABCP‐8, resistant to CBB developed in United States USDK‐CBB‐15, dark red kidney bean, highly resistant to CBB released in United States
Stavely et al., 1997 Stavely et al., 2001
Mutlu et al., 2005 (M. Blair, CIAT, personal communication) (continued)
APPLIED CROP GENOMICS
247
Table XV (continued ) Advanced lines and cultivars developed by marker‐aided breeding
References
CBB, anthracnose, and BCMV Advanced lines with multiple resistance to CBB, BCMV, and anthracnose developed in Canada Rust Rust resistant genes, Ur‐4 and Ur‐5, combined in the BARC‐rust resistant green and waxy bean germplasm lines in Honduras Rust resistant genes, Ur‐4 and Ur‐11, introgressed into navy bean lines BelMiDak‐RR‐1 to 7 in Honduras Rust and anthracnose Five lines resistant to rust and anthracnose developed, with Vi0699 and Vi2599 significantly outyielding controls in Brazil Rust, anthracnose, and angular leaf spot Resistance to anthracnose in TO and AB136; to angular leaf spot in AND277; to rust in Ouro Negro; and to rust and anthracnose in Ouro Negro transferred in Brazil Rust and Bean golden yellow mosaic virus (BGYMV) White‐seeded Snap bean cultivars, BELDADE‐RGMR 4, 5, and 6, possessing resistance to rust and BGYMV released United States Sclerotinia white mold QTL B7 and B8 QTL linked with resistance to white mold transferred into Winchester and Maverick that yielded at par with controls in Canada Pearl millet Downy mildew The parental lines of the original hybrid (HHB 67) improved for downy mildew resistance through MAS and conventional backcross breeding, and new hybrid HHB 67‐2 with improved resistance to downy mildew released in India
http://www.ontariobeans.on.ca. ppyramidingDisease ResistanceGenes.html Stavely and Steinke, 1990; Stavely and McMillan, 1992 Stavely et al., 1994
Faleiro et al., 2004
(M. Blair, CIAT, personal communication)
(M. Blair, CIAT, personal communication)
Miklas et al., 2004 (http://www. whitemoldresearch.com/ presentation2004/Miklas.pdf)
http://www.secheresse.info/article. php3?id‐article¼1919
Rice Amylose content Cadet and Jacinto with unique cooking and processing quality traits released in United States Bacterial blight (BB) Angke and Conde, possessing resistance to BB, produced 20% greater yield over IR64 and released in Indonesia
http://usda‐ars‐beaumont.tamu. edu/marker.html Bustamam et al., 2002
(continued)
248
S. L. DWIVEDI ET AL. Table XV (continued )
Advanced lines and cultivars developed by marker‐aided breeding
References
Resistance to BB transferred in R8006 and R1176 and when crossed to Zhong 9A, the hybrids (Zhongyou 6 and Zhongyou 1176) produced high yield, resistant to BB, and good grain quality in China AR32‐19‐3‐3, AR32‐19‐3‐4, AR32‐4‐3‐1, and AR32‐4‐ 58‐2, all resistant to BB, showed 18–31% yield advantage over PSB Rc28 in Philippines BB resistant hybrids, Guofeng No 2 and Hybrid II You 218 released in China, produced 11–19% greater yield over Shanyou PR 106‐P2 and PR 106‐P9, both resistant to BB, showed 18–22% yield increase over PR 106 in India Blast CS 2, CS 11, CS 18, CS 35, CS 36, CS 62, and CS 67 combining resistance to blast and good agronomic traits developed, with potential to replace CR 203 in Vietnam
Cao et al., 2003
Leung et al., 2004
Leung et al., 2004
Leung et al., 2004
http://www.Vtc.agnet.org/library/ article/rh2003013a.html
Soybean Oil quality Vistive low‐linolenic soybean developed by Monsanto and released for cultivation in United States Wheat Aluminum toxicity Advanced backcross lines tolerant to aluminum developed
http://www.monsanto.com
http://www.dfid‐psp.org/ccstudio/ publications/annualreport/ 2004_aluminium.pdf
Bread‐making quality A wheat cultivar, Burnside, with CWES (Canadian Western Extra Strong) traits developed in Canada
Radovanovic and Cloutier, 2003)
Fusarium head blight (FHB) NILs containing major 3BS QTL and resistant to FHB developed in United States
Zhou et al., 2003b
Rust Resistance to stem (Sr39) and leaf rust (Lr35) incorporated into ‘‘Canada Prairie Spring’’ and ‘‘Canada Western Extra Strong’’ classes of wheat lines in Canada
Gold et al., 1999
Multiple resistance to pest, fungal and viral diseases þ grain quality Several germplasm lines possessing resistance to pest, http://maswheat.ucdavis.edu fungal, and viral diseases, and those with improved grain quality developed in United States
APPLIED CROP GENOMICS
249
released for commercial cultivation. Two rice cultivars with MAS‐derived improvements in amylose content are grown in United States. MAS has also been successful in the development of disease‐resistant hybrids. For example, superior rice hybrids with resistance to bacterial blight in China and pearl millet hybrid with resistance to downy mildew have been released for cultivation. In addition, many advanced lines and improved germplasm combining multiple resistances to diseases or with improved seed quality have been bred, which are now being evaluated in several countries prior to their release as new cultivars (Table XV). Marker‐assisted backcross breeding and marker‐ aided gene pyramiding have been the most frequently used molecular breeding methods to aid the introgression of disease resistance or quality traits into improved genetic backgrounds. MAS has also been used in wide crosses to minimize the linkage drag associated with beneficial traits (see Section IV.D). Although there are only small numbers of reports regarding successful use of MAS in plant breeding, the technology has nevertheless demonstrated its potential as a tool to support conventional genetic enhancement of crops. Large‐scale adoption of MAS technology has already begun for incorporating disease resistance or grain quality in rice (http://www.uark.edu/ua/ ricecap/index.htm), wheat (http://maswheat.ucdavis.edu), barley (http://www. barleycap.org/), and common bean (Kelly et al., 2003; Miklas et al., 2006a) in United States. For example, MAS wheat consortium has developed protocols for more than 40 molecular markers for resistance genes and quality traits and used MABC to incorporate 27 diVerent disease‐ and pest‐resistance genes and 20 alleles with beneficial eVects on bread making and pasta quality into 180 lines adapted to the primary US production regions (http:// maswheat.ucdavis.edu/). Rice researchers in China are using MAS to combine resistance to diseases and improved grain quality in some of their best‐ performing hybrids (Leung et al., 2004). MAS is being used to combine disease resistance and/or grain quality in wheat and common bean in Canada (Radovanovic and Cloutier, 2003; http://www.ontariobeans.on.ca.ppyramiding DiseaseResistanceGenes.html) and for improving wheat, barley, and rice in Australia (Christopher et al., 2004; Eagles et al., 2001; McLauchlan et al., 2001; Ogbonnaya et al., 2001; Paris et al., 2003; Schmierer et al., 2004; http://www.cdesign.com.au/bts2005/pages/papers/134venkatangappaS.pdf). CIMMYT wheat breeding program has already initiated marker‐assisted breeding to introgress gene(s) for resistance to cereal cyst and root lesion nematodes, boron toxicity, Barley yellow dwarf virus, scab, rust, and crown rot as well using Ph1b to promote pairing between alien and wheat chromosomes to accelerate gene transfer from alien species to wheat. Moreover, it is expected that many more successful applications do exist but remain within the confidentiality restriction of commercial breeding companies around the world. Developing countries are not left behind in the use of MAS in crop breeding programs. For example, researchers from the Indian Council of
250
S. L. DWIVEDI ET AL.
Agricultural Research are collaborating with their colleagues at IRRI, CIMMYT, and ICRISAT on the use of MAS in cereal and legume breeding. In fact, the first downy mildew resistant pearl millet hybrid (HHB 67‐2) released in India was bred using MAS by improving the male parent with improved resistance to downy mildew (Hash, 2005). India is testing marker‐ derived submergence‐tolerant lines (Xu et al., 2006), developed through collaboration with IRRI, for their adaptation to deepwater paddy cultivation in eastern India. Development of submergence‐tolerant cultivars using MAS has already been reported from Thailand (Siangliw et al., 2003), and work is in progress to introduce this trait in cultivars adapted in Bangladesh, Laos, the Philippines, and Vietnam. The ultimate goal of this collaboration with IRRI is the development of improved rice inbred and hybrid cultivars with good grain quality and multiple resistances to pests and diseases. MAS‐ derived rice cultivars are already being grown in Indonesia. These marker‐ aided rice cultivars and hybrids have produced on average 11–34% increased yield over popular inbred and hybrid cultivars in Asian countries. This has led to an estimated increase in grain harvest of 0.8 million Mt (worth US $20.5 million) of paddy rice per cropping season in India, Indonesia, the Philippines, and China as a result of the growing bacterial blight resistance present in these inbred and hybrid cultivars (Leung et al., 2004). Many national programs from South America are cooperating with CIAT and advanced research institutes in United States to improve the genetic potential of common bean, the most widely grown pulse crop in that region, by using MAS (Miklas et al., 2006a).
VII. APPROACHES TO ENHANCE THE EFFICIENCY AND SCOPE OF MOLECULAR BREEDING A. STUDYING THE MOLECULAR BASIS OF HETEROSIS Heterosis is defined as the superior performance of an F1 hybrid as compared with its parents. Hybrid cultivars have made significant contribution to world food supply (Duvick, 1999). In the literature, dominance, overdominance, and epistasis have been implicated as the genetic basis of superior hybrid performance. The dominance model attributes increased vigor to the action of favorable dominant alleles from both parents combined in the hybrid, whereas the overdominance model postulates the existence of loci where the heterozygous state is superior to either homozygote (Xiao et al., 1995; Xu, 2003; Yu et al., 1997). Evidence for the role of epistasis (interaction of the favorable alleles at diVerent loci contributed by the two parents) in hybrid vigor have also been reported (Li et al., 2001b; Luo et al., 2001;
APPLIED CROP GENOMICS
251
Stuber et al., 1992; Xu, 2003). The genetic basis of heterosis, heterotic groups, hybrid prediction and hybrid performance, relationships between heterozygosity and genetic distance with hybrid performance and heterosis, and use of MAS in hybrid breeding have been discussed elsewhere (Xu, 2003). The complex nature of heterosis makes it diYcult to partition into individual components because of the epistatic interactions among segregating loci throughout the genome (Li et al., 2001b). To assess the importance of loci with overdominant (ODO) eVects in expression of heterosis, Semel et al. (2006) employed NIL, carrying single marker‐defined chromosome segments from distantly related wild species Solanum pennellii to partition heterosis into defined genomic regions, eliminating a major part of the genome‐wide epistasis. They detected 841 QTL for 35 diverse traits. NILs showing greater reproductive fitness are characterized by the prevalence of ODO QTL, which were virtually absent for the nonreproductive traits. Overdominance results from true overdominance due to allelic interactions of a single gene or from pseudo‐overdominance involving linked loci with dominant alleles in repulsion. In their study, although they detected dominant and recessive QTL for all phenotypic traits, overdominance only for the reproductive traits indicates that pseudo‐overdominance is unlikely to explain heterosis in NIL, thus they favor the true ODO model, a single functional Mendelian locus involved in heterosis. Milborrow (1998) proposed a mechanistic, biochemical interpretation of the superior performance of F1 hybrids in comparison to their homozygous parents. Their interpretation is based on the concept that growth is restricted below the potential maximum by internal genetic factors. In this model, the hybrid vigor is caused by a slight reduction in the strictness of this control mechanism in heterozygotes compared with homozygotes, particularly with respect to metabolism and growth processes. This eVect is believed to be mediated by the presence of changes in regulatory features of certain loci when in the heterozygote state. Among the cereals, heterosis has been exploited in maize, rice, sorghum, and pearl millet to produce superior yielding hybrids that by far dominate the global acreage for each crop. For example, about 95% of US maize acreage is planted to hybrids that exhibit a 15% yield advantage relative to the best open‐pollinated cultivars (Duvick, 1999). A popular hybrid rice cultivar in China (LYP9) produces 20–30% more grains per hectare than other hybrids or inbred rice cultivars (Lu and Zhou, 2000). More recently, an ‘‘immortalized F2’’ population was generated by randomly permutated intermating of 240 RILs from a cross between the parents of Shanyou 63, another widely cultivated hybrid rice cultivar in China. These lines were field evaluated over 2 years and genotyped using 231 polymorphic molecular markers covering the entire rice genome. From this analysis, 33 loci were detected that
252
S. L. DWIVEDI ET AL.
contributed to heterotic eVects in grain yield, tillers per plant, grains per panicle, and 1000‐grain weight (Hua et al., 2003). The heterotic loci showed little overlap with QTL previously identified for the same traits. Thus, in contrast to the Milborrow model (Milborrow, 1998), it appears that in rice there are unique loci conditioning heterosis. Moreover, all kinds of genetic eVects were observed in this study to contribute to heterosis, including partial‐, full‐, and overdominance at the single‐locus level and all three forms of digenic epistatic interactions (additive by additive, dominance by dominance, and additive by dominance). Heterosis eVects at the single‐locus level, in combination with the marginal advantages of double heterozygotes caused by dominance interaction at the two‐loci level, adequately explain the genetic basis of heterosis in Shanyou 3. Using serial analysis of gene expression (SAGE), Bao et al. (2005) surveyed transcriptomes in panicles, leaves, and roots of a super‐hybrid rice (LYP9) in comparison to its parental inbred cultivar genotypes (93‐11 and PA64s). They identified 595 upregulated and 25 downregulated tags in LYP9 that were related to enhancing carbon‐ and nitrogen‐assimilation, including photosynthesis in leaves, nitrogen uptake in roots, and rapid growth in both roots and panicles. This adds a crucial new set of observations for understanding the molecular mechanisms of heterosis and gene regulation networks in rice. In this study, they found massive complementation at the transcript level that further suggests that the underlying mechanisms of heterosis may not be as simple as have been reported from studies of a small number of genes (Birchler et al., 2003). Previous studies using multiple hybrids and their corresponding parents revealed that some diVerential gene expression patterns are significantly correlated with heterosis in wheat (Ni et al., 2000, 2002; Sun et al., 1999, 2004; Wu et al., 2003). However, information on systematic identification and on characterization of diVerentially expressed genes is limited. Yao et al. (2005) used an interspecific hybrid between common wheat (Triticum aestivum L., 2n ¼ 42, AABBDD) line 3338 and spelt (Triticum spelta L., 2n ¼ 42, AABBDD) line 2463, which is highly heterotic both for aerial growth and root related traits. In their research, they included an expression assay using modified suppression subtractive hybridization (SSH) to generate four subtracted cDNA libraries between the wheat hybrid and its parental genotypes. Of the 748 nonredundant cDNAs obtained, 465 cDNAs had high sequence similarity to GenBank entries in diverse functional categories, such as metabolism, cell growth and maintenance, signal transduction, photosynthesis, response to stress, transcription regulation, and others. They further confirmed the expression patterns of 68.2% SSH‐derived cDNAs by reverse Northern blot, while semiquantitative RT‐PCR exhibited similar results (72.2%). This suggests that the genes diVerentially expressed between hybrids and their parents are involved in diverse physiological pathways, which may contribute to heterosis in wheat.
APPLIED CROP GENOMICS
253
Maize inbred lines B73 and Mo17 produce a heterotic F1 hybrid. Based on analysis with 13,999 cDNA microarrays, Swanson‐Wagner et al. (2006) compared global patterns of gene expression in seedlings of the hybrid (B73 Mo17) with those of its parental genotypes. A total of 1367 ESTs were observed to be significantly diVerentially expressed, using an estimated 15% FDR as cutoV. All possible modes of gene action were observed, including additivity, high‐ and low‐parent dominance, underdominance, and overdominance. A total of 1062 of the 1367 ESTs exhibited expression patterns that are not statistically distinguishable from additivity, while the remaining 305 ESTs exhibited nonadditive gene expression. About 181 of the 305 nonadditive ESTs exhibited high parent dominance, 23 ESTs showed low parent dominance, while 44 ESTs displayed underdominance or overdominance. These results suggest that multiple genetic mechanisms, including overdominance, contribute to heterosis. This contrasts with previous studies that reported heterosis was due to gene action of only a small set of maize genes (Auger et al., 2005; Guo et al., 2004; Song and Messing, 2003). Further analysis of allelic variation in gene expression in the maize hybrid and its parental lines (B73 and Mo17) identified a subset of 27 genes that are diVerentially expressed in parental lines. When the transcriptional contribution of each allele from the inbred line was analyzed in the hybrid, the majority of the diVerential expression was observed to be due to cis‐regulatory variation, and not due to diVerences in trans‐acting regulatory factors. This suggests a predominance of additive expression and a lack of epistatic eVects, as genes subject to cis‐ regulatory variation are expected to be expressed at mid‐parent, or additive, levels in the hybrids (Stuper and Springer, 2006). Scheuring et al. (2006) used a 57,000 maize gene‐specific long‐oligonucleotide microarray containing about 32,000 genes to study the diVerential gene expression between a maize hybrid and its parental genotypes (B73 and Mo17). Preliminary analysis revealed that at least 800 genes were expressed at two‐ to ten‐fold higher levels in the hybrid than the parent genotypes. Using Massively Parallel Signature Sequencing (MPSS), an open‐ended mRNA profiling technology, of nearly 400 allelic signature tag pairs, Yang et al. (2006) found that 60% of the genes expressed in meristems of hybrid were significantly diVerent in allele‐specific transcript level as compared to the parental genotypes. This suggests an abundance of cis‐ regulatory polymorphisms aVecting hybrid meristem gene expression. Furthermore, when comparing the expression of the same allele in the hybrid versus inbred parents, they found 50% of the genes expressed at a significantly diVerent level. Such diVerences in expression are likely attributed to the eVect of trans‐acting factors that diVer between the hybrid and inbreds. While cis‐ regulatory variation predicts additive expression, trans‐regulation may result in nonadditive expression in the hybrid. Thus, studying the eVect of transcript regulation at an allele‐specific level provides a diVerent level of understanding of gene regulation than focusing on overall expression in the hybrid.
254
S. L. DWIVEDI ET AL.
With the vast genomic and technological resources available in Arabidopsis thaliana and the occurrence of heterosis in many traits (Meyer et al., 2004 and references therein; Syed and Chen, 2005), Arabidopsis may be the best model for investigating the genetic basis of heterosis (Jansen and Nap, 2001). However, it is heterosis in yield which holds the greatest promise in plant breeding; thus, eVorts must also be focused on validating and/or translating findings in Arabidopsis for greater understanding, and ultimately ability to manipulate, the genetic basis of heterosis in crop plants.
B.
FINE‐MAPPING, CLONING, AND PYRAMIDING OF QTL ASSOCIATED WITH IMPROVED AGRONOMIC TRAITS
Many agronomically important traits including yield are controlled by a few to a large number of genes (QTL), each with varying eVects and diVerent levels of GEI, which together confer a trait with continuous phenotypic variation. With the development of high‐density genetic linkage maps based on DNA markers, it is possible to map QTL of large eVect with a high level of resolution (Paterson et al., 1988). However, it is diYcult to identify all genes underlying QTL because the eVects of many are relatively small and easily confounded by environmental conditions. Selfed lines from backcrosses (advanced backcross lines) are a common method of fine‐ mapping of QTL, where phenotypic diVerences can be more readily identified without the confounding eVects of diverse segregating backgrounds (Darvasi and Soller, 1995; Graham et al., 1997; Saito et al., 2001; Yamamoto et al., 1998). Alternately, NIL provides the means to dissect complex traits into simple Mendelian factors. Each NIL varies for a defined genomic segment containing a target QTL in an otherwise uniform genetic background. NILs are produced by repeatedly backcrossing a donor parent with a recurrent parent in combination with MAS. Comparing the phenotypes of NIL with those of the recurrent and donor parents permits an accurate evaluation of the eVects of the target QTL in an adapted background without the confound factor of interaction with other segregating loci. Developing NIL has the added advantage of providing QTL ILs (with elite agronomic backgrounds) with the minimum of deleterious alleles in the vicinity of target QTL (linkage drag) which can then be used in marker‐assisted pyramiding of QTL with diVerent beneficial eVects. NILs are also useful resources for developing large mapping populations for fine‐mapping and map‐based cloning of specific QTL. Thus, NILs are a uniquely powerful means of linking marker identification, QTL gene isolation, and advanced product development. ILs can also be used for fine‐mapping of QTL (Eshed and Zamir, 1995). Peleman et al. (2005) proposed a method to fine‐map multiple QTL in a single population: QTL are mapped in a relatively small population, and a large population of
APPLIED CROP GENOMICS
255
1000 plants or more is used to derive QTL isogenic recombinants (QIRs). This reduces the number of lines required for phenotyping. LD methods for fine‐mapping may also oVer improved accuracy of QTL detection (Bink and Meuwissen, 2004; Grapes et al., 2004). There a very large number of reports in the literature regarding the identification of putative QTL for traits of agricultural importance in many crops. However, only a few studies have succeeded in fine‐mapping and cloning of those QTL. The earliest examples of successful QTL cloning include a major fruit‐weight QTL of tomato (fw2.2), delimited to a segment of cloned DNA (<150 kb) (Alpert and Tanksley, 1996), and QTL for tomato sugar content (Brix9‐2‐5) to a 484‐bp region within an invertase gene (Lin‐5) (Fridman et al., 2000). With advances made in rice genomics, several QTL associated with agronomic traits have now also been cloned, for example, four QTL for heading date—Hd1, Hd3a, Hd6, and Ehd1 (Doi et al., 2004; Kojima et al., 2002; Takahashi et al., 2001; Yano et al., 2000); QTL for grain number (Gn1a) and grain size (GS3) (Ashikari et al., 2005; Fan et al., 2006); QTL for salt tolerance (SKC1) (Ren et al., 2005); QTL for regeneration ability (PSR1) (Nishimura et al., 2005); and QTL for shattering (Sh4 and qSH1) (Konishi et al., 2006; Li et al., 2006b). Hd1, Hd3a, and Hd6 encode orthologues of CONSTANS (CO) and Flowering locus T (FT) and the a‐subunit of casein kinase 2 (CK2), which are well‐characterized factors for flowering or the circadian clock in Arabidopsis (Hayama and Coupland, 2004; Izawa et al., 2003). However, rice Hd1 promotes flowering under short‐day lengths, while Arabidopsis CO promotes flowering in long‐day conditions (Izawa et al., 2003). Gn1a encodes a cytokinin oxidase/dehydrogenase (OsCKX2), an enzyme that degrades the phytohormone cytokinin. Reduced expression of OsCKX2 causes cytokinin accumulation in inflorescence meristems, which increases the number of reproductive organs, resulting in higher grain yield (Ashikari et al., 2005). GS3 encodes a putative transmembrane protein, and a mutation in this gene induces large grain size, suggesting that GS3 might function as a negative regulator for grain development (Fan et al., 2006). SKC1 encodes a sodium transporter involved in regulating Kþ/Naþ homeostasis under salt stress (Ren et al., 2005). Sh4 encodes an unknown protein that when mutated inhibits the normal development of an abscission layer, necessary for shattering (Li et al., 2006b), similarly an SNP in the 5´ regulatory region of the qSH1 gene causes loss of shattering owing to the absence of abscission layer formation in japonica rice (Konishi et al., 2006). The QTL for grain weight, gw3.1 and gw8.1, have been fine‐mapped in rice, the former in the pericentromeric region of chromosome 3 (93.8‐kb region) (Li et al., 2004a) while the latter on chromosome 8 to about 306.4‐kb region between markers RM23201.CNR151 and RM30000.CNR99 (Xie et al., 2006). The former locus has also been fine‐mapped simultaneously by
256
S. L. DWIVEDI ET AL.
three other groups, and it has been cloned using map‐based cloning (Fan et al., 2006). Similarly, another QTL influencing the number of grains per panicle (gpa7) has been successfully delimited to a 35‐kb genome region on rice chromosome 7 (Tian et al., 2006a). Andaya and Tai (2006) have fine‐ mapped a major QTL, qCTS12, for seedling cold tolerance in rice and successfully delimited it to a region of about 55 kb on the short arm of chromosome 12, with OsGSTZ1 and OsGSTZ2 the most likely candidates gene(s) for qCTS12. VRN1 and VRN2 are the main genes involved in the vernalization response in diploid wheat T. monococcum (Dubcovsky et al., 1998; Tranquilli and Dubcovsky, 1999). However, vernalization in hexaploid wheat (T. aestivum) is controlled by the VRN1 locus (Law et al., 1975; Tranquilli and Dubcovsky, 1999). VRN1 is closely linked to MADS‐box genes AP1 and AGLG1 (similar to Arabidopsis meristem genes AP1 and AGL2, respectively) in a 0.3‐cM interval flanked by genes Cysteine and Cytochrome B5. AP1 is a more likely candidate for VRN1 than AGLG1 (Yan et al., 2003). VRN2 has expression patterns opposite to that of VRN1, and is located 0.04 cM from ZCCT1, the most likely candidate gene for VRN2 (Yan et al., 2004a). Fusarium head blight (FHB) is a devastating disease of wheat worldwide. Waldron et al. (1999) detected a major QTL, Qfhs.ndsu‐3BS, contributing to FHB resistance in Sumai 3 and located in the deletion bin 3BS (Liu and Anderson, 2003). When constructing a fine genetic map of the Qfhs.ndsu‐3BS region that spanned 6.3 cM, Liu et al. (2006) placed Qfhs.ndsu‐3BS into a 1.2‐cM region flanked by STS3B‐189 and STS3B‐206, and redesignated it as Fhb1. Only five major QTL diVerentiate maize from teosinte (Doebley and Stec, 1993). Just two QTL confer the major morphological diVerences between maize and teosinte, which have been dissected into single Mendelian loci: teosinte branched1 (tb1) (Doebley et al., 1995, 1997; Wang et al., 1999) and teosinte glume architecture (tga1) (Dorweiler et al., 1993; Wang et al., 2005b). The gene tb1 suppresses lateral branching (leading to apical dominance), whereas tga1 aVects the hardness of the seed testa (hard casing that envelops the seed in its ancestor teosinte); both the genes were important in the evolution of teosinte to the agronomically suitable maize crop. Vgt1 is a QTL involved in the control of the transition of the apical meristem from the vegetative to the reproductive phase (flowering) that was initially mapped to a region of 5 cM on chromosome bin 8.05 (Vladutu et al., 1999). Using PCR‐ based assays for markers flanking Vgt1 and screening of NIL homozygous for independent crossovers near the QTL, Salvi et al. (2002) conclude that Vgt1 is in a 1.3‐cM region between AFLP13 and AFLP14, ca. 0.3 cM away from AFLP 14. For QTL with small eVects, fine‐scale mapping and positional cloning will be very diYcult in the absence of whole‐genome sequence. However, in these cases, reverse genetics may oVer a solution, through functional genomic
APPLIED CROP GENOMICS
257
analysis of candidate genes that underlie QTL. For example, Liu et al. (2004b) identified five candidate defense response (DR) genes that colocated with QTL for resistance to blast disease and were associated with level of blast resistance. QTL pyramiding is an important strategy for rebuilding the outputs from reductionist genomic research into whole traits of value for crop improvement. Once the desirable QTL have been detected, NIL are generated for each QTL in a common elite genetic background, and the eVect of each QTL individually evaluated. The selected NIL containing the most important QTL for the target trait are subjected to pair‐wise crosses to pyramid two or more QTL for one or more target traits. For example, in rice QTL for increased grain number (Gn1) and QTL for reduced plant height [Ph1(sd1)] were pyramided in the Koshihikari background producing a 23% increase in grain yield while reducing the plant height by 20% compared with Koshihikari (Ashikari et al., 2005). Dissecting QTL to simple Mendelian factors, often through reduction to component traits, and developing NIL for evaluation, selection, and subsequent use in marker‐assisted pyramiding present an eVective strategy for molecular breeding of complex traits.
C. EXPRESSION QTL MAPPING Traditional genetic mapping has largely focused on the identification of loci aVecting one, or at most a few, complex traits. Dissection of the genetics underlying gene expression combines large‐scale microarray analyses of expression profiles and conventional QTL mapping of the same segregating population. In this analysis, the expression profiling is considered a quantitative phenotype aVected by multiple genes and environmental factors (Jansen and Nap, 2001). This approach has facilitated the identification of genomic regions [gene expression QTL (eQTL)] associated with transcript variation in coregulated genes and, when correlated with phenotypic data from a quantitative character, has successfully identified candidate genes by colocalizing gene eQTL and trait QTL (Brem et al., 2002; Klose et al., 2002; Rockman and Kruglyak, 2006; Schadt et al., 2003; Wayne and Mclntyre, 2002). The power of a genetic mapping study depends on the heritability of the trait, the number of individuals included in the analysis, and the genetic dissimilarity among them. In experiments involving microarrays and complex physiological assays, phenotyping can be expensive and time consuming and may impose limits on the sample size. A random selection of individuals may not provide suYcient power to detect linkage until a large sample size is reached. Jin et al. (2004) developed an algorithm for selecting a subset of
258
S. L. DWIVEDI ET AL.
individuals solely on the basis of genotype data that can substantially improve sensitivity compared to a random sample of the same size. The selective phenotyping method involves preferentially selecting individuals to maximize their genotypic dissimilarity while also representing phenotyping extremes. Selective phenotyping is most eVective when prior knowledge of the genetic architecture allows us to focus on specific genetic regions. However, it can also provide modest improvements in eYciency when applied on a whole‐genome basis. Selective phenotyping does not reduce the eYciency of mapping as compared to a random sample in regions that have not been exposed to strong selection pressure. In contrast to selective genotyping, inferences based solely on a selectively phenotyped population of individuals are representative of the whole population. Kendziorski et al. (2006) demonstrated the deficiencies of using conventional single or multiple QTL analyses for the eQTL approach. Instead, they proposed a mixture over markers (MOM) model that shares information across both markers and transcripts. Results from simulation studies indicate that the MOM model is the best at controlling false‐positive associations without sacrificing power of detection. Plants exhibit massive changes in gene expression during morphophysiological and reproductive development as well when exposed to a range of biotic and abiotic stresses. These have been observed as diVerences in transcriptional profiles in rice (Bao et al., 2005; Matsumura et al., 2003; Rabbani et al., 2003; Tang et al., 2005; Wasaki et al., 2006; Yang et al., 2004; Zhu et al., 2003), maize (Kollipara et al., 2002; Yu and Setter, 2003; Zinselmeier et al., 2002), wheat (Gulick et al., 2005; Wilson et al., 2004), barley (Ozturk et al., 2002; Ueda et al., 2002, 2004; Walia et al., 2006), chickpea (Boominathan et al., 2004), potato (Nielsen et al., 2005; Rensink et al., 2005), banana (Coemans et al., 2005), and cassava (Fregene et al., 2004). Variation in transcript abundance is now being associated with gene expression using eQTL analysis in an increasing number of crops. For example, Kirst et al. (2004) dissected the genetic and metabolic network underlying variation in growth in an interspecific backcross population of eucalyptus. QTL analysis of transcript levels of lignin‐related genes showed that their mRNA abundance is regulated by two genetic loci, coordinating genetic control of lignin biosynthesis. These two loci colocalize with QTL for growth, suggesting that the same genomic regions are regulating growth, and lignin content and composition. Using a high‐density oligonucleotide array and phenotypically divergent rice accessions and their transgressive segregants, Hazen et al. (2005) measured the expression of approximately half of the genes in rice (21,000) to associate changes in stress‐regulated gene expression with QTL for osmotic adjustment (OA), which is a known mechanism of drought tolerance. A total of 662 transcripts were observed to be expressed diVerentially between the parental lines. Only 12 genes were induced in the low OA parent (CT9993) at moderate
APPLIED CROP GENOMICS
259
dehydration stress levels, while over 200 genes were induced in the high‐OA parent (IR62266). Sixty‐nine genes were upregulated in all high‐OA lines and nine of those genes were not induced in any of the low‐OA lines, of which four could be annotated as followings: sucrose synthase, a pore protein, a heat shock protein, and an LEA protein. Previous conventional QTL mapping using the same two rice accessions showed that the parental genotypes diVered for five of the OA QTL, that two of these QTL are syntenic with other cereal drought stress QTL (Zhang et al., 2001), and a major OA QTL in the same genomic region on rice chromosome 7 is also reported in a diVerent cross (Lilley et al., 1996). Of the 3954 probes that correspond to this part of the chromosome, few showed a diVerential expression pattern between the high‐ and low‐OA lines. Thus, these preliminary results demonstrate the power of integrating quantitative analysis of gene expression data with genetic map information to identify genetic and metabolic networks that would not have been identified through conventional QTL analysis.
D. SIMULATION AND MODELING OF MAS Some of the most agronomically and economically important traits in most crops have quantitative phenotypic variation, are under polygenic control, and are significantly aVected by the environment. Whole‐plant physiology modeling is becoming an increasingly important tool for partitioning complex traits into their components and understanding how those components interact with each other and contribute to the overall trait expression in diVerent environmental conditions. With a commitment to genomic analysis of component traits, whole‐plant physiology modeling provides a critical link between molecular genetics and crop improvement. Crop models with generic approaches to underlying physiological processes (Wang et al., 2002) provide a means to link phenotype and genotype, through simulation analysis, of an in silico or virtual plant (Tardieu, 2003). In this way, it is possible to dissect the physiological basis of adaptive traits and determine their control at whole‐plant level through modeling, and then to use simulation analysis as a predictive decision‐support tool for molecular breeders. The substantial progress in ‘‘omics’’ technologies for high‐throughput data generation allows researchers to create comprehensive datasets on the mechanisms underlying plant growth and plant responses to perturbation. A plant requires information about its environment and interaction with that environment and uses that information to dictate its adaptive responses that result in the plant phenotype. Significant endeavors in the field of whole‐plant modeling are now being directed at understanding genetic regulation and aiding crop improvement (Chapman et al., 2002, 2003; Cooper et al., 2002; Hammer et al., 2002; Wang et al., 2004, 2005c; Yin et al., 2003, 2004).
260
S. L. DWIVEDI ET AL.
QTL mapping allows the dissection of a phenotype into underlying genetic factors, but it has limited ability to predict how QTL detected in one set of environmental factors or management practices will behave in a new set of conditions (Stratton, 1998). Ecophysiological modeling provides an insight into the factors influencing GEI (Tardieu, 2003), but it does help define the genetic basis for diVerences in response to environmental changes. Combining ecophysiological modeling with genetic mapping provides the opportunity for creating a QTL‐based crop physiology model that could be powerful tool for resolving the genetic basis of complex environment‐dependent yield‐ related traits. For example, using this approach, researchers predicted specific leaf area in barley (Yin et al., 1999), stay‐green response to nitrogen in sorghum (Borrell et al., 2001), leaf‐growth response to temperature and water deficit in maize (Reymond et al., 2003), and preflowering duration in barley (Yin et al., 2005). Hammer et al. (2005) explored whether physiological dissection and integrative modeling of complex traits could link the complexity of the phenotype to underlying genetic systems in a way that could enhance the power of molecular breeding strategies in sorghum. This approach was applied to four key adaptive traits (phenology, osmotic adjustment, transpiration eYciency, and stay‐green) using 547 location‐season combinations and 4235 genotypic expression states derived from allelic variation at 15 loci for each of the 547 environments. The environmental characterization and physiological knowledge helped to dissect and explain gene and environment context dependencies in the data and based on estimated gene eVects to simulate a range of MAS breeding strategies. By removing gene and environment context dependencies, it was possible to devise breeding strategies that generated an enhanced rate of yield improvement over several cycles of selection. Similarly, Messina et al. (2006) combined an ecophysiological model (CROPGRO‐Soybean) with a linear model that predicted cultivar‐ specific parameters as function of E‐loci. This approach predicted 75% of the variance in time to maturity and 54% of the variance in yield. This demonstrates that agricultural genomics data can be eVectively used for predicting cultivar performance and refining crop breeding systems. Innovative simulation models bridge the gap between molecular and conventional plant breeding and will inform both strategic research and tactical breeding decisions (www.generationcp.org/sccv10/sccv10_upload/ modelling_links.pdf). The CGIAR Generation Challenge Program (GCP) is supporting several projects on whole‐plant physiology modeling, QTL E analysis, and simulation of molecular breeding programs that will collectively link physiological and genetic models toward the optimization of marker‐ assisted breeding systems for drought tolerance in cereals. Simulation models integrate molecular information about interaction between genes and simpler traits to allow realistic predictions for more complex traits such as drought tolerance and yield. QuGene software platform (Podlich and Cooper, 1998;
APPLIED CROP GENOMICS
261
http://www.uniquest.com.au) defines gene eVects and builds breeding modules to compare breeding eYciencies. For example, using QuGene software, researchers in Australia developed a breeding module for sorghum incorporating physiological constraints that were implemented by linking QuGene to the Agricultural Production System Simulator (APSIM) cropping systems model (Keating et al., 2003; http://www.apsru.gov.au), thus providing a powerful set of programs that can simulate crop breeding line performance in a given environment and extrapolate the eVects of long‐term selection over many breeding cycles and seasons. Another GCP supported project links QuGene/APSIM with QTL data on maize leaf growth under drought. These projects aim to deliver modeling tools into the hands of molecular breeders and other researchers to extend the scope and impact of their use, particularly with respect to molecular breeding of complex traits such as drought tolerance. Developing and implementing a design‐led breeding system for complex traits require enhanced attention to precision phenotyping, ecophysiological modeling, and marker validation to ensure robustness and selective power. These approaches require the iterative and systemic integration of a range of scientific disciplines, including modelers, physiologists, geneticists, breeders, and molecular biologists. Nevertheless, the first preliminary studies reviewed in this section suggest that a new paradigm in knowledge‐led design‐driven plant breeding is a feasible option and that for the first time genomics may finally realize its potential impact on breeding complex traits is increasingly likely.
VIII. THE ROLE OF COMPUTATIONAL SYSTEMS IN MOLECULAR BREEDING PROGRAMS EVective marker‐aided breeding requires the balance of many diverse elements in order to provide the best compromise between time, cost, and genetic gain: Identify beneficial genetic variation and develop robust marker‐trait
associations EVectively manage and manipulate large amounts of genotype, pedigree,
and phenotype data Select desirable recombinants through an optimum combination (in time
and space) of phenotypic and genotypic data Develop breeding systems that minimize population sizes, number of gen-
erations, and overall costs while maximizing genetic gain for traditional and novel target traits
262
S. L. DWIVEDI ET AL.
In general, MAS works best with simply inherited markers that are inside or flanking markers that are in proximity to the genetic factors aVecting monogenic, oligogenic, and polygenic traits. The journey from the phenotyping‐and‐genotyping of individuals from genetic populations to the identification of marker‐trait associations and onto the application of markers in molecular breeding depends on the sequential use of a number of decision‐support tools that facilitate communication between genomics scientists, geneticists, bioinformaticians, trait specialists, and breeders. In this section, we provide an overview of key decision‐support tools for assisting germplasm evaluation, breeding population management, GEI, genetic map construction, marker‐trait linkage and association analysis, marker‐assisted application, breeding system design and simulation, information management, and other integrated tools needed to support molecular breeding programs (Table XVI).
A. GERMPLASM EVALUATION Marker‐assisted germplasm evaluation (MAGE) aims to complement phenotypic evaluation by helping define the architecture of genetic resources and by identifying germplasm that contains alleles associated with traits of economic importance. Molecular markers can be used to for characterization based on genes, genotypes, or genomes, which provide more accurate and detailed information than classical phenotypic or passport data. Many features revealed by molecular markers, such as unique alleles, allele frequencies, and heterozygosity at marker loci, mirror the genetic structure of germplasm resources and will lead to the identification of useful genes and their transfer into well‐adapted cultivars. MAGE will play an important role in acquisition, distribution, maintenance, and use of germplasm (Bretting and Widrlechner, 1995; Xu, 2003). During germplasm evaluation, molecular markers can be used to (1) diVerentiate cultivars and construct heterotic groups; (2) identify germplasm redundancy, underrepresented alleles, and genetic gaps in current germplasm collections; (3) monitor genetic shifts that occur during germplasm domestication, storage, regeneration, and breeding; (4) screen germplasm for novel and/or superior genes or alleles; and (5) construct a representative subset or core collection (Xu et al., 2003, 2004a). Although computational programs are available for all relevant analyses including computer simulation and resampling (Xu et al., 2004a), a fully integrated, user‐friendly graphical program is needed to bring all these functions together to facilitate decisions through all aspects of germplasm evaluation. Several software packages, such as Statistica, JMAP, SAS, NTSYS, GeneFlow, can be used for the analysis of germplasm evaluation data. This
Table XVI List of Decision Support Tools to Support Molecular Breeding Programs Tool
Clustering, PCA Identify distinct populations and estimate allele frequencies Transform marker data into simple colorful chromosome drawings
http://www.sas.com/ Pritchard et al., 2000a van Berloo, 1999
Classify cultivars and construct heterotic groups; identify germplasm redundancy, underrepresented alleles, and genetic gaps; monitor genetic shifts; screen for novel/superior genes (alleles); construct a representative subset or core collection
Xu et al., 2004a
Breeding population management Hybrid performance BLUP‐based methods prediction Genetic map construction MAPMAKER/EXP MAPDISTO MAP MANAGER CLASSIC JOINMAP GMendel
References
APPLIED CROP GENOMICS
Germplasm evaluation JMAP/SAS Structure GGT (Graphical GenoTypes) GERMPLASM
Function
Bernardo, 1994, 1996
Build linkage map from molecular marker data Build linkage map from molecular marker data with distorted segregation A graphic, interactive program for linkage map construction
Lander et al., 1987 http://mapdisto.free.fr/ Manly, 1993
Combine data derived from several sources into an integrated map Linkage mapping using simulated annealing and multiple pair‐wise methods for F2, BC, DH, RIL, and any generations of SSD
Van Ooijen and Voorrips, 2001 http://www.maizegdb.org/mnl/66/ 45echt.html Lander et al., 1987 Manly and Olsen, 1999 Manly et al., 2001
263
Genotype‐phenotype association MAPMAKER/QTL Map QTL using interval mapping, dealing with simple QTL and several standard populations MAP MANAGER QT A graphic, interactive program for QTL mapping by regression methods MAP MANAGER QTX A graphic, interactive program for QTL mapping using intercross, BC or RIL in plants or animals
(continued )
264
Table XVI (continued ) Tool QTL Cartographer PLABQTL QTL EXPRESS
MCQTL EPISTACY STRAT TASSEL
BQTL (Bayesian QTL mapping) MAS Plabsim
Popmin
QTL mapping using several interval mapping methods with permutation tests to estimate QTL thresholds Identifying QTL using composite interval mapping and QTL environment interaction analysis QTL mapping in outbred populations including line crosses, half‐sib families, nuclear families and sib‐pairs, with permutation tests to determine empirical significance levels and boots‐trapping to estimate empirical confidence intervals of QTL locations Identify QTL using selective and bin mapping by choosing good samples from mapping populations and for locating new markers on preexisting maps QTL mapping using multicross designs A SAS program to test for all possible two‐locus interaction eVects on a QTL using least squares methods Association mapping with incorporated function for structure analysis A comprehensive software for trait analysis by association, evolution, and linkage, including association mapping, diversity estimation and calculating linkage disequilibrium Maximum likelihood estimation of multigene models; Bayesian estimation of multigene models via Laplace Approximations; and interval mapping and composite interval mapping of genetic loci MAS simulation for all common breeding methods. Selection can be carried out at defined loci or for selection indices calculated from allele frequencies at several loci. The simulated data can be analyzed for genetic parameters such as population size, marker density and positions, and selection strategies Numerical optimization of population sizes in marker‐assisted backcross programs
References http://statgen.ncsu.edu/qtlcart/ cartographer.html Utz and Melchinger, 1996 Seaton et al., 2002
Vision et al., 2000 Jourjon et al., 2005 Holland, 1998 Pritchard et al., 2000b Zhang et al., 2006b
Borevitz et al., 2002
Frisch et al., 2000
Hospital and Decoux, 2002
S. L. DWIVEDI ET AL.
MapPop
Function
BCSIM
Simulation for evaluation of marker‐assisted backcross programs
Information management and integrated tools CMTV Display syntenic regions across taxa, combine maps from separate experiments into a consensus map, or project data from diVerent maps into a common coordinate framework using dynamic coordinate translations between source and target maps QTLFinder Integrate QTL and linkage maps into a consensus map; do QTL meta‐analysis and show colocations; construct comparative map of interspecies (or intraspecies) genomes; and compare collinearity of same or similar traits across genomes ICIS (International Crop Kink the gene, gene value, and target environment data with the uniquely Information System) identified germplasm units used and manipulated in breeding programs. It has ICIS as the Genealogy Management System (GMS) to manage data on nomenclature, origin, development and deployment of germplasm and the Data Management System (DMS) to manage and document characterization and evaluation data iMAS (integrated decision Facilitate an integrated, error‐free, and appropriate data analysis from the support system for beginning to end of the molecular breeding pathway, including experimental marker‐assisted plant design, biometric analysis of phenotypic data, linkage and association breeding) mapping, linkage map construction, and MAS
Podlich and Cooper, 1998 Wang et al., 2004
Wang et al., 2004
Sawkins et al., 2004
Yan et al., personal communication
http://www.icics.cgiar.org:8080/
265
http://www.generationcp.org/vw/ Download/ Commisioned_Research_2005/ 33_SP4_MAS.pdf#search ¼‘iMAS% 20marker%20assisted%20selection’
APPLIED CROP GENOMICS
Breeding design and simulation QU‐GENE (QUantitative Simulation platform for quantitative analysis of genetic models including GENEtics) genotype by environment interaction analysis QuCim Identify the best crosses and breeding strategies from mass selection, pedigree system, bulk population system, backcross breeding, top cross (or three‐way cross) breeding, DH breeding, MAS, and many combinations and modifications of these methods QuLine Define genetic models from simple to complex based on simulation experiments to optimize breeding programs and improve breeding eYciency
http://www.dpw.wau.nl/pv/pub/bcsim/ index.htm
266
S. L. DWIVEDI ET AL.
includes the use of principal component or coordinate analysis to identify distinct groups or populations, and for cluster or structure analysis to define population structure. For example, STRUCTURE (Pritchard et al., 2000a) uses multilocus genotype data to investigate population structure, assign individuals to populations, study hybrid zones, identify migrants and admixed individuals, and estimate population allele frequencies in situations where many individuals are migrants or admixed. It can be applied to datasets from most of the commonly used genetic markers, including SSR, RFLP, and SNP.
B. MANAGING BREEDING POPULATIONS Decision‐support tools to help the management of breeding populations are needed to assist in the choice of parental lines, types of crosses, and nature of breeding system. Computational tools may also assist in the establishment and maintenance of heterotic groups, the selection of lines for creation of a synthetic cultivar, the prediction of progeny and hybrid performance, and the monitoring of genomic profiles during population improvement. Genotyping parental lines on a genome‐wide scale, especially when gene‐ based markers are available, provide an opportunity for establishing parent– hybrid performance relationships at the molecular level. Genome‐wide heterozygosity and specific combinations of alleles (linkats) may be useful determinants in some crops for maximizing heterosis and hybrid vigor. Melchinger and Gumber (1998) used a multistage procedure to identify heterotic groups, which consists of the following steps: (1) grouping the germplasm based on genetic similarity, (2) selection of representative genotypes (e.g., two or four lines or one population) from each subgroup for producing diallel crosses, (3) evaluation of diallel crosses among the subgroups together with parents, and (4) selection of the most promising cross combinations as potential heterotic patterns. The ability to use molecular markers to predict hybrid performance would greatly enhance the eYciency of hybrid breeding programs. Development of a reliable method for predicting hybrid performance or heterosis without generating and testing hundreds or thousands of single cross combinations has been the goal of numerous studies using marker data and combinations of marker and phenotypic data, particularly in maize and rice. The best linear unbiased prediction (BLUP) procedure has been used for decades for evaluating the genetic merit of animals, especially dairy cattle. Intrapopulation, additive genetic models have traditionally been used for BLUP in animal breeding (Henderson, 1975). Bernardo (1994, 1996) used BLUP in maize breeding with interpopulation genetic models that involve both
APPLIED CROP GENOMICS
267
general combining ability and specific combining ability and found that BLUP is useful for routine prediction of single‐cross performance. The predicted performance of single crosses may subsequently be used to predict the performance of F2 tester combinations, three‐way crosses, or double crosses. Along with the pedigree relationship, BLUP can use trait data, or both trait and marker data, for prediction. A synthetic cultivar is developed by intercrossing selected clones or inbred lines, with seed production of the cultivar through open‐pollination. MAS can be used to develop synthetic cultivars by mixing inbred lines that have been bred by MAS or by mixing individual plants derived from any stage of MAS. With genotypic information available across the whole genome for all the selected individuals or inbred lines, synthetic cultivars can be created to contain complementary genotypes, fixed heterozygosity, and the best combinations of genetic structure.
C. GENETIC MAP CONSTRUCTION Genetic maps can be constructed using segregating populations of diVerent types, which have diVerent advantages depending on the species and level of polyploidy. MAPMAKER/EXP is the most frequently used software for map construction (Lander et al., 1987). Various maps can be generated based on populations derived from diVerent crosses or the same population evaluated in diVerent environments. These maps can be integrated into a single or consensus map. JOINMAP is used to construct genetic linkage maps for several types of mapping populations. It can combine (join) data from several sources into an integrated map, with several other functions, including LG determination, automatic phase determination for outbred full‐sib family, several diagnostics, and map charts (Van Ooijen and Voorrips, 2001). GMendel uses simulated annealing and multiple pair‐wise methods for locus ordering. All markers within an LG are used simultaneously to estimate a locus order that provides maps equivalent to those found by MAPMAKER and JOINMAP. It can be used to build maps using F2, backcross, DHL, RIL, and in any generation of SSD lines. Other software packages in use are MAPDISTO (http://mapdisto.free.fr/) and MAP MANAGER CLASSIC (Manly, 1993) that perform specific functions.
D. IDENTIFYING MARKER‐TRAIT ASSOCIATIONS Establishing a highly significant genotype–phenotype association is one of the prerequisites for MAS. Linkages or associations between target traits or genes and molecular markers are detected based on genetic linkage or
268
S. L. DWIVEDI ET AL.
assocition mapping experiments. Decision‐support tools required for genotype–phenotype association include (1) statistical methods and tools to establish, validate, and compare genotype–phenotype associations through linkage mapping, LD, or AM, and in silico mapping, using single or multiple genetic populations, genetic resources, or breeding populations; (2) statistical methods and tools for identification of genetic background eVects, QTL alleles at multiple loci, and multiple alleles at a single locus; (3) tools facilitating the validation of candidate gene markers with linked markers in order to generate functional markers; and (4) tools facilitating management of genetic populations, linkage maps, and related data. A widely used QTL mapping software is QTL Cartographer (http://statgen.ncsu.edu/qtlcart/ cartographer.html), which implements several statistical approaches to analysis of multiple marker data including composite interval mapping (CIM) and multiple interval mapping. The interaction between diVerent QTL can also be estimated. Another populated QTL mapping software is PLABQTL that uses CIM with many functions common to those of QTL Cartographer. QTL can be localized and characterized in populations derived from a biparental cross. Simple interval mapping (SIM) and CIM are performed using a fast multiple regression procedure. PLABQTL can also be used to analyze QTL environment interactions (Utz and Melchinger, 1996). For mapping with populations from outbreeding species, QTL EXPRESS can be used to map QTL using line crosses, half‐sib families, nuclear families, and sib‐pairs (Seaton et al., 2002). EPISTACY is a SAS‐based program which can test pair‐wise epistatic (interaction) eVects on a quantitative trait using QTL‐mapping datasets (Holland, 1998). Other softwares for mapping QTL include MAPMAKER/QTL (Lander et al., 1987), MAP MANAGER QT (Manly and Olsen, 1999), MAP MANAGER QTX (Manly et al., 2001), MapPop (Vision et al., 2000), and MCQTL (Jourjon et al., 2005). Software packages are also now available for mapping genetic traits using Bayesian approaches. For example, BQTL performs (1) maximum likelihood estimation of multigene models, (2) Bayesian estimation of multigene models via Laplace approximations, and (3) interval mapping and CIM of genetic loci (Borevitz et al., 2002), while BLADE is used for Bayesian analysis of haplotypes for LD mapping (Liu et al., 2001b; Lu et al., 2003). AM or LD mapping, using unstructured populations, is gaining increasing credibility over traditional QTL mapping using genetic populations (see Section II.D). However, softwares are needed that analyze and remove the eVect of population structure. STRAT uses a structured association method for AM, enabling valid case‐control studies even in the presence of population structure (Pritchard et al., 2000b). The software TASSEL has been released, which performs a variety of genetic analyses, including AM, diversity estimation, and LD analysis (Zhang et al., 2006b). The association
APPLIED CROP GENOMICS
269
analysis between genotypes and phenotypes can be performed by either a general linear model or a mixed linear model. The general linear model allows users to analyze complex field designs, environmental interactions, and epistatic interactions. The mixed model is especially designed to handle polygenic eVects at multiple levels of relatedness, including pedigree information. These new analyses should permit association analysis in a wide‐range plant and animal species.
E. MARKER‐ASSISTED SELECTION Many factors influence the eYciency of MAS in plant breeding programs (see Section VI.A and B). Decision‐support tools are needed to determine sample size for foreground and background selection, for estimation of genetic gains (response to selection), for construction of selection indices for multiple traits and whole‐genome selection, for estimation and graphical display of RGC of selected individuals at each generation of introgression, for identification of desirable plants based on both phenotype and genotype information, for cost‐benefit analysis, and for marker‐aided simulations studies. There has been much interest in the development of software that simulates MAS using genetic models. Early eVorts had somewhat limited results, for example, GREGOR simulates MAS based only on predefined genetic linkage maps, and is thus restricted in its value for simulation of MAS in breeding programs (Tinker and Mather, 1993). More recently, Plabsim was developed for the simulation of MAS programs, with the following features: (1) simulations can be made for any diploid genome with an arbitrary number of loci at arbitrary positions on an arbitrary number of chromosomes; (2) the implemented reproduction schemes include all common breeding methods; (3) an arbitrary number of selection steps can be combined with a specified selection strategy and selection can be carried out for genotypes at defined loci, or for selection indices calculated from allele frequencies at several loci; and (4) the simulated data can be analyzed for a broad range of genetic parameters including population size, marker density and positions, and selection strategies on the genetic composition of the breeding product and on the required number of marker data points (Frisch et al., 2000). Other software packages related to MAS include Popmin for the numerical optimization of population sizes in marker‐assisted backcross programs (Hospital and Decoux, 2002), GGT for displaying molecular marker data into simple colorful graphical representations of chromosome haplotypes (van Berloo, 1999), and BCSIM for evaluation of marker‐assisted backcross programs (http://www.dpw.wau.nl/pv/pub/bcsim/index.htm).
S. L. DWIVEDI ET AL.
270
F.
GEI ANALYSIS
Computational tools are needed to assist in dealing with many complex issues related to the eVect of the environment, particularly regarding complex traits, including: To separate genetic (G) eVects from the environment (E) and GEI
interaction To incorporate environmental and genotypic variables into statistical
models to explain GEI To define target populations and genotypes for a given environment To determine subsets of genotypes and sites with negligible crossover
eVects to identify subgroups of sites and genotypes with similar response to maximize response to selection To develop selection indices using phenotypic and marker data to select the best genotypes To study genetic diversity of crop genotypes associated with the target traits and perform AM To study gene expression under target conditions using microarray technology Podlich and Cooper (1998) developed QU‐GENE software for carrying out quantitative genetic analyses of GEI in crop breeding and this has become an increasingly widely utilized decision‐support tool in breeding programs. Statistical models have been refined in order to incorporate pedigree information (or coeYcient of parentage) among genotypes when modeling GEI (Crossa et al., 2006). It is likely that these will soon be further refined using whole‐plant physiology models.
G.
BREEDING DESIGN AND SIMULATION
The major objective of plant breeding programs is to develop new cultivars superior to those currently available in a given target production environment (TPE). Designing eVective breeding systems requires information about target genes, donor germplasm, and proposed elite recurrent parents. This can then be combined with evaluation data on the target biological characteristics, breeding objectives for the TPE, in order to optimize the breeding procedure and selection methods through modeling and simulation analysis. This type of analysis will also predict the desirable target genotype and the probability of successfully generating new cultivars through the proposed breeding system. QU‐GENE, a simulation platform based on quantitative genetic models, facilitates the simulation of actual breeding programs through its two‐stage process (Podlich and Cooper, 1998).
APPLIED CROP GENOMICS
271
The genetics and breeding simulation tool (QuLine and QuCim) has the potential to utilize vast and varied genetic information. QuLine is capable of defining genetic models ranging from simple to complex inheritance. QuCim can be used to identify the best crosses and breeding strategies by predicting cross performance and comparing diVerent selection methods. Using simulation experiments, breeders may optimize their breeding programs and thereby greatly improve the breeding eYciency (Wang et al., 2004). Almost all eVorts in this field have been focused on genetic models, thus none provides the facility to carry out such as cost benefit analysis or integrate whole‐plant physiology models.
H.
INFORMATION MANAGEMENT AND INTEGRATED TOOLS
Crop informatics has become a prerequisite in molecular breeding because breeding‐related information is increasing at such a high rate that collecting, storing, mining, and manipulating such a large amount of information for selection decisions would not be possible without appropriate statistical, biometrical, and informatics tools. An integrated breeding tool is therefore needed to rapidly collect, analyze, and represent breeding‐related data in the short‐time window available for most selection decisions. In addition, computational tools are required to translate and integrate research outputs into a usable form for plant breeding programs. International Crop Information System (ICIS) is open‐source community developed software that has been evolving over many years. ICIS can link gene, gene value, and target environment data with the uniquely identified germplasm units used and manipulated in breeding programs (http://www. icis.cgiar.org:8080/). ICIS has a modular structure with a core consisting of Genealogy Management System (GMS) that manages data on nomenclature, origin, development, and deployment of germplasm and the Data Management System (DMS) that manages and documents characterization and evaluation data. Specialized user interfaces deliver data views and decision‐support tools to crop scientists from diVerent disciplines, which can access common data resources leading to eYcient use and reuse of research data. ICIS databases tailored to diVerent crops are also being developed for separate ICIS implementations. ICIS has also embedded a parallel structure of central and local versions that provides local read/write capabilities, allowing data generated locally to be merged and harmonized with the central database at the local user’s discretion. Some of the issues that need to be further integrated into ICIS to meet breeding requirements include: (1) a database for all environmental characterization data such as climate, soil, and abiotic stress information; (2) data‐mining tools for all breeding purposes such as GEI and identification of novel alleles
272
S. L. DWIVEDI ET AL.
and genetic variation; (3) modeling breeding processes and selection schemes using multiple sources of breeding information to eliminate some field and laboratory tests required for making selection decision, which may be critical for complex traits; and (4) linkage to major public databases with appropriate data comparison and mining tools to enable extraction of useful information through comparative analysis of the specific breeding program data with global research outputs. Researchers need eYcient and intuitive tools to help identify common genomic regions, and, where possible, specific genes involved in influencing the expression of target traits across diverse germplasm and growing conditions. Sawkins et al. (2004) developed the comparative map and trait viewer (CMTV) that can help integrate various kinds of genomic maps. Its major strength is in the comparative display of LGs or chromosomes across diVerent species, populations, or evaluation environments and link information associated with diVerent objects on the maps. These correspondences could then be displayed as graphical lines linking corresponding loci between maps in order to illustrate syntenic relationships. Alternatively, they could be used to construct a consensus map using these common markers as anchors from which the positions of other markers could be interpolated. However, the current version of this software stops short of being able to carry out combined analysis across the maps to be compared. In contrast, QTLFinder can carry out this type of QTL meta‐analysis. This software integrates QTL from separate experiments and linkage maps into a consensus map. QTLFinder can also construct comparative maps across species using sequence similarity, and compare the colinearity of same or similar traits across genomes (Jianbing Yan, China Agricultural University, Beijing, personal communication). An integrated decision support system for marker‐assisted plant breeding iMAS (a GCP‐supported software) will be released by the end of 2006 (Subhash Chandra, ICRISAT, personal communication), is expected to assist the development and application of marker‐assisted plant breeding by integrating the best freely available quality software required for the journey from phenotyping‐and‐genotyping of individuals to identification and application of trait‐linked markers. iMAS will provide simple‐to‐understand‐and‐use online decision‐support guidelines to help the user correctly use this software, and correctly interpret the outputs. Software identified for inclusion in iMAS includes IRRISTAT (for experimental design, biometric analysis of phenotypic data, and AM), GMendel and MapDisto (for linkage map construction), PlabQTL and QTL Cartographer (for QTL analysis), PopMin (for estimating sample size for foreground and background selection), GGT (for estimation and display of RGC of selected individuals), and TASSEL (for AM). Many support tools are available for use with functional genomic data, but these are yet to be fully explored for direct application in breeding
APPLIED CROP GENOMICS
273
programs. These tools are for sequence comparison, handling and analysis, microarray data treatment and analysis, motif alignment and search, and comparative genomics. Various softwares such as EHAP (http://wpicr.wpic. pitt.edu/WPICCompGen/ehap__v1.htm), DPPH (Bafna et al., 2003), HAPLOVIEW (Barrett et al., 2005), HAPLOT (Gu et al., 2005), HAP (1) (http:// research.calit2.net/hap/), and HAP (2) (Zhao, 2004) have been developed for haplotype analysis using SNP data. It is likely that these approaches will soon be widely used by molecular breeders across diverse crops as sequence and expressional data become increasingly available.
IX. FUTURE PROSPECTS FOR THE MOLECULARIZATION OF PUBLIC CROP IMPROVEMENT Plant breeding is the science, art, and business of improving plants for human benefit (Bernardo, 2002) The rate, scale, and scope of uptake of genomics in crop breeding programs have continually lagged behind expectations. This is little diVerent to the adoption of quantitative genetics, mechanization, and computerization during the last century. This is partly due to the long product development cycle in plant breeding and in turn the long‐term nature of feedback from the market regarding the impact of any changes in the cultivar development pipeline. Thus, although molecularization of plant breeding is the fourth natural paradigm shift for crop improvement programs, we must assume that the introduction of MAS and the breeding with transgenic germplasm will be a gradual stepwise process. At the same time, there is considerable and immediate need for computational tools to help breeders more eVectively translate and integrate the outputs from bioscience research and to help eYciently select the best technology interventions and associated breeding systems for their target traits and markets. With the availability of comprehensive and robust facilitating and decision‐support tools, it is expected that plant breeders will become much more responsive to the emergence of new technologies. Polymorphic DNA markers and genetic maps are now available for most important food crops, albeit in varying numbers and levels of genomic saturation (Tables VI–VIII and X; Dwivedi et al., 2005). Similarly, the genetic control of many agronomic traits is well characterized in many crops, and marker‐trait linkages have been reported for a diverse array of traits in a large number of crops (Section II). A critical mass of reports of
274
S. L. DWIVEDI ET AL.
validation in diVerent genetic backgrounds and environments is naturally only just beginning to emerge (Section III; Tables XI and XII). Nevertheless, MAS is now being practiced in most well‐researched crops (Section IV.A–D; Tables XIII and XIV). In the private sector, molecular breeding applications are still dominated by MAS for transgene introgression and to a lesser extent for backcross programs for simple traits (Section V). Thus, only a very small proportion of marker genotyping is currently being used for complex traits where it has been long since highlighted that MAS will have its greatest impact. In the short term, we expect the greatest growth in MAS of mono‐ and oligogenic traits that are diYcult or expensive to screen using conventional phenotyping methods (Section VI). In the medium term, we envisage that a number of emerging technologies will facilitate a gradual shift from MAS for individual simply inherited traits to more holistic molecular breeding strategies (Section VII). It is only at this point that we expect to see a significant increase of interest in the application of MAS for polygenic traits. However, there are a number of technical and logistical hurdles that must be overcome before genomic tools can assist the breeding of such complex targets. Traditionally, the heritability of quantitative traits was the most common predictor of genetic gains for diVerent plant breeding methods. DNA markers may be used today to accelerate and enhance overall breeding methods by combining DNA marker and phenotyping data in a selection index. The best current success stories of MAS in plant breeding tend to focus on traits that are diYcult to screen and controlled by one or few genes. However, more recently there have been a number of successes in pyramiding a range of diVerence sources of biotic and abiotic stress resistances (Table XIV). This engenders hope for the potential of MAS to improve important quantitative traits, particularly when accelerating the use of new sources of variation in elite germplasm. DNA markers will also be useful tools for early testing. However, geneticists and plant breeders will still need to deal with LD while using MAS in recurrent selection, especially when using polymorphic markers arising from mapping populations, which tend to be from diverse parents, and thus may not be relevant for target breeding materials. The power of MAS will also continue to rely heavily on the accuracy and precision of phenotyping, and the characterization and evaluation of germplasm in the field. Issues such as the error term to test for the significance of a QTL, detecting small eVects with narrow genetic variance, or the number of QTL not related to genetic variance or divergence of parents are all under‐researched areas that need priority attention by geneticists. Addressing these issues will allow plant breeders to define the optimum number of individuals/lines and markers to be used in their MAS programs. Plant breeders are ready to apply MAS for quantitative traits when the genetic gain and time or cost eYciency from doing so are clearly higher than
APPLIED CROP GENOMICS
275
through conventional selection methods. Initial emphasis in this area should be on traits for which a robust cost‐eVective phenotyping system is not available for the target trait. To quickly reach this stage requires a paradigm shift in strategy among the marker‐trait identification community: from eVorts to identify all QTL influencing the target trait to a focus on identification of a few QTL having the largest eVect on the target trait. QTL of major eVect may be easier to detect (in the right genetic material), and be less influenced by GEIs and genetic background eVects. Of great importance will be a shift away from analysis of entire genetic populations to an emphasis on selected individuals with extreme phenotypes from relevant breeding populations and genetic stocks and likely, pooled DNA analysis using the selected individuals. Of equal importance will be a shift from linked markers to diagnostic gene‐based markers, which will generally be SNP based and thus readily scalable for high‐throughput haplotyping. Detailed cost‐benefit analysis of various elements of DNA marker development and application, including the cost of the required genotyping platforms and professional expertise, needs to be assessed at the earliest possible stage. This is particularly important at this time when most public plant breeding programs are not adequately funded or poorly equipped to reach a critical threshold of marker assay throughput. Molecular breeding consortia accessing joint venture genotyping hubs or commercial service providers appear to be an increasingly realistic option where those facilities can provide the right quality, quantity, and timelines of service to fit the given breeding system. In the last decade, computational tools have rapidly evolved to provide solutions for the data acquisition, management, analysis, and visualization needs arising from the development and widespread use of high‐throughput genomics technologies. Plant breeders expect that informatics will assist with the development of diagnostic tools for identification of the best breeding systems, optimization of the best crosses, and selection of the best ensuing segregating progeny. Likewise, bioinformatic research should identify causative alleles and estimate breeding values or relative risks in the context of breeding populations. Moreover, besides assisting with candidate genes, bioinformatics should provide plant breeders with information regarding LD and epistatic and pleiotropic eVects of the allele in the target breeding population. Statistical methods will assist in estimating and predicting allele eVects which should be updated as the alleles are assessed in distinct breeding backgrounds and across other environments. Information on breeding values provided by DNA markers may enable identification of DNA markers for further use in a more robust MAS system. Geneticists can use DNA markers to dissect complex epitasis eVects, which may arise as an outcome of selection‐induced variation. For example, a minor or neutral QTL may become a major QTL when selection brings changes that create the most appropriate genetic background for interaction
276
S. L. DWIVEDI ET AL.
with that target QTL. The release of genetic variability through capitalizing on epistasis may allow a more extended response to selection than that currently resulting solely from additive variance. Genotype‐by‐environment interaction (GEI) occurs when the eVects of the environment, the genotype or both, are nonadditive. GEI may lead to divergence, convergence, or crossover performance of genotypes across the environments; that is, the distinct performance among genotypes depends on the environment (location, year, cropping season). Linear mixed models are used for modeling GEI and assisting the grouping of environments and genotypes. Factorial and partial least squares regressions incorporate external environmental and genotypic covariables directly into the model. These are useful tools for gaining more insight into the genetics of the target trait by adding molecular marker data associated with quantitative trait variation in the model for interpreting GEI. With more and more information accumulating from genotyping and phenotyping, integration of these diverse datasets with environmental characterization data will help establish genetic models for GEI and apply them to crop improvement. Molecular markers could further explain some of the GEI variabilities and assist in breeding for low‐heritability traits. For example, Paterson et al. (1991) suggested that, for a low‐heritability trait such as soluble solids in tomato, the phenotype of F3 progeny could be predicted more accurately from the QTL genotype of the F3 parent than from the phenotype of the F2 individual. Applied genomic tools are being used to unravel the molecular mechanism of heterosis, classifying germplasm with distinct heterotic groups, predicting hybrid performance, understanding the relationships between heterozygosity and genetic distance with hybrid performance and heterosis. All these will lead to a better understanding of the genes regulating the network of diverse physiological pathways that control the expression of hybrid vigor. This will undoubtedly lead to enhanced use of MAS for the development of superior yielding hybrids. So far, various hypotheses have been proposed to explain the genetic mechanisms of heterosis, each being supported to some extent by diVerent experimental data. Considering that heterosis may mediate its eVect at various levels and developmental stages for diVerent traits, it is feasible that there is no single genetic model or hypothesis that can be used to explain all heterotic eVects observed in hybrids across traits, crops, and breeding systems (Xu, 2003). Molecular markers will provide new insights into heterosis as it becomes feasible to carry out genome‐wide analysis of parental lines across large numbers of hybrids, germplasm accessions, and breeding materials. Plants exhibit massive changes in gene expression during morphophysiological and reproductive development as well as when exposed to a range of biotic and abiotic stresses (Section VII.C). A new field of genetics of global gene expression has emerged based on the application of traditional
APPLIED CROP GENOMICS
277
techniques of linkage and association analysis for the thousands of transcripts measured by microarrays. Dissecting the architecture of quantitative traits in this way connects DNA sequence variation with phenotypic variation, and is improving our understanding of transcriptional regulation and regulatory variation (Rockman and Kruglyak, 2006). A range of decision‐support tools are needed to facilitate communication among scientists involved in diVerent elements of the crop improvement product development pipeline. While there are a number of computational tools to carry out various functions in the research domain, it is essential that these tools are integrated into a common platform to assist their eVective deployment in crop improvement. iMAS (www.generationcp.org), an integrated decision support systems for marker‐assisted plant breeding, is a preliminary attempt to create a publicly available computational platform to assist the development and application of marker‐assisted plant breeding. iMAS currently integrates freely available software for the journey from phenotyping‐and‐genotyping of individuals to identification and application of trait‐linked markers. iMAS also provides simple‐to‐understand‐and‐use online decision‐support guidelines to help the user correctly operate these softwares, and correctly interpret the outputs. It has been argued that genetically modified food is the next great scientific and technological revolution in agriculture and the only eYcient and cheap way to feed a growing population in a shrinking world. Genetic transformation is particularly important for transfer of genes from distant species. In many cases, genetic transformation will be the only mechanism for harnessing the outputs of large‐scale whole‐genome research, particularly in model systems. At the same time, rapidly accumulating information about crop genomes is allowing scientists to identify genes associated with beneficial traits in ‘‘crop relatives.’’ Marker‐assisted introgression of these beneficial alleles into existing cultivars will be increasingly critical for eYcient use of exotic genetic variation in breeding programs. Thus, the intimate integration of MAS and genetic transformation approaches in field breeding programs will be an important challenge for the future success of public sector crop improvement. Using molecular biology tools and outputs, researchers will be able to broaden the scope of breeding goals, improve the rate and precision of genetic gain toward specific trait targets, and significantly reduce the time needed to breed new cultivars. However, there is still much work to be done in understanding the ‘‘choreography’’ of molecular breeding to the extent required to reach a knowledge‐led design‐based plant breeding paradigm. For example, the relationship between single genetic loci, complex genetic traits, and environmental factors all diVerentially interact to aVect the development of the plant, its response to biotic and abiotic stresses, and ultimately the yield. Over the next decade, MAS technologies will become cheaper and easier to apply at large scale, and knowledge from genomics
278
S. L. DWIVEDI ET AL.
research will become more readily translated into breeding tools and integrated into breeding systems. These advances will empower plant breeders around the world to use molecular breeding approaches as part of a much larger systemic and holistic approach to sustainable agricultural development (http://www.washingtonpost.com/wp‐dyn/content/article/2006/07/ 03/AR2006070300922.html). Plant breeders in the twentieth century accomplished improvements in crop performance through knowledge and application of scientific advances in genetics research. However, a substantial proportion of genetic progress also resulted from pragmatic practice of the art of plant breeding. The crop genetic enhancers in this twenty‐first century will harness the outputs of bioscience research (especially genomics) in order to address the challenge of doubling food production sustainable on same land area (1.5 billion ha) by 2050. To substantially contribute to achieving this goal, it will be necessary to build holistic knowledge and implementation systems to understand, predict, and manipulate the interaction of genes and gene networks. This should lead to the eYcient improvement of a wide range of important agronomic traits that will be introduced into commercial cultivars by an increasingly controlled and targeted coordination of recombination throughout the breeding system. DNA markers will therefore play a dual role through aiding genetic analysis of the underlying basis of important traits, and for assisting in the selection of promising progeny that after validation through field testing may become new cultivars in farmers’ fields.
ACKNOWLEDGMENTS Sangam Dwivedi is grateful to the GCP for financial support during the first stages of manuscript development, ICRISAT library staV for assistance in literature searches and sourcing reprints; to Naveen Puppala of New Mexico State University (NMSU), United States for his support during the later stages of manuscript development; and to Valerie Pipkin (NMSU) for help in preparing the manuscript for final submission.
REFERENCES Ablett, G., Hill, H., and Henry, R. J. (2006). Sequence polymorphism discovery in wheat microsatellite flanking regions using pyrophosphate sequencing. Mol. Breed. 17, 281–289. Aert, R., Sa´gi, L., and Volckaert, G. (2004). Gene content and density in banana (Musa acuminata) as revealed by genomic sequencing of BAC clones. Theor. Appl. Genet. 109, 129–139.
APPLIED CROP GENOMICS
279
Ahmad, M. (2000). Molecular marker‐assisted selection of HMW glutenin alleles related to wheat bread quality by PCR‐generated DNA markers. Theor. Appl. Genet. 101, 892–896. Ahmadi, N., Albar, L., Pressoir, G., Pinel, A., Fargette, D., and Ghesquiere, A. (2001). Genetic basis and mapping of the resistance to rice yellow mottle virus. III. Analysis of QTL eYciency in introgressed progenies confirmed the hypothesis of complementary epistasis between two resistance QTL. Theor. Appl. Genet. 103, 1084–1092. Allison, D. B. (1997). Transmission‐disequilibrium tests for quantitative traits. Am. J. Hum. Genet. 60, 676–690. Alpert, K. B., and Tanksley, S. D. (1996). High‐resolution mapping and isolation of a yeast artificial chromosome contig containing fw2.2: A major fruit weight quantitative trait locus in tomato. Proc. Natl. Acad. Sci. USA 93, 15503–15507. Andaya, V. C., and Tai, T. H. (2006). Fine mapping of the qCTS12 locus, a major QTL for seedling cold tolerance in rice. Theor. Appl. Genet. 113, 467–475. Andersen, J. R., and Lu¨bberstedt, T. (2003). Functional markers in plants. Trends Plant Sci. 8, 554–560. Anderson, J. A., Stack, R. W., Liu, S., Waldron, B. L., Fjeld, A. D., Coyne, C., Moreno‐ Sevilla, B., Fetch, J. M., Song, Q. J., Cregan, P. B., and Frohberg, R. C. (2001). DNA markers for fusarium head blight resistance QTL in two wheat populations. Theor. Appl. Genet. 102, 1164–1168. Anderson, J. V., Delseny, M., Fregene, M. A., Jorge, V., Mba, C., Lopez, C., Restrepo, S., Soto, M., Piegu, B., Verdier, V., Cooke, R., Tohme, J., et al. (2004). An EST resource for cassava and other species of Euphorbiaceae. Plant Mol. Biol. 56, 527–539. Andrews, D. J., and Anand Kumar, K. (1996). Use of West African pearl millet landrace Iniadi in cultivar development. Plant Genet. Resour. Newsl. 105, 15–22. Arru, L., Faccini, N., Govoni, C., Cattivelli, L., Pecchioni, N., Delogu, G., Stanca, A. M., and Vale, G. (2003). The PCR‐based marker MWG2018 linked to the RDG2A leaf stripe resistance gene is a useful tool for assessing barley resistance in breeding programs. Crop Sci. 43, 1036–1042. Arumuganathan, K., and Earle, E. D. (1991). Nuclear DNA content of some important plant species. Plant Mol. Biol. Rep. 9, 208–219. Arzani, A., Peng, J. H., and Lapitan, N. L. V. (2004). DNA and morphological markers for a Russian wheat aphid resistance gene. Euphytica 139, 167–172. Ashikari, M., and Matsuoka, M. (2002). Application of rice genomics to plant biology and breeding. Bot. Bull. Acad. Sin. 43, 1–11. Ashikari, M., Sakakibara, H., Lin, S., Yamamoto, T., Takashi, T., Nishimura, A., Angeles, R. E., Qian, Q., Kitano, H., and Matsuoka, M. (2005). Cytokinin oxidase regulates rice grain production. Science 309, 741–745. Ashkenazi, V., Chani, E., Lavi, U., Levy, D., Hillel, J., and Veilleux, R. E. (2001). Development of microsatellite markers in potato and their use in phylogenetic and fingerprinting analyses. Genome 44, 50–62. Ası´ns, M. J. (2002). Present and future of quantitative trait locus analysis in plant breeding. Plant Breed. 121, 281–291. Auger, D. L., Gray, A. D., Ream, T. S., Kato, A., Coe, E. H., Jr., and Birchler, J. A. (2005). Nonadditive gene expression in diploid and triploid hybrids of maize. Genetics 169, 389–397. Avila, C. M., Nadal, S., Moreno, M. T., and Torres, A. M. (2006). Development of a simple PCR‐based marker for the determination of growth habits in Vicia faba L. using a candidate gene approach. Mol. Breed. 17, 185–190. Ayoub, M., Armstrong, E., Bridger, G., Fortin, M. G., and Mather, D. E. (2003). Marker‐based selection in barley for a QTL region aVecting a‐Amylase activity of malt. Crop Sci. 43, 556–561.
280
S. L. DWIVEDI ET AL.
Azanza, F., Barzur, A., and Juvik, J. A. (1996a). Variation in sweet corn kernel characteristics associated with stand establishment and eating quality. Euphytica 87, 7–18. Azanza, F., Tadmor, Y., Klein, B. P., Rocheford, T. R., and Juvik, J. A. (1996b). Quantitative trait loci influencing chemical and sensory characteristics of eating quality in sweet corn. Genome 39, 40–50. Babu, E. R., Mani, V. P., and Gupta, H. S. (2004). Combining high quality protein and hard endosperm traits through phenotypic and marker assisted selection.Fourth International Crop Science Congress, Australia( http://www.regional.org.au/au/cs/2004/symposia/5/2/ 1321_babu.htm). Bafna, V., Gusfield, D., Lancia, G., and Yooseph, S. (2003). Haplotyping as perfect phylogeny: A direct approach. J. Comput. Biol. 10, 323–340. Bao, J. B., Lee, S., Chen, C., Zhang, X.‐Q., Zhang, Y., Liu, S.‐Q., Clark, T., Wang, J., Cao, M.‐L., Yang, H.‐M., Wang, S. M., and Yu, J. (2005). Serial analysis of gene expression study of a hybrid rice strain (LYP9) and its parental cultivars. Plant Physiol. 138, 1216–1231. Barone, A. (2004). Molecular marker‐assisted selection for potato breeding. Am. J. Potato Res. 81, 111–117. Barone, A., Ritter, E., Schachtschabel, U., Debener, T., Salamini, F., and Gebhardt, C. (1990). Localization by restriction fragment length polymorphism mapping in potato of a major dominant gene conferring resistance to the potato cyst nematode Globodera rostochiensis. Mol. Gen. Genet. 224, 177–182. Barrett, J. C., Fry, B., Maller, J., and Daly, M. J. (2005). Haploview: Analysis and visualization of LD and haplotype maps. Bioinformatics 21, 263–265. Barriere, Y., and Argillier, O. (1993). Brown‐midrib genes of maize: A review. Agronomie 13, 865–876. Barrie`re, Y., Guillet, C., GoVner, D., and Pichon, M. (2003). Genetic variation and breeding strategies for improved cell digestibility in annual forage crops: A review. Anim. Res. 52, 193–228. Batley, J., Barker, G., O’Sullivan, H., Edwards, K. J., and Edwards, D. (2003). Mining for single nucleotide polymorphisms and insertions/deletions in maize expressed sequence tag data. Plant Physiol. 132, 84–91. Bearzoti, E., and Vencovsky, R. (1998). Methodology estimation of the proportion of genetic variance explained by molecular markers. Genet. Mol. Biol. 21, 557–566. Bearzoti, E., and Vencovsky, R. (2002). Simulation of marker‐assisted recurrent selection in autogamous species. Crop Breed. Appl. Biotechnol. 2, 1–10. Beaver, J. S., Miklas, P. N., Kelly, J. D., Steadman, J. R., and Rosas, J. C. (1998). Registration of PR9357‐ 107 small red dry bean germplasm resistant to BCMV, BCMNV, and rust. Crop Sci. 38, 1408–1409. Beavis, W. D. (1998). QTL analysis: Power, precision and accuracy. In ‘‘Molecular Dissection of Complex Traits’’ (A. H. Paterson, Ed.), pp. 145–162. CRC Press, Boca Raton, FL. Bedell, J. A., Budiman, M. A., Nunberg, A., Citek, R. W., Robbins, D., Jones, J., Flick, E., Rohlfing, T., Fries, J., Bradford, K., McMenamy, J., Smith, M., et al. (2005). Sorghum genome sequencing by methylation filtration. PLoS Biol. 3, 0103–0115. Benjamini, Y., and Yekutieli, D. (2005). Quantitative trait loci analysis using the false discovery rate. Genetics 171, 783–790. Bennetzen, J. L. (1998). The evolution of grass genome organization and function. Symp. Soc. Exp. Biol. 51, 123–126. Bennetzen, J. L., Schrick, K., Springer, P. S., Brown, W. E., and SanMiguel, P (1994). Active maize genes are unmodified and flanked by diverse classes of modified, highly repetitive DNA. Genome 37, 565–576. Bernacchi, D., Beck‐Bunn, T., Emmatty, D., Eshed, Y., Inai, S., Lopez, J., Petiard, V., Sayama, H., Uhlig, J., Zamir, D., and Tanksley, S. D. (1998a). Advanced backcross QTL
APPLIED CROP GENOMICS
281
analysis of tomato: II. Evaluation of near‐isogenic lines carrying single‐donor introgressions for desirable wild QTL‐ alleles derived from Lycopersicon hirsutum and L. pimpinellifolium. Theor. Appl. Genet. 97, 170–180. Bernacchi, D., Beck‐Bunn, T., Eshed, Y., Lopez, J., Petiard, V., Uhlig, J., Zamir, D., and Tanksley, S. D. (1998b). Advanced backcross QTL analysis in tomato. I. Identification of QTLs for traits of agronomic importance from Lycopersicon hirsutum. Theor. Appl. Genet. 97, 381–397. Bernardo, R. (1994). Prediction of maize single‐cross performance using RFLPs and information from related hybrids. Crop Sci. 34, 20–25. Bernardo, R. (1996). Best linear unbiased prediction of maize single‐cross performance. Crop Sci. 36, 50–56. Bernardo, R. (2002). ‘‘Breeding for Quantitative Traits in Plants.’’ Stemma Press, Woodbury, Minnesota. Bernardo, R. (2004). What proportion of declared QTL in plants are false? Theor. Appl. Genet. 109, 419–424. Bhattramakki, D., Dong, J., Chhabra, A. K., and Hart, G. E. (2000). An integrated SSR and RFLP linkage map of Sorghum bicolor (L.) Moench. Genome 43, 988–1002. Bhattramakki, D., Dolan, M., Hanafey, M., Wineland, R., Vaske, D., Register, J. C., III, Tingey, S. V., and Rafalski, A. (2002). Insertion‐delition polymorphisms in 30 regions of maize genes occur frequently and can be used as highly informative genetic markers. Plant. Mol. Biol. 48, 539–547. Bidinger, F. R., Serraj, R., Rizvi, S. M. H., Howarth, C., Yadav, R. S., and Hash, C. T. (2005). Field evaluation of drought tolerance QTL eVects on phenotype and adaptation in pearl millet (Pennisetum glaucum (L.) R. Br.) top cross hybrids. Field Crop Res. 94, 14–32. Bink, M., and Meuwissen, T. H. E. (2004). Fine mapping of quantitative trait loci using linkage disequilibrium in inbred plant populations. Euphytica 137, 95–99. Birchler, J. A., Auger, D. L., and Riddle, N. C. (2003). In search of the molecular basis of heterosis. Plant Cell 15, 2236–2239. Blaszczyk, L., Chelkowski, J., Korzun, V., Kraic, J., Ordon, F., Ovesna, J., Purnhauser, L., Tar, M., and Vida, G. (2004). Verification of STS markers for leaf rust resistance genes by seven European laboratories. Cell. Mol. Biol. Lett. 9, 805–817. Bohn, M., Groh, S., Khairallah, M. M., Hoisington, D. A., Utz, H. F., and Melchinger, A. E. (2001). Re‐evaluation of the prospects of marker‐assisted selection for improving insect resistance against Diatraea spp. In tropical maize by cross validation and independent validation. Theor. Appl. Genet. 103, 1059–1067. Bonnett, D. G., Rebetzke, G. J., and Spielmeyer, W. (2005). Strategies for eYcient implementation of molecular markers in wheat breeding. Mol. Breed. 15, 75–85. Boominathan, P., Shukla, R., Kumar, A., Manna, D., Negi, D., Praveen, K. V., and Chattopadhyay, D. (2004). Long term transcript accumulation during the development of dehydration adaptation in Cicer arietinum. Plant Physiol. 135, 1608–1620. Boreck, I. B., and Suarez, B. K. (2001). Linkage and association: Basic concepts. Adv. Genet. 42, 45–66. Borevitz, J. O., Maloof, J. N., Lutes, J., Dabi, T., Redfern, J. L., Trainer, G. T., Werner, J. D., Asami, T., Berry, C. C., Weigel, D., and Chory, J. (2002). Quantitative trait loci controlling light and hormone response in two accessions of Arabidopsis thaliana. Genetics 160, 683–696. Borrell, A. K., Hammer, G. L., and van Oosterom, E. (2001). Staygreen: A consequence of the balance between supply and demand for nitrogen during grain filling? Ann. Appl. Biol. 138, 91–95. Bost, B., de Vienne, D., Hospital, F., Moreau, L., and Dillmann, C. (2001). Genetic and nongenetic basis for the L‐shaped distribution of quantitative trait loci eVects. Genetics 157, 1773–1787.
282
S. L. DWIVEDI ET AL.
Bouchez, A., Hospital, F., Causse, M., Gallais, A., and Charcosset, A. (2002). Marker‐assisted introgression of favorable alleles at quantitative trait loci between maize elite lines. Genetics 162, 1945–1959. Bowers, J. E., Abbey, C., Anderson, S., Chang, C., Draye, X., Hoppe, A. H., Jessup, R., Lemke, C., Lennington, J., Li, Z., Lin, Y.‐R., Liu, S.‐C., et al. (2003). A high‐density genetic recombination map of sequence‐tagged site for Sorghum, as a framework for comparative structural and evolutionary genomics of tropical grains and grasses. Genetics 165, 367–386. Bowman, J. G. P., Blake, T. K., Surber, L. M. M., Habernicht, D. K., and Bockelman, H. (2001). Feed‐quality variation in the barley core collection of the USDA National small grain collection. Theor. Appl. Genet. 41, 863–870. Bradbury, L. M. T., Henry, R. J., Jin, Q., Reinke, R. F., and Waters, D. L. E. (2005). A perfect marker for fragrance genotyping in rice. Mol. Breed. 16, 279–283. Bra¨utigam, M., Lindlo¨f, A., Zakhrabekova, S., Gharti‐Chhetri, G., Olsson, B., and Olsson, O. (2005). Generation and analysis of 9,792 EST sequences from cold acclimated oat, Avena sativa. BMC Plant Biol. 5, 18. doi:10.1186/1471‐2229‐5‐18. Brem, R. B., Yvert, G., Clinton, R., and Kruglyak, L. (2002). Genetic dissection of transcriptional regulation in budding yeast. Science 296, 752–755. Breseghello, F., and Sorrells, M. (2006a). Association mapping of kernel size and milling quality in wheat (Triticum aestivum L.) cultivars. Genetics 172, 1165–1177. Breseghello, F., and Sorrells, M. (2006b). Association analysis as a strategy for improvement of quantitative traits in plants. Crop Sci. 46, 1323–1330. Bretting, P. K., and Widrlechner, M. P. (1995). Genetic markers and plant genetic resource management. Plant Breed. Rev. 13, 11–86. Brick, M. A., Byrne, P. F., Schwartz, H. F., Ogg, J. B., Otto, K., Fall, A. L., and Gilbert, J. (2006). Reaction to three races of Fusarium wilt in the Phaseolus vulgaris core collection. Crop Sci. 46, 1245–1252. Briney, A., Wilson, R., Potter, R. H., Barclay, I., Crosbie, G., Appels, R., and Jones, M. G. K. (1998). A PCR‐based marker for selection of starch and potential noodle quality in wheat. Mol. Breed. 4, 427–433. Broughton, W. J., Herna´ndez, G., Blair, M., Beebe, S., Gepts, P., and Vanderleyden, J. (2003). Beans (Phaseolus spp.)—model food legumes. Plant Soil 252, 55–128. Brown, A. H. D. (1989). Core collections: A practical approach to genetic resources management. Genome 31, 818–824. Brown, S. M., Hopkins, M. S., Mitchell, S. E., Senior, M. L., Wang, T. Y., Duncan, R. R., Gonzalez‐Candelas, F., and Kresovich, S. (1996). Multiple methods for the identification of polymorphic simple sequence repeats (SSRs) in sorghum [Sorghum bicolor (L.) Moench]. Theor. Appl. Genet. 93, 190–198. Budak, H., Pedraza, F., Cregan, P. B., Baenziger, P. S., and Dweikat, I. (2003). Development and utilization of SSRs to estimate the degree of genetic relationships in a collection of pearlmillet germplasm. Crop Sci. 43, 2284–2290. Buhariwalla, H. K., Jarret, R. L., Jayashree, B., Crouch, J. H., and Ortiz, R. (2005a). Isolation and characterization of microsatellite markers from Musa balbisiana. Mol. Ecol. Notes 5, 327–330. Buhariwalla, H. K., Jayashree, B., Eeshwar, K., and Crouch, J. H. (2005b). Development of ESTs from chickpea roots and their use in diversity analysis of the Cicer genus. BMC Plant Biol. 5, 16. DOI:10.1186/1471‐2229‐5‐16. Burow, M. D., Simpson, E. C., Starr, J. L., and Paterson, A. H. (2001). Transmission genetics of chromatin from a synthetic amphidiploid to cultivated peanut (Arachis hypogaea L.): Broadening the gene pool of a monophyletic polyploidy species. Genetics 159, 823–837. Burr, B., Evola, S. V., Burr, F. A., and Backemann, J. S. (1983). The application of restriction fragment length polymorphism to plant breeding. In ‘‘Genetic Engineering’’ (J. K. Setlow and A. Hollaender, Eds.), Vol. 5, pp. 45–49. Plenum Press, New York.
APPLIED CROP GENOMICS
283
Burr, B., Burr, F. A., Thompson, K. H., Alberston, M. C., and Stubber, C. W. (1988). Gene mapping with recombinant inbreds in maize. Genetics 118, 519–526. Bustamam, M., Tabien, R. E., Suwarmo, A., Abalos, M. C., Kadir, T. S., Ona, I., Bernardo, M., VeraCruz, C. M., and Leung, H. (2002). Asian rice biotechnology network: Improving popular cultivars through marker‐assisted backcrossing by the NARES. Abstract of the Int. Rice Congress, 16–22 September 2002, Beijing China, (www.irri.org/irc2002/index.htm). Buteler, M. I., Jarret, R. L., and LaBonte, D. R. (1999). Sequence Characterization of microsatellites in diploid and polyploidy Ipomoea. Theor. Appl. Genet. 99, 123–132. Cagampang, G. B., Perez, C. M., and Juliano, B. O. (1973). A gel consistency test for eating quality of rice. J. Sci. Food Agric. 24, 1589–1594. Cahill, D. J., and Schmidt, D. H. (2004). Use of marker‐assisted selection in a product development breeding program. ‘‘Fourth International Crop Science Congress, Brisban, Australia, 26 September–1 October 2004’’ (T. Fischer, Ed.). (www.cropscience.org.au). Cakir, M., Gupta, S., Platz, G. J., Ablett, G. A., Loughman, R., Emebiri, L. C., Poulsen, D., Li, C. D., Lance, R. C. M., Galwey, N. W., Jones, M. G. K., and Appels, R. (2003). Mapping and validation of the genes for resistance to Pyrenophora teres f. teres in barley (Hordium vulgare L.). Aust. J. Agric. Res. 54, 1369–1377. Campos, H., Cooper, A., Habben, J. E., Edmeades, G. O., and Schussler, J. R. (2004). Improving drought tolerance in maize: A view from industry. Field Crop Res. 90, 19–34. Canci, P. C., Nduulu, L. M., Muehlbauer, G. J., Dill‐Macky, R., Rasmussoon, D. C., and Smith, K. P. (2004). Validation of quantitative trait loci for fusarium head blight and kernel discoloration in barley. Mol. Breed. 14, 91–204. Cao, L.‐Y., Zhuang, J.‐Y., Zhan, X.‐D., Zheng, K.‐L., and Cheng, S.‐H. (2003). Hybrid rice resistant to bacterial blight developed by marker assisted selection. Chinese J. Rice Sci. 17, 184–186. Castro, A. J., Chen, X., Hayes, P. M., and Johnston, M. (2003a). Pyramiding quantitative trait locus (QTL) alleles determining resistance to barley stripe rust: EVects on resistance at seedling stage. Crop Sci. 43, 651–659. Castro, A. J., Chen, X., Corey, A., Filichkina, T., Hayes, P. M., Mundt, C., Richardson, K., Sandoval‐Islas, S., and Vivar, H. (2003b). Pyramiding and validation of quantitative trait locus (QTL) alleles determining resistance to barley stripe rust: EVects on adult plant resistance. Crop Sci. 43, 2234–2239. Causse, M., Santoni, S., Damerval, C., Maurice, A., Charcosset, A., Deatrick, J., and de Vienne, D. (1996). A composite map of expressed sequences in maize. Genome 39, 418–432. Causse, M. A., Fulton, T. M., Cho, Y. G., Ahn, S. N., Chunwongse, J., Wu, K., Xiao, J., Yu, Z., Ronald, P. C., Harrington, S. E., Second, G., McCouch, S. R., et al. (1994). Saturated molecular map of the rice genome based on an interspecific backcross population. Genetics 138, 1251–1274. Chaitieng, B., Kaga, A., Tomooka, N., Isemura, T., Kuroda, Y., and Vaughan, D. A. (2006). Development of a black gram [Vigna mungo (L.) Hepper] linkage map and its comparison with an azuki bean [Vigna angularis (Willd.) Ohwi and Ohashi] linkage map. Theor. Appl. Genet. 113, 1261–1269. Chandler, P. M., Marrion‐Poll, A., Ellis, M., and Gubler, F. (2002). Mutants at the Slender1 locus of barley cv Himalaya. Molecular and physical characterization. Plant Physiol. 129, 181–190. Chao, S., Baysdorfer, C., Heredia‐Diaz, O., Musket, T., Xu, G., and Coe, E. H., Jr. (1994). RFLP mapping of partially sequenced leaf cDNA clones in maize. Theor. Appl. Genet. 88, 717–721. Chapman, S. C., Hammer, G. L., Podlich, D. W., and Cooper, M. (2002). Linking biophysical and genetic models to integrate physiology, molecular biology and plant breeding. In ‘‘Quantitative Genetics, Genomics and Plant Breeding’’ (M. S. Kong, Ed.), pp. 167–187. CABI, Wallingford, UK.
284
S. L. DWIVEDI ET AL.
Chapman, S. C., Cooper, M., Podlich, D. W., and Hammer, G. L. (2003). Evaluating plant breeding strategies by simulating gene action and dryland environment eVects. Agron. J. 95, 99–113. Chavarriaga‐Aguirre, P., Maya, M. M., Bonierbale, M. W., Kresovich, S., Fergene, M. A., Tohme, J., and Kochert, G. (1998). Microsatellites in cassava (Manihot esculenta Crantz): Discovery, inheritance and variability. Theor. Appl. Genet. 97, 493–501. Chavarriaga‐Aguirre, P., Maya, M. M., Tohme, J., Duque, M. C., Carlos, I., Bonierbale, M. W., Kresovich, S., and Kochert, G. (1999). Using microsatellites, isozymes and AFLPs to evaluate genetic diversity and redundancy in the cassava core collection and to assess the usefulness of DNA‐based markers to maintain germplasm collections. Mol. Breed. 5, 263–273. Chen, D.‐H., dela Vina, M., Inukai, T., Mackill, D. J., Ronald, P. C., and Nelson, R. J. (1999). Molecular mapping of the blast resistance gene, Pi44(t), in a line derived from a durably resistant rice cultivar. Theor. Appl. Genet. 98, 1046–1053. Chen, M., Presting, G., Barbazuk, W. B., Goicoechea, J. L., Blackmon, B., Fang, G., Kim, H., Frisch, D., Yu, Y., Higingbottom, S., Phimphilai, J., Phimphilai, D., et al. (2002). An integrated physical and genetic map of the rice genome. Plant Cell 14, 537–545. Chen, S., Lin, X. H., Xu, C. G., and Zhang, A. (2000). Improvement of bacterial blight resistance of Minghui 63, an elite restorer line of hybrid rice by marker‐assisted selection. Crop Sci. 40, 239–244. Chen, S., Xu, C. G., Lin, X. H., and Zhang, Q. (2001). Improving bacterial blight resistance of 6078, an elite restorer line of hybrid rice, by molecular marker‐assisted selection. Plant Breed. 120, 133–137. Cheng, Z., Presting, G. G., Buell, C. R., Wing, R. A., and Jiang, J. (2001). High resolution pachytene chromosome mapping of bacterial artificial chromosomes anchored by genetic markers reveals the chromosome location and the distribution of genetic recombination along chromosome 10 of rice. Genetics 157, 1749–1757. Chin, E. C. L., Senior, M. L., Shu, H., and Smith, J. S. C. (1996). Maize simple repetitive DNA sequences: Abundance and allelic variation. Genome 39, 866–873. Ching, A., Caldwell, K. S., Jung, M., Dolan, M., Smith, O. S., Tingey, S., Morgante, M., and Rafalski, A. J. (2002). SNP frequency, haplotype structure and linakge disequilibrium in elite maize inbred lines. BMC Genet. 3, 19–33. Cho, R. J., Mindrinos, M., Richards, D. R., Sapolsky, R. J., Anderson, M., Drenkard, E., Dewdney, J., Reuber, T. L., Stammers, M., Federspiel, N., Theologis, A., Yang, W.‐H., et al. (1999). Genome‐wide mapping with biallelic markers in Arabidposis thaliana. Nat. Genet. 23, 203–207. Cho, Y. G., Eun, M. Y., McCouch, S. R., and Chae, Y. A. (1994). The semidwarf gene, sd‐1, of rice (Oryza sativa L.). II. Molecular mapping and marker‐assisted selection. Theor. Appl. Genet. 89, 54–59. Cho, Y. G., Ishii, T., Temnykh, S., Chen, X., Lipovich, L., McCouch, S. R., Park, W. D., Ayres, N., and Cartinhour, S. (2000). Diversity of microsatellites derived from genomic libraries and GeneBank sequences in rice. Theor. Appl. Genet. 100, 713–722. Choi, H.‐K., Melissa, A. L., Doyle, J., and Cook, D. R. (2006). Development of nuclear gene‐ derived markers linked to legume species. Mol. Gen. Genom. 276, 56–70. Christopher, M., Cordeiro, G., Waters, D., and Henry, R. (2004). Marker assisted selection in rice improvement. RIRDC Publication No 04/011 (http://www.rirdc.gov.au/reports/ric/ 04‐011sum.html). Coe, E., Cone, K., McMullen, M., Chen, S.‐S., Davis, G., Gardiner, J., Liscum, E., Polacco, M., Paterson, A., Sanchez‐Villeda, H., Soderlund, C., and Wing, R. (2002). Access to the maize genome: An integrated physical and genetic map. Plant Physiol. 128, 9–12.
APPLIED CROP GENOMICS
285
Coemans, B., Matsumura, H., Terauchi, R., Remy, S., Swennen, R., and Sa´gi, L. (2005). SuperSAGE combined with PCR walking allows global gene expression profiling of banana (Musa acuminata), a non‐model organism. Theor. Appl. Genet. 111, 1118–1126. Collard, B. C. Y., and Mackill, D. J. (2007). Marker‐assisted selection: An approach for precision plant breeding in the 21st century. Phil. Trans. R. Soc. B. Rev. (in press). Collins, H. M., Panozzo, J. F., Logue, S. J., JeVeries, S. P., and Barr, A. R. (2003). Mapping and validation of chromosome regions associated with high malt extract in barley (Hordeum vulgare L.). Aust. J. Agric. Res. 54, 1223–1240. Colton, L. M., Groza, H. I., Wielgus, S. M., and Jiang, J. (2006). Marker‐assisted selection for the broad‐ spectrum potato late blight resistance conferred by gene RB derived from a wild potato species. Crop Sci. 46, 589–594. Conaway‐Bormans, C. A., Marchetti, M. A., Johnson, C. W., McClung, A. M., and Park, W. D. (2003). Molecular markers linked to the rice blast resistance gene Pi‐z in rice for use in marker‐assisted selection. Theor. Appl. Genet. 107, 1014–1020. Concibido, V. C., Denny, R. L., Lange, D. A., Orf, J. H., and Young, N. D. (1996). RFLP mapping and molecular marker‐assisted selection of soybean cyst nematode resistance in PI 209332. Crop Sci. 36, 1643–1650. Concibido, V. C., La Vallee, B., Mclaird, P., Pineda, N., Meyer, J., Hummel, L., Yang, J., Wu, K., and Delannay, X. (2003). Introgression of a quantitative trait locus for yield from Glycine soja into commercial soybean cultivars. Theor. Appl. Genet. 106, 575–582. Concibido, V. C., Diers, B. W., and Arelli, P. R. (2004). A decade of QTL mapping for cyst nematode resistance in soybean. Crop Sci. 44, 1121–1131. Cone, K. C., McMullen, M. D., Bi, I. V., Davis, G. L., Yim, Y.‐S., Gardiner, J. M., Polacco, M. L., Sanchez‐Villeda, H., Fang, Z., Schroeder, S. G., Havermann, S. A., Bowers, J. E., et al. (2002). Genetic, physical, and informatics resources for maize. On the road to integrated map. Plant Physiol. 130, 1598–1605. Cooper, M., Chapman, S. C., Podlich, D. W., and Hammer, G. L. (2002). The GP problem: Quantifying gene‐to‐phenotype relationships. In Silico Biol. 2, 151–164. Cooper, M., Smith, O. S., Graham, G., Arthur, L., Feng, L., and Podlich, D. W. (2004). Genomics, Genetics, and Plant Breedings: A private sector perspective. Crop Sci. 44, 1907–1913. Costa, J. M., Corey, A., Hayes, P. M., Jobet, C., Kleinhofs, A., Kopisch‐Obusch, A., Kramer, S. F., Kudrna, D., Li, M., Riera‐Lizarazu, O., Sato, K., Szucs, P., et al. (2001). Molecular mapping of the Oregon Wolfe barleys: A phenotypically polymorphic double‐ haploid population. Theor. Appl. Genet. 103, 415–424. Courtois, B., Chaitep, W., Moolsri, S., Sinha, P. K., Trebuil, G., and Yadav, R. (1996). Drought resistance and germplasm improvement. Ongoing research in the Upland rice research consortium. In ‘‘Upland Rice Research in Partnership IRRI’’ (C. Piggin, B. Curtois, and V. Schmit, Eds.), pp. 154–175. IRRI Discussion Paper Series 16, Los Ban˜os, the Philippines. Coventry, S. J., Collins, H. M., Barr, A. R., JeVerues, S. P., Chalmers, K. J., Logue, S. J., and Langridge, P. (2003). Use of putative QTLs and structural genes in marker‐assisted selection for diastatic power in malting barley (Hordeum vulgare L.). Aust. J. Agric. Res. 54, 1241–1250. Cregan, P. B., Mudge, J., Fickus, E. W., Danesh, D., Denny, R., and Young, N. D. (1999). Two simple sequence repeat markers to select for soybean cyst nematode resistance conditioned by the rhg1 locus. Theor. Appl. Genet. 99, 811–818. Crosbie, T. M., Eathington, S. R., Johnson, G. R., Edwards, M., Reiter, R., Stark, S., Mohanty, R. G., Oyervides, M., Buehler, R. E., Walker, A. K., Dobert, R., Delannay, X., et al. (2006). Plant breeding: Past, present, and future. In ‘‘Plant Breeding: The Arnel R. Hallauer International Symposium’’ (K. R. Lamkey and M. Lee, Eds.). Blackwell Publishing, Ames, Iowa.
286
S. L. DWIVEDI ET AL.
Crossa, J., Burguen˜o, J., Cornelius, P. L., McLaren, G., Trethowan, R., and Krischnamachari, A. (2006). Modeling genotype x environment interaction using additive genetic covariance of relatives for predicting breeding values of wheat genotypes. Crop Sci. 46, 1722–1733. Crouch, H. K., Crouch, J. H., Jarret, R. L., Cregan, P. B., and Ortiz, R. (1998a). Segregation at microsatellite loci in haploid and diploid gametes of Musa. Crop Sci. 98, 211–217. Crouch, J. H., Vuylsteke, D., and Ortiz, R. (1998b). Perspectives on the application of biotechnology to assist the genetic enhancement of plantain and banana (Musa spp.). Electronic J. Biotechnol. 1, 1–12(http://www.ejb.org/content/vol1/issue1/full/2/). Darvasi, A., and Soller, M. (1995). Advanced intercross lines, an experimental population for fine genetic mapping. Genetics 141, 1199–1207. Datta, K., Baisakh, N., Thet, K. M., Tu, J., and Datta, S. K. (2002). Pyramiding transgenes for multiple resistance in rice against bacterial blight, yellow stem borer, and sheath blight, Theor. Appl. Genet. 106, 1–8. Davis, G. L., McMullen, M. D., Baysdorfer, C., Musket, T., Grant, D., Staebell, M., Xu, G., Polacco, M., Koster, L., Melia‐Hancock, S., Houchins, K., Chao, S., et al. (1999). A maize map standard with sequenced core markers, grass genome reference points and 932 expressed sequence tagged sites (ESTs) in a 1736‐locus map. Genetics 152, 1137–1172. de Bustos, A., Rubio, P., Soler, C., Garcia, P., and Jouve, N. (2001). Marker‐assisted selection to improve HMW‐glutenins in wheat. Euphytica 119, 69–73. Dekkers, J. C. M., and Hospital, F. (2002). The use of molecular genetics in the improvement of agricultural populations. Nat. Rev. Genet. 3, 22–32. Dellaporta, S. L., Wood, J., and Hicks, J. B. (1983). A plant DNA minipreparation: Version II. Plant Mol. Biol. Rep. 1, 19–21. Delmer, D. P. (2005). Agriculture in the developing world: Connecting innovations in plant research to downstream application. Proc. Natl. Acad. Sci. USA 102, 15739–15746. Demeke, T., Laroche, A., and Gaudet, D. A. (1996). A DNA marker for the Bt‐10 common bunt resistant gene in wheat. Genome 39, 51–56. Deng, H. W., Chen, W. M., and Reskker, R. R. (2001). Population admixture: Detection by Hardy‐ Weinberg test and its quantitative eVects on linkage disequilibrium methods for localizing genes underlying complex traits. Genetics 157, 885–897. Deu, M., and Glaszmann, J. C. (2004). Linkagedisequilibrium in sorghum. Plant and Animal Genomes XII Conference, PAG‐XII: (W10). Deu, M., Rattunde, F., and Chantereau, J. (2006). A global view of genetic diversity in cultivated sorghums using a core collection. Genome 49, 168–180. de Vicente, M. C., and Tanksley, S. D. (1993). QTL analysis of transgressive segregation in an interspecific tomato cross. Genetics 134, 585–596. Doebley, J., and Stec, A. (1993). Inheritance of the morphological diVerences between maize and teosinte: Comparison of results for two F2 population. Genetics 134, 559–570. Doebley, J., Stec, A., and Gustus, C. (1995). Teosinte branched 1 and the origin of maize: Evidence for epistasis and the evolution of dominance. Genetics 141, 333–346. Doebley, J., Stec, A., and Hubbard, L. (1997). The evolution of apical dominance in maize. Nature 386, 485–488. Doi, K., Izawa, T., Fuse, T., Yamanouchi, U., Kubo, T., Shimatani, Z., Yano, M., and Yoshimura, A. (2004). Ehd1, a B‐type response regulator in rice, confers short‐day promotion of flowering and controls FT‐like gene expression independently of Hd1. Genes Dev. 18, 926–936. Dong, Y. S., Cao, Y. S., Zhang, X. Y., Liu, S. C., Wang, L. F., You, G. X., Pang, B. S., Li, L. H., and Jia, J. Z. (2003). Establishment of candidate core collections in Chinese common wheat germplasm. J. Plant Genet. Resour. 4, 1–8. Dorweiler, J., Stec, A., Kermicle, J., and Doebley, J. (1993). Teosinte glume architecture 1: A genetic locus controlling a key step in maize evolution. Science 262, 233–235.
APPLIED CROP GENOMICS
287
Douglass, S. K., Juvik, J. A., and Spilltstoesser, W. E. (1993). Sweet corn seedling emergence and variation in kernel carbohydrate reserves. Seed Sci. Technol. 21, 433–445. D’Ovidio, R. (1993). Single‐seed PCR of LMW glutenin genes to distinguish between durum wheat cultivars with good and poor technological properties. Plant Mol. Biol. 22, 1173–1176. Draye, X., Lin, Y.‐R., Qian, X.‐Y., Bowers, J. E., Burow, G. B., Morell, P. L., Peterson, D. G., Presting, G. G., Ren, S.‐X., Wing, R. A., and Paterson, A. H. (2001). Toward integration of comparative genetic, physical, diversity, and cytomolecular maps for grasses and grains, using the sorghum genome as a foundation. Plant Physiol. 125, 1325–1341. Dreher, K., Morris, M., Khairallah, M., Ribaut, J. M., Pandey, S., and Srinivasan, G. (2002). Is marker‐assisted selection cost‐eVective compared with conventional plant breeding methods? The case of quality protein Maize. In ‘‘Economic and Social Issues in Agricultural Biotechnology’’ (R. E. Evenson, V. Santaniello, and D. Zilberman, Eds.), pp. 203–236. CABI, Wellingford, New York, USA. Dreher, K., Khairallah, M., Ribaut, J. M., and Morris, M. (2003). Money matters (I): Costs of field and laboratory procedures associated with conventional and marker‐assisted maize breeding at CIMMYT. Mol. Breed. 11, 221–234. Dubcovsky, J., Lijavetzky, D., Appendino, L., and Tranquilli, G. (1998). Comparative RFLP mapping of Triticum monococcum genes controlling vernalization requirement. Theor. Appl. Genet. 97, 968–974. Durrett, R. T., Chen, K. Y., and Tanksley, S. D. (2002). A simple formula useful for positional cloning. Genetics 160, 353–355. Duvick, D. N. (1999). Heterosis: Feeding the people and protecting natural resources. In ‘‘Genetics and Exploitation of Heterosis in Crops’’ (J. G. Coors, and S. Pandey, Eds.), pp. 19–29. American Society of Agronomy, Madison, Wisconsin, USA. Dvorak, J., Luo, M. C., and Yang, Z. L. (1998). Restriction fragment length polymorphism and divergence in the genomic regions of high and low recombination in self‐fertilizing and cross‐fertilizing Aegilops species. Genetics 148, 423–434. Dweikat, I., Ohm, H., Mackenzie, S., Paterson, F., Cambron, S., and RatcliVe, R. (1994). Association of DNA marker with Hessian fly resistance gene H9 in wheat. Theor. Appl. Genet. 89, 964–968. Dwivedi, S. L., Crouch, J. H., Nigam, S. N., Ferguson, M. E., and Paterson, A. H. (2003). Molecular breeding of groundnut for enhanced productivity and food security in the semi‐ arid tropics: Opportunities and challenges. Adv. Agron. 80, 154–221. Dwivedi, S. L., Blair, M., Upadhyaya, H. D., Serraj, R., Balaji, J., Buhariwalla, H. K., Ortiz, R., and Crouch, J. H. (2005). Using genomics to exploit grain legume biodiversity in crop improvement. Plant Breed. Rev. 26, 176–357. Dwivedi, S. L., Bertioli, D. J., Crouch, J. H., Valls, J. F., Upadhyaya, H. D., Fa´vero, A., Moretzsohn, M., and Paterson, A. H. (2006). Peanut. In ‘‘Genome Mapping and Molecular Breeding in Plants, Volume 2: Oiseeds’’ (C. Kole, Ed.). Springler‐Verlag, Berlin Heidelberg, Germany. Dwivedi, S. L., Stalker, H. T., Blair, M. W., Bertioli, D., Upadhyaya, H. D., Nielen, S., and Ortiz, R. (2007). Enhancing crop gene pools with beneficial traits using wild relatives. Plant Breed. Rev. 30 (in press). Eagles, H. A., Bariana, H. S., Ogbonnaya, F. C., Rebetzke, G. J., Hollamby, G. J., Henry, R. J., Henschke, P. H., and Carter, M. (2001). Implementation of markers in Australian wheat breeding. Aust. J. Agric. Res. 52, 1349–1356. Eathington, S. R. (2005). Practical applications of molecular technology in the development of commercial maize hybrids. In ‘‘Proceedings of the 60th Annual Corn and Sorghum Seed Research Conferences.’’ American Seed Trade Association, Washington DC. Ellis, M. H., Spielmeyer, W., Gale, K. R., Rebetzke, G. J., and Richards, R. A. (2002). ‘Perfect’ markers for the Rht‐B1b and Rht‐D1b dwarfing genes in wheat. Theor. Appl. Genet. 105, 1038–1042.
288
S. L. DWIVEDI ET AL.
Eshed, Y., and Zamir, D. (1994). A genomic library of Lycopersicon pennellii in L. esculentum: A tool for fine mapping of genes. Euphytica 79, 175–179. Eshed, Y., and Zamir, D. (1995). An introgression line population of Lycopersicon pennellii in the cultivated tomato enables the identification and fine mapping of yield‐associated QTL. Genetics 141, 1147–1162. Eujayl, I., Sorrels, M. E., Baum, M., Wolters, P., and Powell, W. (2002). Isolation of EST‐ derived microsatellite markers for genotyping the A and B genomes of wheat. Theor. Appl. Genet. 104, 399–407. Faleiro, F. G., Ragagnin, V. A., Moreira, M. A., and de Barros, E. G. (2004). Use of molecular markers to accelerate the breeding of common bean lines resistant to rust and anthracnose. Euphytica 138, 213–218. Falque, M., Decousset, L., Dervins, D., Jacob, A.‐M., Joets, J., Martinant, J.‐P., RaVoux, X., Ribie`re, N., Ridel, C., Samson, D., Charcosset, A., and Murigneux, A. (2005). Linkage mapping of 1454 new maize candidate gene loci. Genetics 170, 1957–1966. Fan, C., Xing, Y., Mao, H., Lu, T., Han, B., Xu, C., Li, X., and Zhang, Q. (2006). GS3, a major QTL for grain length and weight and minor QTL for grain width and thickness in rice, encodes a putative transmembrane protein. Theor. Appl. Genet. 112, 1164–1171. Fasoula, V. A., Harris, D. K., and Boerma, H. R. (2004). Validation and designation of quantitative trait loci for seed protein, seed oil, and seed weight from two soybean populations. Crop Sci. 44, 1218–1225. Fauquet, C. M., and Tohme, J. (2004). The global cassava partnership for genetic improvement. Plant Mol. Biol. 86, v–x(editorial). Faure´, S., Noyer, J. L., Horry, J. P., Bakry, F., Lanaud, C., and de Leo´n, D. G. (1993). A molecular marker‐based linkage map of diploid bananas (Musa acuminata). Theor. Appl. Genet. 87, 517–526. Feltus, F. A., Wan, J., Schulze, S. R., Estill, J. C., Jiang, N., and Paterson, A. H. (2004). An SNP resource for rice genetics and breeding based on subspecies indica and japonica genome alignments. Genome Res. 14, 1812–1819. Feltus, F. A., Singh, H. P., Lohithaswa, H. C., Schulze, S. R., Silva, T. D., and Paterson, A. H. (2006). A comparative genomic strategy for targeted discovery of single‐nucleotide polymorphisms and conserved‐noncoding sequences in orphan crops. Plant Physiol. 140, 1183–1191. Fernandes, J., Brendel, V., Gai, X., Lal, S., Chandler, V. L., Elumalai, R. P., Galbraith, D. W., Pierson, E. A., and Walbot, V. (2002). Comparison of RNA profiles based on maize expressed sequence tag frequency analysis and micro‐array hybridization. Plant Physiol. 128, 896–910. Fjellstrom, R., Conaway‐Bormans, C. A., McClung, A. M., Marchetti, M. A., Shank, A. R., and Park, W. D. (2004). Development of DNA markers suitable for marker‐assisted selection of three Pi genes conferring resistance to multiple Pyricularia grisea pathotypes. Crop Sci. 44, 1790–1798. Fjellstrom, R., McClung, A. M., and Shank, A. R. (2006). SSR markers closely linked to the Pi‐z locus are useful for selection of blast resistance in a broad array of rice germplasm. Mol. Breed. 17, 149–157. Flint‐Garcia, S. A., Thornsberry, J. M., and Buckler, E. S., IV. (2003). Structure of linkage disequilibrium in plants. Annu. Rev. Plant Biol. 54, 357–374. Fregene, M., Angel, F., Gomez, R., Rodriguez, F., Chavarriaga, P., Roca, W., Tohme, J., and Bonierbale, M. (1997). A molecular genetic map of cassava (Manihot esculenta Crantz). Theor. Appl. Genet. 95, 431–441. Fregene, M., Matsumura, H., Akano, A., Dixon, A., and Terauchi, R. (2004). Serial analysis of gene expression (SAGE) of host‐plant resistance to the cassava mosaic disease (CMD). Plant Mol. Biol. 56, 563–571.
APPLIED CROP GENOMICS
289
Fridman, E., Pleban, T., and Zamir, D. (2000). A recombination hotspot delimits a wild species quantitative trait locus for tomato sugar content to 484 bp within an invertase gene. Proc. Natl. Acad. Sci. USA 97, 4718–4723. Fridman, E. F., Carrari, F., Liu, Y.‐S., Fernie, A. R., and Zamir, D. (2004). Zooming in on a quantitative trait locus for tomato sugar content to 484 bp within an invertase gene. Proc. Natl. Acad. Sci. USA 97, 4718–4723. Frisch, M., Bohn, M., and Melchinger, A. E. (1999a). Minimum sample size and optimal positioning of flanking markers in marker‐assisted backcrossing for transfer of a target gene. Crop Sci. 39, 967–975. Frisch, M., Bohn, M., and Melchinger, A. E. (1999b). Comparison of selection strategies for marker‐assisted backcrossing of a gene. Crop Sci. 39, 1295–1301. Frisch, M., Bohn, M., and Melchinger, A. E. (2000). PLABSIM: Software for simulation of marker‐assisted backcrossing. J. Hered. 91, 86–87. Fu, H., and Dooner, H. K. (2002). Intraspecific violation of genetic colinearity and its implications in maize. Proc. Natl. Acad. Sci. USA 99, 9573–9578. Fu, Y. B., Peterson, G. W., Williams, D., Richards, K. W., and Fetch, J. M. (2005). Patterns of AFLP variation in a core subset of cultivated hexaploid oat germplasm. Theor. Appl. Genet. 111, 530–539. Fulton, T. M., Beck‐Bunn, T., Emmatty, D., Eshed, Y., Lopez, J., Petiard, V., Uhlig, J., Zamir, D., and Tanksley, S. D. (1997). QTL analysis of an advanced backcross of Lycopersicon peruvianum to the cultivated tomato and comparisons with QTLs found in other wild species. Theor. Appl. Genet. 95, 881–894. Gallais, A., Dillmann, C., and Hospital, F. (1997). An analytical approach of marker assisted selection with selection on markers only. In ‘‘Advances in Biometrical Genetics. Proceedings of the Tenth Meeting of the EUCARPIA Section Biometrics in Plant Breeding’’ (R. Krajewski and Z. Kaczmarek, Eds.), pp. 111–116. Institute of Plant Genetics, Polish Academy of Sciences, Poznan. Gaut, B. S., Le Thierry I0 , E. M., Peek, A. S., and Saukins, M. C. (2000). Maize as a model for the evolution of plant nuclear genomes. Proc. Natl. Acad. Sci. USA 97, 7008–7015. Gebhardt, C., Ritter, E., Barone, A., Debener, T., Walkemeier, B., Schachtschabel, U., Kaufmann, H., Thompson, R. D., Bonierbale, M. W., Ganal, M. W., Tanksley, S. D., and Salamini, F. (1991). RFLP map of potato and their alignment with the homoeologous tomato genome. Theor. Appl. Genet. 83, 49–57. Gebhardt, C., Bellin, D., Henselewski, H., Lehmann, W., Schwarzfischer, J., and Valkonen, J. P. T. (2006). Marker‐assisted combination of major genes for pathogen resistance in potato. Theor. Appl. Genet. 112, 1458–1464. Gibbon, B. C., and Larkins, B. A. (2005). Molecular genetic approaches to developing quality protein maize. Trends Genet. 21, 227–233. Gilchrist, E. J., and Haughn, G. W. (2005). TILLING without a plough: A new method with applications for reverse genetics. Curr. Opin. Plant Biol. 8, 1–5. Gill, B. S., Appels, R., Botha‐Oberholster, A.‐M., Buell, C. R., Bennetzen, J. L., Chalhoub, B., Chumley, F., Dvorˇa´k, J., Iwanaga, M., Keller, B., Li, W., McCombie, W. R., et al. (2004). A workshop report on wheat genome sequencing: International genome research on wheat consortium. Genetics 168, 1087–1096. Gimelfarb, A., and Lande, R. (1994). Simulation of marker‐assisted selection in hybrid populations. Genet. Res. 63, 39–47. Gimelfarb, A., and Lande, R. (1995). Marker‐assisted selection and marker‐QTL associations in hybrid populations. Theor. Appl. Genet. 91, 522–528. GoV, S. A., Ricke, D., Lan, T.‐H., Presting, G., Wang, R., Dunn, M., Glazebrook, J., Sessions, A., Oeller, P., Varma, H., Hadley, D., Hutchinson, D., et al. (2002). A draft sequence of the rice genome (Oryza sativa L. ssp. japonica). Science 296, 92–100.
290
S. L. DWIVEDI ET AL.
Gold, J., Harder, D., Townley‐Smith, F., Aung, T., and Procunier, J. (1999). Development of a molecular marker for rust resistance gene Sr39 and Lr35 in wheat breeding lines. Electron. J. Biotechnol. 2(1), (http://www.ejb.org/content/vol2/issue1/full/1). Graham, G. I., WolV, D. W., and Stuber, C. W. (1997). Characterization of a yield quantitative trait locus on chromosome five of maize by fine mapping. Crop Sci. 37, 1601–1610. Grapes, L., Dekkers, J. C. M., Rothschild, M. F., and Fernando, R. L. (2004). Comparing linkage disequilibrium‐based methods for fine mapping quantitative trait loci. Genetics 166, 1561–1570. Gu, S., Pakstis, A. J., and Kidd, K. K. (2005). HAPLOT: A graphical comparison of haplotype blocks, tagSNP sets and SNP variation for multiple populations. Bioinformatics 21, 3938–3939. Gu, W., Weeden, N., Yu, J., and Wallace, D. (1995). Large‐scale, cost‐eVective screening of PCR products in marker‐assisted selection. Theor. Appl. Genet. 91, 465–470. Guillaumie, S., Linossier, L., Torney, V., Robert, N., and Ravel, S. (2004). Co‐location between a gene encoding for the bZip factor SPA and an eQTL for high molecular weight glutenin subunit in wheat (Triticum aestivum). Genome 47, 705–713. Guillet‐Claude, C., Birolleau‐Touchard, C., Manicacci, D., Rogowsky, P. M., Rigau, J., Murigneux, A., Martinant, J.‐P., and Barrie`re, Y. (2004). Nucleotide diversity of the ZmPox3 maize peroxidase gene: Relationships between a MITE insertion in exon 2 and variation in forage maize digestibility. BMC Genet. 5, 19, doi: 10.1186/1471‐2156‐5‐19. Gulick, P. J., Drouin, S., Yu, Z., Danyluk, J., Poisson, G., Monroy, A. F., and Sarhan, F. (2005). Transcriptome comparison of winter and spring wheat responding to low temperature. Genome 48, 913–923. Guo, M., Rupe, M. A., Zinselmeier, C., Habben, J., Bowen, B. A., and Smith, O. S. (2004). Allelic variation of gene expression in maize hybrids. Plant Cell 16, 1707–1716. Guo, S.‐W. (2000). Genetic mapping of complex traits: Promises, problems, and prospects. Theor. Popul. Biol. 57, 1–11. Gupta, P. K., Rustgi, S., Sharma, S., Singh, R., Kumar, N., and Balyan, H. S. (2003). Transferable EST‐ SSR markers for the study of polymorphism and genetic diversity in bread wheat. Mol. Genet. Genomics 270, 315–323. Gupta, P. K., Rustgi, S., and Kulwal, P. L. (2005a). Linkage disequilibrium and association studies in higher plants: Present status and future prospects. Plant Mol. Biol. 57, 461–485. Gupta, S. K., Charpe, A., Koul, S., Prabhu, K. V., and Haq, Q. M. R. (2005b). Development and validation of molecular markers linked to an Aegilops umbellulata‐derived leaf rust‐resistance gene, Lr9, for marker‐assisted selection in bread wheat. Genome 48, 823–830. Gur, A., and Zamir, D. (2004). Unused natural variation can lift yield barriers in plant breeding. PLoS Biol. 2, 1610–1615. Haberer, G., Young, S., Bharati, A. K., Gundlach, H., Raymond, C., Fuks, G., Butler, E., Wing, R. A., Rounsley, S., Birren, B., Nusbaum, C., Mayer, K. F. X., et al. (2005). Structure and architecture of the maize genome. Plant Physiol. 139, 1612–1624. Halward, T. M., Stalker, H. T., and Kochert, G. (1993). Development of an RFLP linkage map in diploid peanut species. Theor. Appl. Genet. 87, 379–384. Ha¨ma¨la¨inen, J. H., Watanabe, K. N., Valkonen, J. P. T., Arihira, A., Plaisted, R. L., Pehu, E., Miller, L., and Slack, L. A. (1997). Mapping and marker‐assisted selection for a gene for extreme resistance to potato virus Y. Theor. Appl. Genet. 94, 192–197. Hamblin, M. T., and Aquadro, C. F. (1999). DNA sequence variation and the recombinational landscape in Drosophila pseudoobscura: A study of the second chromosome. Genetics 153, 859–869. Hammer, G. L., KropV, M. J., Sinclair, T. R., and Porter, J. R. (2002). Future contribution of crop modeling: From heuristics and supporting decision making to understanding genetic regulation and aiding crop improvement. Eur. J. Agron. 18, 15–31.
APPLIED CROP GENOMICS
291
Hammer, G. L., Chapman, S., van Oosterom, E., and Podlich, D. W. (2005). Trait physiology and crop modeling as a framework to link phenotypic complexity to underlying genetic systems. Aust. J. Agric. Res. 56, 947–960. Han, O. K., Kaga, A., Isemura, T., Wang, X. W., Tomooka, N., and Vaughan, D. A. (2005). A genetic linkage map for azuki bean [Vigna angularis (Willd.) Ohwi & Ohashi]. Theor. Appl. Genet. 111, 1278–1287. Harushima, Y., Yano, M., Shomura, A., Sato, M., Shimano, T., Kuboki, Y., Yamamoto, T., Lin, S. Y., Antonio, B. A., Parco, A., Kajiya, H., Huang, N., et al. (1998). A high‐density rice genetic linkage map with 2275 markers using a single F2 population. Genetics 148, 479–494. Hash, C. T. (2005). Opportunities for application of molecular markers for sustainable crop production in stress environments: Sorghum and pearlmillet. In ‘‘International Conference on Sustainable Crop Production in Stress Environments: Management and Genetic Options,’’ pp. 113 (abstract). Jawaharlal Nehru Krishi Vishwa Vidyalaya, Jabalpur, India. Hash, C. T., Rizivi, S. M. H., Serraj, S., Bidinger, F., Vadez, V., Sharma, A., Howarth, C. J., and Yadav, R. S. (2004). Field assessment of back‐cross derived hybrids validates a major pearlmillet drought tolerance QTL. In ‘‘49th Annual Meeting of Crop Science Society of America,’’ p. 211 (abstract). Seattle, Washington, USA. Haussmann, B. I. G., Hess, D. E., Seetharama, N., Welz, H. G., and Geiger, H. H. (2002). Construction of a combined sorghum linkage map from two recombinant inbred populations using AFLP, SSR, RFLP, and RAPD markers, and comparison with other sorghum maps. Theor. Appl. Genet. 105, 629–637. Hayama, R., and Coupland, G. (2004). The molecular basis of diversity in the photoperiodic flowering responses of Arabidopsis and rice. Plant Physiol. 135, 677–684. Hazen, S. P., Pathan, M. S., Sanchez, A., Baxter, I., Dunn, M., Estes, B., Chang, H.‐S., Zhu, T., Kreps, J. A., and Nguyen, H. T. (2005). Expression profiling of rice segregating for drought tolerance QTL using a rice genome array. Funct. Integr. Genomics 5, 104–116. Hedden, P. (2003). The genes of the green revolution. Trends Genet. 19, 5–19. Hehl, R., Faurie, E., Hesselbach, J., Salamini, F., Whitham, S., Baker, B., and Gebhardt, C. (1999). TMV resistance gene N homologues are linked to Synchytrium endobioticum resistance in potato. Theor. Appl. Genet. 98, 379–386. Helguera, M., Soria, V. M., Khan, I. A., Kolmer, J., and Dubcovsky, J. (2005). PCR markers for Triticum speltoides leaf rust resistance gene Lr51 and their use to develop isogenic hard red spring wheat lines. Crop Sci. 45, 728–734. Henderson, C. R. (1975). Best linear unbiased estimation and prediction under a selection model. Biometrics 31, 423–447. Herna´ndez‐Sa´nchez, J., Visscher, P., Plastow, G., and Haley, C. (2003). Candidate gene analysis for quantitative traits using the transmission disequilibrium test: The example of the Melanocortin 4‐ receptor in pigs. Genetics 164, 637–644. Heun, M., Kennedy, A. E., Anderson, J. A., Lapitan, N. L. V., Sorrels, M. E., and Tamksley, S. D. (1996). Construction of a restriction fragment length polymorphism map of barley (Hordeum vulgare). Genome 34, 437–447. Hittalmani, S., Parco, A., Mew, T. V., Zeigler, R. S., and Huang, N. (2000). Fine mapping and DNA marker‐assisted pyramiding of the three major genes for blast resistance in rice. Theor. Appl. Genet. 100, 1121–1128. Holland, J. B. (1998). EPISTACY: A SAS program for detecting tow‐locus epistasis interactions using genetic marker information. J. Hered. 89, 374–375. Holland, J. B. (2004). Implementation of molecular markers for quantitative traits in breeding programs—challenges and opportunities. In ‘‘New Directions for a Diverse Planet.’’ Proceedings of Fourth International Crop Science Congress, Brisbane, Australia. Hori, K., Kobayashi, T., Shimizu, A., Sato, K., Takeda, K., and Kawasaki, S. (2003). EYcient construction of high‐density linkage map and its application to QTL analysis in barley. Theor. Appl. Genet. 107, 806–813.
292
S. L. DWIVEDI ET AL.
Hospital, F., and Charcosset, M. (1997). Marker‐assisted introgression of quantitative loci. Genetics 147, 1469–1485. Hospital, F., and Decoux, G. (2002). Popmin: A program for the numerical optimization of population sizes in marker‐assisted programs. J. Hered. 93, 383–384. Hospital, F., Chevalet, C., and Mulsant, P. (1992). Using markers in gene introgression breeding programs. Genetics 132, 1119–1210. Hospital, F., Moreau, L., Lacoudre, F., Charcosset, A., and Gallais, A. (1997). More on the eYciency of marker‐assisted selection. Theor. Appl. Genet. 95, 1181–1189. Hu, J., and Vick, B. A. (2003). Target region amplification polymorphism: A novel marker technique for plant genotyping. Plant Mol. Biol. Rep. 21, 289–294. Hu, J.‐J., Nakatani, M., Mizuno, K., and Fujimura, T. (2004a). Development and characterization of microsatellite markers in sweetpotato. Breed. Sci. 54, 177–188. Hu, J.‐J., Nakatani, M., Lalusin, A.‐G., and Fujimura, T. (2004b). New microsatellite markers developed from reported Ipomoea trifida sequences and their application to sweetpotato and its related wild species. Sci. Hortic. 102, 375–386. Hua, J., Xing, Y., Wu, W., Xu, C., Sun, X., Yu, S., and Zhang, Q. (2003). Single‐locus heterotic eVects and dominance by dominance interactions can adequately explain the genetic basis of heterosis in an elite rice hybrid. Proc. Natl. Acad. Sci. USA 100, 2574–2579. Huama´n, Z., Aguilar, C., and Ortiz, R. (1999). Selecting a Peruvian sweetpotato core collection on the basis of morphological, eco‐geographical, and disease and pest reaction data. Theor. Appl. Genet. 98, 840–844. Huama´n, Z., Ortiz, R., Zhang, D., and Rodrı´guez, F. (2000). Isozyme analysis of entire and core collection of Solanum tuberosum subsp. andigena potato cultivars. Crop Sci. 40, 273–276. Huang, N., Angeles, E. R., Domingo, J., Magpantay, G., Singh, S., Zhang, G., Kumaravadivel, N., Bennet, J., and Khush, G. S. (1997). Pyramiding of bacterial blight resistance genes in rice: Marker‐ assisted selection using RFLP and PCR. Theor. Appl. Genet. 95, 313–320. Huang, X. Q., Hsam, S. L. K., Zeller, F. J., Wenzel, G., and Mohler, V. (2000). Molecular mapping of the wheat powdery mildew resistance gene Pm24 and marker validation for molecular breeding. Theor. Appl. Genet. 101, 407–414. Igartua, E., Edney, M., Rossnagel, B. G., Spaner, D., Legge, W. G., Scoles, G. J., Eckstein, P. E., Penner, G. A., Tinker, N. A., Briggs, K. G., Falk, D. E., and Mather, D. E. (2000). Marker‐ based selection of QTL aVecting grain and malt quality in two‐row barley. Crop Sci. 40, 1426–1433. Ikeda, A., Ueguchi‐Tanaka, M., Sonoda, Y., Kitano, H., Koshioka, M., Futsuhara, Y., Matsuoka, M., and Yamaguchi, J. (2001a). slender rice, a constitutive gibberellin response mutant, is caused by a null mutation of the SLR1 gene, an ortholog of the height‐regulating gene GAI/RGA/RHT/D8. Plant Cell 13, 999–1010. Ikeda, N., Bautista, N. S., Yamada, T., Kamijima, O., and Ishii, T. (2001b). Ultra‐simple DNA extraction method for marker‐assisted selection using microsatellite markers in rice. Plant Mol. Biol. Rep. 19, 27–32. IRGSP (2005). The map‐based sequence of the rice genome. Nature 436, 793–800. Ishimaru, K. (2003). Identification of a locus increasing rice yield and physiological analysis of its function. Plant Physiol. 122, 1083–1090. Izawa, T., Takahashi, Y., and Yano, M. (2003). Comparative biology comes into bloom: Genomic and genetic comparison of flowering pathways in rice and Arabidopsis. Curr. Opin. Plant Biol. 6, 113–120. Jaccoud, D., Peng, K., Feinstein, D., and Kilian, A. (2001). Diversity arrays: A solid state technology for sequence information independent genotyping. Nucleic Acids Res. 29, e25. Jannink, J. L., and Walsh, B. (2002). Association mapping in plant populations. In ‘‘Quantitative Genetics, Genomics and Plant Breeding’’ (M. S. Kong, Ed.), pp. 59–68. CABI, Wallingford, UK.
APPLIED CROP GENOMICS
293
Jansen, R. C., and Nap, J.‐P. (2001). Genetical genomics: The added value from segregation. Trends Genet. 17, 388–391. Jarret, R. L., and Bowen, N. (1994). Simple sequence repeats (SSRs) for sweetpotato germplasm characterization. Plant Genet. Resour. Newsl. 100, 9–11. JeVeries, S. P., Pallotta, M. A., Paull, J. G., Karakousis, A., Kretschmer, J. M., Manning, S., Islam, A. K. M. R., Langridge, P., and Chalmers, K. J. (2000). Mapping and validation of chromosome regions conferring boron toxicity tolerance in wheat (Triticum aestivum). Theor. Appl. Genet. 101, 767–777. JeVeries, S. P., King, B. J., Barr, A. R., Warner, P., Logue, S. J., and Langridge, P. (2003). Marker‐assisted backcross introgression of the Yd2 gene conferring resistance to barley yellow dwarf virus in barley. Plant Breed. 122, 52–56. Ji, W., Li, Y., Li, J., Dai, C.‐H., Wang, X., Bai, X., Cai, H., Wang, L., and Zhu, Y.‐M. (2006). Generation and analysis of expressed sequence tags from NaCL‐treated Glycine soja. BMC Plant Biol. 6, 4. Jiang, G. H., Xu, C. G., Tu, J. M., Li, X. H., He, Y. Q., and Zhang, Q. F. (2004). Pyramiding of insect‐ and disease‐resistance genes into an elite indica, cytoplasm male sterile restorer line of rice, Minghui 63. Plant Breed. 123, 112–116. Jin, C., Lan, H., Attie, A. D., Bulutuglo, D., Churchill, G. A., and Yandell, B. S. (2004). Selective phenotyping for increased eYciency in genetic mapping studies. Genetics 168, 2285–2293. Johnson, G. R. (2004). Marker assisted selection. Plant Breed. Rev. 24(Pt. 1), 293–310. Joseph, M., Gopalakrishnan, S., Sharma, R. K., Singh, V. P., Singh, A. K., Singh, N. K., and Mohapatra, T. (2003). Combining bacterial blight resistance and Basmati quality characteristics by phenotypic and molecular marker‐assisted selection in rice. Mol. Breed. 13, 1–11. Jourjon, M. F., Jasson, S., Marcel, J., Ngom, B., and Mangin, B. (2005). MCQTL: Multi‐allelic QTL mapping in multi‐cross design. Bioinformatics 21, 128–130. Juliano, B. O. (1985). ‘‘Rice Chemistry and Technology,’’ 2nd ed. American Association of Cereal Chemists, Incorporated, St. Paul, Minnesota, USA. Kanazin, V., Talbert, H., See, D., DeCamp, P., Nevo, E., and Glake, T. (2002). Discovery and single‐ nucleotide polymorphisms in barley (Hordeum vulgare L.). Plant Mol. Biol. 48, 529–537. Kapoor, R. L., Karkar, P. S., Khairwal, I. S., Bainiwal, C. R., Nijhawan, D. C., and Yadav, H. P. (1989). Bajra hybrid HHB67: A major breakthrough. Haryana Farm. 18, 17–21. Kasha, K. J., and Kao, K. N. (1970). High frequency haploid production in barley (Hordeum vulgare L.). Nature 225, 874–876. Katiyar, S. K., and Bennett, J. (2001). Biotechnology for gall midge resistance: From molecular tagging to gene pyramiding. In ‘‘Rice Research for Food Security and Poverty Alleviation’’ (S. Peng and B. Hardy, Eds.), pp. 692. Proceedings of International Rice Research Conference 31 March–3 April 2000. Keating, B. A., Carberry, P. S., Hammer, G. L., Probert, M. E., Robertson, M. J., Holzworth, D., Huth, N. I., Hargreaves, J. N. G., Meinke, H., Hochman, Z., McLean, G., Verburg, K., et al. (2003). An overview of APSIM, a model designed for farming system simulation. Eur. J. Agron. 18, 267–288. Kelly, J. D., Gepts, P., Miklas, P. N., and Coyne, D. P. (2003). Tagging and mapping of genes and QTL molecular marker‐assisted selection for traits of economic importance in bean and cowpea. Field Crop Res. 82, 135–154. Kendziorski, C. M., Chen, M., Yuan, M., Lan, H., and Attie, A. D. (2006). Statistical methods for expression quantitative trait loci (eQTL) mapping. Biometrics 62, 19–27. Kirst, M., Myburg, A. A., De Leo´n, J. P. G., Kirst, M. E., Scott, J., and SederoV, R. (2004). Coordinated genetic regulation of growth and lignin revealed by quantitative trait locus analysis of cDNA microarray data in an interspecific backcross of eucalyptus. Plant Physiol. 135, 2368–2378.
294
S. L. DWIVEDI ET AL.
Klein, P. E., Klein, R. R., Cartinhour, S. W., Ulanch, P. E., Dong, J., Obert, J. A., Morishige, D. T., Schlueter, S. D., Childs, K. L., Ale, M., and Mullet, J. E. (2000). A high‐throughput AFLP‐based method for constructing integrated genetic and physical maps: Progress toward a sorghum genome map. Genome Res. 10, 789–807. Klimyuk, V. I., Carroll, B. J., Thomas, C. M., and Jones, J. D. G. (1993). Alkali treatment for rapid preparation of plant material for reliable PCR analysis. Plant J. 3, 493–494. Klos, K. L. E., Paz, M. M., Marek, L. P., Cregan, P. B., and Shoemaker, R. C. (2000). Molecular markers useful for detecting resistance to brown stem rot in soybean. Crop Sci. 40, 1445–1452. Klose, J., Nock, C., Herrmann, M., Stu¨hler, K., Marcus, K., Blu¨ggel, M., Krause, E., Schalkwyk, L. C., Rastan, S., Brown, S. D. M., Bu¨ssow, K., Himmelbauer, H., et al. (2002). Genetic analysis of mouse brain proteome. Nat. Genet. 30, 385–393. Knapp, S. J. (1998). Marker‐assisted selection as a strategy for increasing the probability of selecting superior genotypes. Crop Sci. 38, 1164–1174. Kobayashi, S., Araki, E., Osaki, M., Khush, G. S., and Fukuta, Y. (2006). Localization, validation and characterization of plant‐type QTLs on chromosomes 4 and 6 in rice (Oryza sativa L.). Field Crop Res. 96, 106–112. Kobiljski, B., Quarrie, S., Dencˇic´, S., Kirby, J., and Ivegesˇ, M. (2002). Genetic diversity of the Novi Sad wheat core collection revealed by microsatellites. Cell. Mol. Biol. 7, 685–694. Koebner, R. (2003). MAS in cereals: Green for maize, amber for rice, still red for wheat and barley. Marker Assisted Seletion: A fast track to increase genetic gain in plant and animal breeding?Session 1: MAS in plants (http://www.fao.org/biotech/docs/koebner.pdf). Koebner, R. M. D., Powell, W., and Donini, P. (2001). Contributions of DNA molecular marker technologies to the genetics and breeding of wheat and barley. Plant Breed. Rev. 21, 181–220. Kojima, S., Takahashi, Y., Kobayashi, Y., Monna, L., Sasaki, T., Araki, T., and Yano, M. (2002). Hd3a, a rice ortholog of the Arabidopsis FT gene, promotes transition to flowering downstream of Hd1 under short‐day conditions. Plant Cell Physiol. 43, 1096–1105. Kollipara, K. P., Saab, I. N., Wych, R. D., Lauer, M. J., and Singletary, G. W. (2002). Expression profiling of reciprocal maize hybrids divergent for cold germination and desiccation tolerance. Plant Physiol. 129, 974–992. Kong, L., Dong, J., and Hart, G. E. (2000). Characteristics, linkage map positions, and allelic diVerentiation of Sorghum bicolor (L.) Moench DNA simple‐sequence repeats (SSRs). Theor. Appl. Genet. 101, 438–448. Konieczny, A., and Ausubel, F. A. (1993). A procedure for mapping Arabidopsis mutations using co‐dominant ecotype‐specific PCR‐based markers. Plant J. 4, 403–410. Konishi, S., Izawa, T., Lin, S. Y., Ebana, K., Fukuta, Y., Sasaki, T., and Yano, M. (2006). An SNP caused loss of seed shattering during rice domestication. Science 312, 1392–1396. Kosambi, D. D. (1944). The estimation of map distances from recombination values. Ann. Eugen. 12, 172–175. Kota, R., Rudd, S., Facius, A., Kolesov, G., Theil, T., Zhang, H., Stein, N., Mayer, K., and Graner, A. (2003). Snipping polymorphisms from large EST collections in barley (Hordeum vulgare L.). Mol. Genet. Genomics 270, 24–33. Kraakman, A. T. W., Martı´nez, F., Mussiraliev, B., van Eeuwijk, F. A., and Niks, R. E. (2006). Linkage disequilibrium mapping of morphological, resistance, and other agronomically relevant traits in modern spring barley cultivars. Mol. Breed. 17, 41–58. Kriegner, A., Cervantes, J. C., Burg, K., Mwanga, R. O. M., and Zhang, D. P. (2003). A genetic linkage map of sweetpotato (Ipomoea batatas (L) Lam) based on AFLP markers. Mol. Breed. 11, 169–185. Kroymann, J., and Mitchell‐Olds, T. (2005). Epistasis and balanced polymorphism influencing complex trait variation. Nature 435, 95–98. Kuchel, H., Ye, G.‐Y., Fox, R., and JeVeries, S. (2005). Genetic and economic analysis of a targeted marker‐assisted wheat breeding strategy. Mol. Breed. 16, 67–78.
APPLIED CROP GENOMICS
295
Kumpatla, S. P., and Mukhopadhyay, S. (2005). Mining and survey of simple sequence repeats in expressed sequence tags of dicotyledonous species. Genome 48, 985–998. Kwon, Y.‐S., Eun, M. Y., and Sohn, J.‐K. (2001). Marker‐assisted selection for identification of plant regeneration ability of seed‐derived calli in rice (Oryza sativa L.). Mol. Cells 12, 103–106. Kynast, R. G., Riera‐Lizarazu, O., Vales, M. I., Okagaki, R. J., Maquieira, S. B., Chen, G., Ananiev, E. V., Odland, W. E., Russell, C. D., Stec, A. O., Livingston, S. M., Zaia, H. A., et al. (2001). A complete set of monogenic individual chromosome additions to the oat genome. Plant Physiol. 125, 1216–1127. Lande, R., and Thompson, R. (1990). EYciency of marker‐assisted selection in the improvement of quantitative traits. Genetics 124, 743–756. Lander, E. S., Green, P., Abrahamson, J., Barlow, A., Daly, M. J., Lincoln, S. E., and Newburg, L. (1987). MAPMAKER: An interactive computer package for constructing primary genetic linkage maps of experimental and natural populations. Genomics 1, 174–181. Lander, E. S., Linton, M., Birren, B., Nusbaum, C., Zody, M. C., Baldwin, J., Devon, K., Dewar, K., Doyle, M., FitzHugh, W., Funke, R., Gage, D., et al. (2001). Initial sequencing and analysis of human genome. Nature 409, 860–921. Landi, P., Sanguineti, M. C., Salvi, S., Giuliani, S., Billotti, M., Maccaferri, M., Conti, S., and Tuberosa, R. (2005). Validation and characterization of a major QTL aVecting leaf ABA concentration in maize. Mol. Breed. 15, 291–303. Langridge, P., Lagudah, E. S., Holton, T. A., Appels, R., Sharp, P. J., and Chalmers, K. J. (2001). Trends in genetic and genome analyses in wheat: A review. Aust. J. Agric. Res. 52, 1043–1077. Larson, S. R., Kadyrzhanova, D., McDonald, C., Sorrells, M., and Blake, T. K. (1996). Evaluation of barley chromosome‐3 yield QTL in a backcross F2 population using STS‐ PCR. Theor. Appl. Genet. 93, 618–625. Latha, R., Rubia, L., Bennett, J., and Swaminathan, M. S. (2004). Allele mining for stress tolerance genes in Oryza species and related germplasm. Mol. Biotechnol. 27, 101–108. Law, C. N., Worland, A. J., and Giorgi, B. (1975). The genetic control of ear emergence time by chromosomes 5A and 5D of wheat. Heredity 36, 49–58. Lee, G. J., Boerma, H. R., Villagarcia, M. R., Zhou, X., Carter, T. E., Jr., Li, Z., and Gibbs, M. O. (2004). A major QTL conditioning salt tolerance in S‐100 soybean and descendent cultivars. Theor. Appl. Genet. 109, 1610–1619. Leung, H., Wu, J., Liu, B., Bustaman, M., Sridhar, R., Singh, K., Redona, E., Quang, V. D., Zhang, K., Bernardo, M., Wang, G., Leach, J., et al. (2004). Suscetainable disease resistance in rice: Current and future strategies. (http://www.cropscience org.au/icsc2004/symposia/3/ 7/1022_leungh.htm). Li, C., Zhou, A., and Sang, T. (2006b). Rice domestication by reducing shattering. Science 311, 1936–1939. Li, C. D., Rossnagel, B. G., and Scoles, G. J. (2000). The development of oat microsatellite markers and their use in identifying relationships among Avena species and oat cultivars. Theor. Appl. Genet. 101, 1259–1268. Li, J., Thomson, M., and McCouch, S. R. (2004a). Fine mapping of a grain‐weight quantitative trait locus in the pericentromeric region of rice chromosome 3. Genetics 168, 2187–2195. Li, J. Z., Sjakste, T. G., Ro¨der, M. S., and Ganal, M. W. (2003b). Development and genetic mapping of 127 new microsatellite markers in barley. Theor. Appl. Genet. 107, 1021–1027. Li, Y., Shi, Y., Cao, Y., and Wang, T. (2004b). Establishment of a core collection for maize germplasm preserved in Chinese national gene bank using geographic distribution and characterization data. Genet. Resour. Crop Evol. 51, 845–852. Li, Z., Jakkula, L., Hussey, R. S., Tamulonis, J. P., and Boerma, H. R. (2001a). SSR mapping and confirmation of the QTL from PI96354 conditioning soybean resistance to southern root‐knot nematode. Theor. Appl. Genet. 103, 1167–1173.
296
S. L. DWIVEDI ET AL.
Li, Z. K., Luo, L. J., Mei, H. W., Wang, D. L., Shu, Q. Y., Tabien, R., Zhong, D. B., Ying, C. S., Stansel, J. W., Khush, G. S., and Patrson, A. H. (2001b). Overdominant epistatic loci are the primary genetic basis of inbreeding depression and heterosis in rice. I. Biomass and grain yield. Genetics 158, 1737–1753. Li, Z. K., Yu, S. B., Lafitte, H. R., Huang, N., Courtois, B., Hittalmani, S., Vijayakumar, C. H. M., Liu, G. F., Wang, G. C., Shashidhar, H. E., Zhuang, J. Y., Zheng, K. L., et al. (2003a). QTL x environment interactions in rice. I. Heading date and plant height. Theor. Appl. Genet. 108, 141–153. Li, Z.‐K., Fu, B.‐Y., Gao, Y.‐M., Xu, J.‐L., Ali, J., Lafitte, H. R., Jiang, Y.‐Z., Rey, D., Vijayakumar, C. H. M., Maghirang, R., Zheng, T.‐Q., and Zhu, L.‐H (2005). Genome‐ wide introgression lines and their use in genetic and molecular dissection of complex phenotypes in rice (Oryza sativa L.). Plant Mol. Biol. 59, 33–52. Li, Z. K., Arif, M., Zhong, D. B., Fu, B. Y., Xu, J. L., Domingo‐Rey, J., Ali, J., Vijayakumar, C. H. M., Yu, S. B., and Khush, G. S. (2006a). Complex genetic networks underlying the defensive system of rice (Oryza sativa L.) to Xanthomonas oryzae pv. oryzae. Proc. Natl. Acad. Sci. USA 103, 7994–7999. Liang, F., Deng, Q., Wang, Y., Xiong, Y., Jin, D., Li, J., and Wang, B. (2004). Molecular marker‐assisted selection for yield‐enhancing genes in the progeny of ‘‘9311 x O. rufipogon’’ using SSR. Euphytica 139, 159–165. Lilley, J. M., Ludlow, M. M., McCouch, S. R., and O’Toole, J. C. (1996). Locating QTL for osmotic adjustment and dehydration tolerance in rice. J. Exp. Bot. 47, 1427–1436. Lin, S. C., and Min, S. K. (1991). ‘‘Rice Varieties and Their Genealogy in China (in Chinese).’’ Sanghai Science and Technolgy Press, Sanghaai. Lin, Y.‐R., Draye, X., Qian, X., Ren, S., Zhu, L.‐H., Tomkins, J., Wing, R., Li, Z., and Paterson, A. H. (2000). Locus‐specific contig assembly in highly‐duplicated genomes, using the BAC‐RF method. Nucleic Acid Res. 28, e23. Lindblad‐Toh, K., Winchester, E., Daly, M., Wang, D., Hirschhorn, J. N., Laviolette, J. P., Ardie, K., Reich, D. E., Robinson, E., Sklar, P., Shah, N., Thomas, D., et al. (2000). Large‐ scale discovery and genotyping of single‐nucleotide polymorphisms in the mouse. Nat. Genet. 24, 381–386. Little, R. R., Hilder, G. B., and Dawson, E. H. (1958). DiVerential eVects of dilute alkali on 25 varieties of milled white rice. Cereal Chem. 35, 111–126. Liu, S., and Anderson, J. A. (2003). Marker‐assisted evaluation of Fusarium head blight resistant wheat germplasm. Crop Sci. 43, 760–766. Liu, B., Zhang, S. H., Zhu, X. Y., Yang, Q. Y., Wu, S. Z., Mei, M. T., Mauleon, R., Leach, J., Mew, T., and Leung, H. (2004b). Candidate defense genes as predictors of quantitative blast resistance in rice. Mol. Plant Microbe. Interact. 17, 1146–1152. Liu, F., von Bothmer, R., and Salomon, B. (1999). Genetic diversity among East Asian accessions of the barley core collection as revealed by six isozyme loci. Theor. Appl. Genet. 98, 1226–1233. Liu, F., von Bothmer, R., and Salomon, B. (2000a). Genetic diversity in Eurpoean accessions of the barley core collection as detected by isozyme electrophoresis. Genet. Resour. Crop Evol. 47, 571–581. Liu, F., Sun, G.‐L., Salomon, B., and von Bothmer, R. (2001a). Distirbution of allozymic alleles and genetic diversity in the American barley core collection. Theor. Appl. Genet. 102, 606–615. Liu, J., Liu, D., Tao, W., Li, W., Wang, S., Chen, P., Cheng, S., and Gao, D. (2000b). Molecular marker‐facilitated pyramiding of diVerent genes for powdery mildew resistance in wheat. Plant Breed. 119, 21–24. Liu, J. S., Sabatti, C., Teng, J., Keats, B. J. B., and Risch, K. (2001b). Bayesian analysis of haplotypes for linkage disequilibrium mapping. Genome Res. 11, 1716–1724. Liu, P. Y., Zhu, J., and Lu, Y. (2004a). Marker‐assisted selection in segregating generations of self‐ fertilizing crops. Theor. Appl. Genet. 109, 370–376.
APPLIED CROP GENOMICS
297
Liu, Q.‐Q., Li, Q.‐F., Cai, X.‐L., Wang, H.‐M., Tang, S.‐Z., Yu, H.‐X., Wang, Z.‐Y., and Gu, M.‐H. (2006). Molecular marker‐assisted selection for improved cooking and eating quality of two elite parents of hybrid rice. Crop Sci. 46, 2354–2360, doi:10.2135/cropsci2006.03.0180. Liu, S., GriVey, C. A., and Saghai Maroof, M. A. (2001c). Identification of molecular markers associated with adult plant resistance to powdery mildew in common wheat cultivar Massey. Crop Sci. 41, 1268–1275. Liu, X., Gu, M., Han, Y., Ji, Q., Gu, S., Zhang, R., Li, X., Chen, J., Korban, S. S., and Xu, M. (2004c). Developing gene tagged markers for functional analysis of starch‐synthesizing genes in rice (Oryza sativa L.). Euphytica 135, 345–353. Liu, Z., Anderson, J. A., Hu, J., Friesen, T. L., Rasmussen, J. B., and Faris, J. D. (2005). A wheat intervarietal linkage map based on microsatellite and target region amplified polymorphism markers and its utility for detecting quantitative trait loci. Theor. Appl. Genet. 111, 782–794. Liu, Z.‐W., Biyashev, R. M., and Saghai Maroof, M. A. (1996). Development of simple sequence repeat DNA markers and their integration into a barley linkage map. Theor. Appl. Genet. 93, 869–876. Lopez, C., Jorge, V., Pie´gu, B., Mba, C., Cortes, D., Restrepo, S., Soto, M., Laudie´, M., Berger, C., Cooke, R., Delseny, M., Tohme, J., et al. (2004). A unique catalogue of 5700 expressed genes in cassava. Plant Mol. Biol. 56, 541–554. Lopez, C., Pie´gu, B., Cooke, R., Delseny, M., Tohme, J., and Verdier, V. (2005). Using cDNA and genomic sequences as tools to develop SNP strategies in cassava (Manihot esculenta Crantz). Theor. Appl. Genet. 110, 425–431. Lozoya‐Saldana, H., Belmar‐Diaz, C., Bradeen, J. M., and Helgeson, J. P. (2005). Chatacterization of Phytophthora infestans isolates infecting transgenic and somatic hybrid potatoes resistant to pathogen in the Toluca Valley, Mexico. Am. J. Potato Res. 82, 79(abstract). Lu, C., and Zhou, J. (2000). Breeding and utilization of two‐line interspecific hybrid rice Liangyou Peijiu. Hybrid Rice 15, 4–5. Lu, X., Niu, T., and Liu, J. S. (2003). Haplotype information and linkage disequilibrium mapping for single nucleotide polymorphisms. Genome Res. 13, 2112–2117. Lu¨bberstedt, T., Melchenger, A. E., Fa¨hr, S., Klein, D., Dally, A., and WesthoV, P. (1998). QTL mapping in test crosses of flint lines of maize: III. Comparison across populations for forage traits. Crop Sci. 38, 1278–1289. Lu¨bberstedt, T. L., Zein, I., Andersen, J. R., Wenzel, G., Kru¨tzfeldt, B., Eder, J., Quzunova, M., and Chun, S. (2005). Development and application of functional markers in maize. Euphytica 146, 101–108. Lund, M. S., Sorensen, P., Gulbrandsten, B., and Sorensen, D. A. (2003). Multitrate fine mapping of quantitative trait loci using combined linkage disequilibria and linkage analysis. Genetics 163, 405–410. Luo, L. J., Li, Z. K., Mei, H. W., Shu, Q. Y., Tabien, R., Zhong, D. B., Ying, C. S., Stansel, J. W., Khush, G. S., and Paterson, A. H. (2001). Overdominant epistatic loci are the primary genetic basis of inbreeding depression and heterosis in rice. II. Grain yield components. Genetics 158, 1755–1771. Lynch, M. N., and Walsh, B. (1998). ‘‘Genetics Analysis of Quantitative Traits.’’ Sinauer Associates, Sunderland, MA. Maccaferri, M., Sanguineti, M. C., Noli, E., and Tuberosa, R. (2005). Population structure and long‐range linkage disequilibrium in a durum wheat elite collection. Mol. Breed. 15, 271–289. Mackill, D. J. (2006). Breeding for resistance to abiotic stresses in rice: The value of quantitative trait loci. In ‘‘Plant Breeding: The Arnel R. Hallauer International Symposium’’ (K. R. Lamkey and M. Lee, Eds.), pp. 201–212. Blackwell Pub, Ames, IA. Mackill, D. J., and McNally, K. L. (2004). A model crop species: Molecular markers in rice. In ‘‘Molecular Marker Systems in Plant Preeding and Crop Improvement’’ (H. Lo¨rz and G. Wenzel, Eds.), Vol. 55, pp. 39–54. Springer Verlag, Heidelberg.
298
S. L. DWIVEDI ET AL.
Mahalakshmi, V., Ng, N., Lawson, M., and Ortiz, R. (2007a). Cowpea [Vigna unguiculata (L.) Walp.] core collection defined by geographical and agro‐botanical discriptors. Plant Genet. Resour. :Characterization and Utilization. (in press). Mahalakshmi, V., Ng, Q., Atalobhor, J., Ogunsola, D., Lawson, M., and Ortiz, R. (2007b). Development of a West African Yam Dioscorea spp. core collection. Genet. Resour. Crop Evol., doi: 10.1007/S10722‐006‐92034. (in press). Malosetti, M., and Abadie, T. (2001). Sampling strategy to develop a core collection of Uruguyan maize landraces based on morphological traits. Genet. Resour. Crop Evol. 48, 381–390. Malysheva‐Otto, L. V., Ganal, M. W., and Ro¨der, M. S. (2006). Analysis of molecular diversity, population structure and linkage disequilibrium in a worldwide survey of cultivated barley germplasm (Hordeum vulgare L.). BMC Genet. 7, 6. Manly, K. F. (1993). A Macintosh program for storage and analysis of experimental genetic mapping data. Mamm. Genome 4, 303–313. Manly, K. F., and Olsen, J. M. (1999). Overview of QTL mapping software and introduction to Map Manager QT. Mamm. Genome 10, 327–334. Manly, K. F., Cudmore, R. H., and Meer, J. M. (2001). Map Manager QTX, cross‐paltform software for genetic mapping. Mamm. Genome 12, 930–932. Martienssen, R. A., Rabinowicz, P. D., O’Shaughnessy, A., and McCombie, W. R. (2004). Sequencing the maize genome. Curr. Opin. Plant Biol. 7, 102–107. Martin, E. R., Monks, S. A., Warren, L. L., and Kaplan, N. L. (2000). A test for linkage and association in general pedigrees: The pedigree disequilibrium test. Am. J. Hum. Genet. 67, 146–154. Matsumura, H., Reich, S., Ito, A., Saitoh, H., Kamoun, S., Winter, P., Kahl, G., Reuter, M., Kruger, D. H., and Terauchi, R. (2003). Gene‐expression analysis of plant host‐pathogen interactions by SuperSAGE. Proc. Natl. Acad. Sci. USA 100, 15718–15723. Matus, I., Corey, A., Filichkin, T., Hayes, P. M., Vales, M. I., Kling, J., Riera‐Lizarazu, O., Sato, K., Powell, W., and Waugh, R. (2003). Development and characterization of recombinant chromosome substitution lines (RCSLs) using Hordeum vulgare subsp. spontaneum as a source of donor alleles in a Hordeum vulgare subsp. vulgare background. Genome 46, 1010–1023. Matz, E. C., Burr, F. A., and Burr, B. (1995). Molecular map based on TxCM and COxTx recombinant inbred families. Maize Genet. Coop. Newsl. 69, 257–267. Mazur, B. J. (1995). Commercializing the products of plant biotechnology. Trends Biotechnol. 13, 319–323. Mba, R. E. C., Stephenson, P., Edwards, K., Melzer, S., Nkumbira, J., Gullberg, U., Apel, K., Gale, M., Tohme, J., and Fregene, M. (2001). Simple sequence repeat (SSR) markers survey of the cassava (Manihot esculenta Crantz) genome: Towards an SSR‐based molecular genetic map of cassava. Theor. Appl. Genet. 102, 21–31. McCouch, S. R., Teytelman, L., Xu, Y., Lobos, K. B., Clare, K., Walton, M., Fu, B., Maghirang, R., Li, Z., Xing, Y., Zhang, Q., Kono, I., et al. (2002). Development and mapping of 2240 new SSR markers for rice (Oryza sativa L.). DNA Res. 9, 199–207. McLauchlan, A., Ogbonnaya, F. C., Hollingsworth, B., Carter, M., Gale, K. R., Henry, R. J., Holton, T. A., Morell, M. K., Rampling, L. R., Sharp, P. J., Shariflou, M. R., Jones, M. G. K., et al. (2001). Development of robust PCR‐based DNA markers for each homoeo‐allele of granule‐ bound starch synthase and their application in wheat breeding programs. Aust. J. Agric. Res. 52, 1409–1416. McNally, K. L., Wang, H., Naredo, E. B., Raghavan, C., Atienza, G., and Leung, H. (2006). ‘‘EcoTILLING in rice. Plant and Animal Genomes XIV Conference. W 258 (Allele mining)’’. San Diego, CA. Mejlhede, N., Kyjovska, Z., Backes, G., Burhenne, K., Rasmussen, S. K., and Jahoor, A. (2006). EcoTILLING for the identification of allelic variation in the powdery mildew resistance genes mlo and Mla of barley. Plant Breed. 125, 461–467.
APPLIED CROP GENOMICS
299
Melchinger, A. E., and Gumber, P. K. (1998). Overview of heterosis and heterotic groups in agronomic crops. In ‘‘Concepts and Breeding of Heterosis in Crop Plants’’ (K. R. Lamkey and J. E. Staub, Eds.), pp. 29–44. CSSA, Madison, WI. Melchinger, A. E., Utz, H. F., and Scho¨n, C. C. (1998). Quantitative trait locus (QTL) mapping using diVerent testers and independent population samples in maize reveals low power of QTL detection and large bias in estimates of QTL eVects. Genetics 149, 383–403. Melotto, M., Afanador, L., and Kelly, J. D. (1996). Development of a SCAR marker linked to the I gene in common bean. Genome 39, 1216–1219. Melotto, M., Monteiro‐Vitorello, C. B., Bruschi, A. G., and Camargo, L. E. A. (2005). Competitive bioinformatics analysis of genes expressed in common bean (Phaseolus vulgaris L.) seedlings. Genome 48, 562–570. Menz, M. A., Klein, R. R., Mullet, J. E., Obert, J. A., Unruh, N. C., and Klein, P. E. (2002). A high‐density genetic map of Sorghum bicolor (L.) Moench based on 2926 AFLP, RFLP, and SSR markers. Plant Mol. Biol. 48, 483–499. Mertz, E. T., Bates, L. S., and Nelson, O. E. (1964). Mutant gene that changes protein composition and increases lysine content of maize endosperm. Science 145, 279–280. Mesfin, A., Smith, K. P., Dill‐Macky, R., Evans, C. K., Waugh, R., Gustus, C. D., and Muehlbauer, G. J. (2003). Quantitative trait loci for Fusarium head blight resistance in barley detected in a two‐ rowed by six‐rowed population. Crop Sci. 43, 307–318. Messina, C. D., Jones, J. W., Boote, K. J., and Vallejos, C. E. (2006). A gene‐based model to simulate soybean development and yield response to environment. Crop Sci. 46, 456–466. Messing, J., Bharti, A. K., Karlowski, W. M., Gundblach, H., Kim, H. R., Yu, Y., Wei, F., Fuks, G., Soderlund, C. A., Mayer, K. F. X., and Wing, R. A. (2004). Sequence composition and genome organization of maize. Proc. Natl. Acad. Sci. USA 101, 14349–14354. Meuwissen, T. H. E., and Goddard, M. E. (2004). Mapping multiple QTL using linkage disequilibrium and linkage analysis information and multitrait data. Genet. Sel. Evol. 36, 261–269. Meuwissen, T. H. E., Hayes, B. J., and Goddard, M. E. (2001). Prediction of total genetic value using genome‐wide dense marker maps. Genetics 157, 1819–1829. Meyer, R. C., To¨rje´k, O., Becher, M., and Altmann, T. (2004). Heterosis of biomass production in Arabidopsis: Establishment during early development. Plant Physiol. 134, 1813–1823. Michalek, W., Weschke, W., Pleissner, K.‐P., and Graner, A. (2002). EST analysis in barley defines a unigene set comprising 4,000 genes. Theor. Appl. Genet. 104, 97–103. Mignouna, H. D., Mank, R. A., Ellis, T. H. N., van den Bosch, N., Asiedu, R., Ng, S. Y. C., and Peleman, J. (2002a). A genetic linkage map of Guinaea yam (Dioscorea rotundata Poir.) based on AFLP markers. Theor. Appl. Genet. 105, 716–725. Mignouna, H. D., Abang, M. M., Onasanya, A., Agindotan, B., and Asiedu, R. (2002b). Identification and potential use of RAPD markers linked to Yam mosaic virus resistance in white yam (Dioscorea rotundata). Ann. Appl. Biol. 140, 163–169. Mignouna, H. D., Abang, M. M., Onasanya, A., and Asiedu, R. (2002c). Identification and application of RAPD markers for anthracnose resistance in wateryam (Dioscorea alata). Ann. Appl. Biol. 141, 61–66. Mignouna, H. D., Abang, M. M., and Asiedu, R. (2003a). Harnessing modern biotechnology for tropical tuber crop improvement: Yam (Dioscorea spp.) molecular breeding. Afr. J. Biotechnol. 2, 478–485. Mignouna, H. D., Abang, M. M., and Fagbemi, S. A. (2003b). A comparative assessment of molecular marker assays (AFLP, RAPD, and SSR) for white yam (Discorea rotundata Poir.) germplasm characterization. Ann. Appl. Biol. 142, 269–276. Miklas, P. N., and Kelly, J. D. (2002). The use of MAS to develop pinto bean germplasm possessing Co‐42 gene for anthracnose resistance. Annu. Rep. Bean Improv. Coop. 45, 68–69.
300
S. L. DWIVEDI ET AL.
Miklas, P. N., Larsen, R. C., Riley, R., and Kelly, J. D. (2000). Potential marker‐assisted selection for bc‐I2 resistance to bean common mosaic potyvirus in common bean. Euphytica 116, 211–219. Miklas, P. N., Kelly, J. D., and Singh, S. P. (2003). Registration of anthracnose‐resistant pinto bean germplasm line USPT‐ANT‐1. Crop Sci. 43, 1889–1890. Miklas, P., Bosak, K., and Grafton, K. (2004). Marker‐assisted backcrossing of QTL for resistance to Sclerotinia white mold in pinto bean (http://www.whitemoldresearch.com/ presentation2004/Miklas.pdf). Miklas, P. N., Kelly, J. D., Beebe, S. E., and Blair, M. W. (2006a). Common bean breeding for resistance against biotic and abiotic stresses: From classical to MAS breeding. Euphytica 147, 105–131. Miklas, P. N., Hu, J., Niklas, J. G., and Larsen, K. M. (2006b). Potential application of TRAP (Targeted Region Amplified Polymorphism) markers for mapping and tagging disease resistance traits in common bean. Crop Sci. 46, 910–916. Milborrow, B. V. (1998). A biochemical mechanism for hybrid vigour. J. Exp. Bot. 49, 1063–1071. Milbourne, D., Meyer, R. C., Collins, A. J., Ramsay, L. D., Gebhardt, C., and Waugh, R. (1998). Isolation, characterization and mapping of simple sequence repeat loci in potato. Mol. Gen. Genet. 259, 233–245. Miyahara, K. (1999). Analysis of LGC1, low glutelin mutant of rice. Gamma Field Symp. 38, 43–52. Mohan, M., Nair, S., Bhagwat, A., Krishna, T. G., Yano, M., Bhatia, C. R., and Sasaki, T. (1997). Genome mapping, molecular markers and marker‐assisted selection in crop plants. Mol. Breed. 3, 87–103. Mohler, V., and Jahoor, A. (1996). Allele‐specific amplification of polymorphic sites for the detection of powdery mildew resistance loci in cereals. Theor. Appl. Genet. 93, 1078–1082. Moreau, L., Charcosset, A., Hospital, F., and Gallais, A. (1998). Marker‐assisted selection eYciency in populations of finite size. Genetics 148, 1353–1365. Moretzsohn, M. C., Leoi, L., Proite, K., Guimara˜es, P. M., Leal‐Bertioli, S. C. M., Gimenes, M. A., Martins, W. S., Valls, J. F. M., Grattapadaglia, D., and Bertioli, D. J. (2005). A microsatellite‐based, gene‐rich linkage map for the AA genome of Arachis (Fabaceae). Theor. Appl. Genet. 111, 1060–1071. Morrell, P. L., Toleno, D. M., Lundy, K. E., and Clegg, M. T. (2005). Low levels of linkage disequilibrium in wild barley (Hordeum vulgare ssp. spontaneum) despite high rate of self fertilization. Proc. Natl. Acad. Sci. USA 102, 2442–2447. Morris, M., Dreher, K., Ribaut, J. M., and Khairallah, M. (2003). Money matters (II): Costs of maize inbred line conversion schemes at CIMMYT using conventional and marker‐assisted selection. Mol. Breed. 11, 235–247. Mu, J., Zhou, H., Zhao, S., Xu, C., Yu, S., and Zhang, Q. (2004). Development of contiguous introgression lines covering entire genome of the sequenced Japonica rice. In ‘‘Proceedings of the Fourth International Crop Science Congress.’’ Brisbane, Australia (http://www. cropscience.org.au/icsc2004). Muehlbauer, G. J., Specht, J. E., Thomas‐Compton, M. A., Staswick, P. E., and Bernard, R. L. (1988). Near‐isogenic lines—A potential source in the integration of conventional and molecular genetic linkage maps. Crop Sci. 28, 729–735. Mutlu, N., Miklas, P. N., Steadman, J. R., Vidaver, A. K., Lindgren, D. T., Reiser, J., Coyne, D. P., and Paster‐Corrales, M. A. (2005). Registration of common bacterial blight resistant pinto bean germplasm line ABCP‐8. Crop Sci. 45, 805–806. Naess, S. K., Bradeen, J. M., Wielgus, S. M., Haberlach, G. T., McGrath, J. M., and Helgeson, J. P. (2000). Resistance to late blight in Solanum bulbocastanum is mapped to chromosome 8. Theor. Appl. Genet. 101, 697–704.
APPLIED CROP GENOMICS
301
Naik, S., Gill, K. S., Rao, V. S. P., Gupta, V. S., Tamhankar, S. A., Pujar, S., Gill, B. S., and Ranjekar, P. K. (1998). Identification of a STS marker linked to the Aegilops speltoides‐ derived leaf rust resistance gene Lr28 in wheat. Theor. Appl. Genet. 97, 535–540. Nair, S., Bentur, J. S., Rao, U. P., and Mohan, M. (1995). DNA markers tightly linked to a gall midge resistance gene (Gm2) are potentially useful for marker‐aided selection in rice breeding. Theor. Appl. Genet. 91, 68–73. Nair, S. K., Prasanna, B. M., Garg, A., Rathore, R. S., Setty, T. A. S., and Singh, N. N. (2005). Identification and validation of QTL conferring resistance to sorghum downy mildew (Peronosclerospora sorghi) and Rajasthan downy mildew (P. heteropogoni) in maize. Theor. Appl. Genet. 110, 1384–1392. Narayanan, N. N., Baisakh, N., Cruz, C. M. V., Gnanamanickam, S. S., Datta, K., and Datta, S. K. (2002). Molecular breeding for the development of blast and bacterial blight resistance in rice cv. IR50. Crop Sci. 42, 2072–2079. Narayanan, N. N., Baisakh, N., Oliva, N. P., VeraCruz, C. M., Gnanamanickam, S. S., Datta, K., and Datta, S. K. (2004). Molecular breeding: Marker‐assisted selection combined with biolistic transformation for blast and bacterial blight resistance in Indica rice (cv. CO39). Mol. Breed. 14, 61–71. Nasu, S., Suzuki, J., Ohta, R., Hasegawa, K., Yui, R., Kitazawa, N., Monna, L., and Minobe, Y. (2002). Search for and analysis of single nucleotide polymorphisms (SNPs) in rice (Oryza sativa, O. rufipogon) and establishment of SNP markers. DNA Res. 9, 163–171. Nelson, O. E. (2001). Maize: The long trail to QTM. In ‘‘Encyclopedia of Genetics’’ (E. C. R. Reeve and I. Black, Eds.), pp. 657–660. Fitzroy Dearborn. Ni, Z. F., Sun, Q. X., and Liu, Z. Y. (2000). Identification of a hybrid‐specific expressed gene encoding novel RNA‐binding protein in wheat seedling leaves using diVerential display of mRNA. Mol. Gen. Genet. 263, 934–938. Ni, Z. F., Sun, Q. X., Wu, L. M., and Xie, C. J. (2002). DiVerential gene expression between wheat hybrids and their parental inbreds in primary roots. Acta Bot. Sin. 44, 457–462. Nicot, N., Chiquet, V., Gandon, B., Amilhat, L., Legeai, F., Leroy, P., Bernard, M., and Sourdille, P. (2004). Study of simple sequence repeat (SSR) markers from wheat expressed sequence tags (ESTs). Theor. Appl. Genet. 109, 800–805. Niebur, W. S., Rafalski, J. A., Smith, O. S., and Cooper, M. (2004). Applications of genomics technologies to enhance rate of genetic progress for yield of maize within a commercial breeding program. In ‘‘New Directions for a Diverse Planet: Proceedings of the Fourth International Crop Science Congress’’ (T. Fischer, Ed.). Brisbane, Australia (www. cropscience.org.au). Nielsen, K. L., Grønkjaer, K., Welinder, K. G., and Emmersen, J. (2005). Global transcript profiling of potato tuber using LongSAGE. Plant Biotechnol. J. 3, 175–185. Nishimura, A., Ashikari, M., Lin, S., Takashi, T., Angeles, E. R., Yamamoto, T., and Matsuoka, M. (2005). Isolation of a rice regeneration quantitative trait loci gene and its application to transformation. Proc. Natl. Acad. Sci. USA 102, 11940–11944. O’Donoughue, L. S., Kianian, S. F., Rayapati, P. J., Penner, G. A., Sorrells, M. E., Tanksley, S. D., Phillips, R. L., Rines, H. W., Lee, M., Fedak, G., Molnar, S. J., HoVman, D., et al. (1995). A molecular linkage map of cultivated oat. Genome 38, 368–380. Ogbonnaya, F. E., Subrahmanyam, N. C., Moullet, O., De Majnik, J., Eagles, H. A., Brown, J. S., Eastwood, R. F., Kollmorgen, J., Appels, R., and Lagudah, E. S. (2001). Diagnostic DNA markers for cereal cyst nematode resistance in bread wheat. Aust. J. Agric. Res. 52, 1367–1374. Ojiambo, P. S., Nyanapah, J. O., Lung’aho, C., Karinga, J. K., and Kidanemariam, H. M. (2000). Comparing diVerent ecological models in field evaluations of selected genotypes from Solanum tuberosum CIP population A for resistance to Phytophthora infestans (Mont.) de Bary in Kenya. Euphytica 111, 211–218.
302
S. L. DWIVEDI ET AL.
Okada, Y., Kanatani, R., Arai, S., and Ito, K. (2003a). A CAPS marker that distinguishes the barley yellow mosaic disease resistance locus rym1 derived from Chinese landrace Mokusekko 3. J. Inst. Brew. 109, 103–105. Okada, Y., Kanatani, R., Arai, S., Asakura, T., and Ito, K. (2003b). Production of a novel virus‐ resistant barley line introgression to the rym1 locus with high malting quality using DNA marker assisted selection. J. Inst. Brew. 109, 99–102. Okogbenin, E., Marin, J., and Fergene, M. (2006). An SSR‐based molecular genetic map of cassava. Euphytica 147, 433–440. Olson, M., Hood, L., Cantor, C., and Botsetin, D. (1989). A common language for physical mapping of the human genome. Science 245, 1434–1435. Ordon, F., Bauer, E., Friedt, W., and Graner, A. (1995). Marker‐based selection for the ym4 BaMMV‐resistance gene in barley using RAPDs. Agronomie 15, 481–485. Oriero, C. E., Odunola, O. A., Lokko, Y., and Ingelbrecht, I. (2006). Analysis of B‐genome derived simple sequence repeat (SSR) markers in Musa spp. Afr. J. Biotechnol. 5, 126–128. Ortiz, R. (2003). An international public partnership for genetic enhancement of cowpea using a holistic approach to biotechnology. Genomic Proteomic Technol. 3, 45–47. Ortiz, R., Madsen, S., Ruiz‐Tapia, E. N., Jacobsen, S. E., Mujica‐Sa´nchez, A., Christiansen, J. L., and Stølen, O. (1999). Validating a core collection of Peruvian quinoa germplasm. Genetic Resour. Crop Evol. 46, 285–290. Ozturk, Z. N., Talame, V., Deyhoyos, M., Maichalowski, C. B., Galbriath, D. W., Gozukirmizi, N., Tuberosa, R., and Bohnert, H. J. (2002). Monitoring large‐scale changes in transcript abundance in drought‐ and salt‐stressed barley. Plant Mol. Biol. 48, 551–573. Palaisa, K. A., Morgante, M., Williams, M., and Rafalski, A. (2003). Contrasting eVect of selection on sequence diversity and linkage disequilibrium at two phytoene synthase loci. Plant Cell 15, 1795–1806. Palmer, L. E., Rabinowicz, P. D., O’Shaughnessy, A. L., Balija, V. S., Nascimento, L. U., Dike, S., de la Bastide, M., Martienssen, R. A., and McCombie, W. R. (2003). Maize genome sequencing by methylation filtration. Science 302, 2115–2117. Paltridge, N. G., Collins, N. C., Bendahmane, A., and Symons, R. H. (1998). Development of YLM, a co‐dominant PCR marker closely linked to the Yd2 gene for resistance to barley yellow dwarf disease. Theor. Appl. Genet. 96, 1170–1177. Paran, I., and Michelmore, R. W. (1993). Development of reliable PCR‐based markers linked to downy mildew resistance gene in lettuce. Theor. Appl. Genet. 85, 985–993. Paris, M., Potter, R. H., Lance, R. C. M., Li, C. D., and Jones, M. G. K. (2003). Typing Mlo alleles for powdery mildew resistance in barley by single nucleotide polymorphism analysis using MALDI‐ToF mass spectrometry. Aust. J. Argic. Res. 54, 1343–1349. Paterson, A. H., Lander, E. S., Hewitt, J. D., Paterson, S., Lincoln, S. E., and Tanksley, S. D. (1988). Resolution of quantitative traits into Mendelian factors, using a complete linkage map of restriction fragment length polymorphism. Nature 335, 721–726. Paterson, A. H., Damon, S., Hewitt, J. D., Zamir, D., Rabinowitch, H. D., Lincoln, S. E., Lander, E. S., and Tanksley, S. D. (1991). Mendelian factors underlying quantitative trait in tomato: Comparison across species, generations, and environments. Genetics 127, 181–197. Payne, P. I. (1987). Genetics of wheat storage proteins and their eVect of allelic variation on bread‐making quality. Annu. Rev. Plant Physiol. 38, 141–153. Peleman, J. D., Wye, C., Zethof, J., Sorensen, A. P., Verbakel, H., van Oeveren, J., Gerats, T., and van der Voort, J. R. (2005). Quantitative trait locus (QTL) isogenic recombinant analysis: A method for high‐resolution mapping of QTL within a single population. Genetics 171, 1341–1352. Peng, J., Richard, D. E., Hartley, N. M., Murphy, G. P., Devos, K. M., Flintham, J. E., Beales, J., Fish, L. J., Worland, A. J., Pelica, F., Sudhakar, D., Christou, P., et al. (1999). ‘Green revolution’ genes encode mutant gibberellin response modulators. Nature 400, 256–261.
APPLIED CROP GENOMICS
303
Perry, J. A., Wang, T. L., Welham, T. J., Gardner, S., Pike, J. M., Yoshida, S., and Parniske, M. (2003). A TILLING reverse genetics tool and a web‐accessible collection of mutants of the legume Lotus japonicus. Plant Physiol. 131, 866–871. Pestsova, E. G., Borner, A., and Ro¨der, M. S. (2001). Development of a set of Triticum aestivum—Aegilops tauschii introgression lines. Hereditas 135, 139–143. Pestsova, E. G., Borner, A., and Ro¨der, M. S. (2006). Development and QTL assessment of Triticum aestivum‐Aegilops tauschii introgression lines. Theor. Appl. Genet. 112, 634–647. Peterson, D. G., Schulze, S. R., Sciara, E. B., Lee, S. A., Bowers, J. E., Nagel, A., Jiang, N., Tibbitts, D. C., Wessler, S. R., and Paterson, A. H. (2002). Integration of Cot analysis, DNA cloning, and high throughput sequencing facilitate genome characterization and gene discovery. Genome Res. 12, 795–807. Picoult‐Newberg, L., Ideker, T. E., Pohl, M. G., Taylor, S. L., Donaldson, M. A., Nickerson, D. A., and Boyce‐Jacino, M. (1999). Mining SNPs from EST databases. Genome Res. 9, 167–174. Pinson, S. R. M., Capdevielle, F. M., and Ord, J. H. (2005). Confirming QTL and finding additional loci conditioning sheat blight resistance in rice using recombinant inbred lines. Crop Sci. 45, 503–510. Pinstrup‐Anderson, P., Pandya‐Lorch, R., and Rosegrant, M. W. (1999). World food prospects: Critical issues for the early twenty‐first century. 2020 Vision. Food policy report. International Food Policy Research Institute, Washington, DC, October 1999, 32pp. Podlich, D. W., and Cooper, M. (1998). QU‐GENE: A simulation platform for quantitative analysis of genetic models. Bioinformatics 14, 632–653. Podlich, D. W., Winkler, C. R., and Cooper, M. (2004). Mapping As You Go (MAYG): An eVective approach for marker‐assisted selection of complex traits. Crop Sci. 44, 1560–1571. Portyanko, V. A., HoVman, D. L., Lee, M., and Holland, J. B. (2001). A linkage map of hexaploid oat based on grass anchor DNA clones and its relationship to other oat maps. Genome 44, 249–265. Potokina, E., Caspers, M., Prasad, M., Kota, R., Zhang, H., Sreenivasulu, N., Wang, M., and Graner, A. (2004). Functional association between malting quality trait components and cDNA array based on expression patterns in barley (Hordeum vulgare L.). Mol. Breed. 14, 153–170. Powell, W., Machray, G. C., and Provan, J. (1996). Polymorphism revealed by simple sequence repeats. Trends Plant Sci. 7, 215–222. Prasad, M., Kumar, N., Kulwal, P. L., Roder, M. S., Balyan, H. S., Dhaliwal, H. S., and Gupta, P. K. (2003). QTL analysis for grain protein content using SSR markers and validation studies using NILs in bread wheat. Theor. Appl. Genet. 106, 659–667. Prasanna, B. M., Vasal, S. K., Kassahun, B., and Singh, N. N. (2001). Quality protein maize. Curr. Sci. 81, 1308–1319. Price, A. H. (2006). Believe it or not, QTLs are accurate. Trends Plant Sci. 11, 213–216. Pritchard, J. K., Stephens, M., and Donnelly, P. (2000a). Inference of population structure using multilocus genotype data. Genetics 155, 945–959. Pritchard, J. K., Stephens, M., Rosenberg, N. A., and Donnelly, P. (2000b). Association mapping in structured populations. Am. J. Human Genet. 67, 170–181. Qi, X., Lindup, S., Pittaway, T. S., Allouis, S., Gale, M. D., and Devos, K. M. (2001). Development of simple sequence repeat markers from bacterial artificial chromosomes without subcloning. BioTechniques 31, 355–362. Qi, X., Pittaway, T. S., Lindup, S., Liu, H., Waterman, E., Padi, F. K., Hash, C. T., Zhu, J., Gale, M. D., and Devos, K. M. (2004). An integrated genetic map and a new set of simple sequence repeat markers for pearlmillet, Pennisetum glaucum. Theor. Appl. Genet. 109, 1485–1493. Quarrie, S. A., Steed, A., Calestani, C., Semikhodskii, A., Lebreton, C., Chinoy, C., Steele, N., Pljevljakusic´, D., Waterman, E., Weyen, J., Schondelmaier, J., Habash, D. Z., et al. (2005).
304
S. L. DWIVEDI ET AL.
A high‐density genetic map of hexaploid wheat (Triticum aestivum L.) from the cross Chinese Spring x SQ1 and its use to compare QTLs for grain yield across a range of environments. Theor. Appl. Genet. 110, 865–880. Rabbani, M. A., Maruyama, K., Abe, H., Khan, M. Y., Katsura, K., Ito, Y., Yoshiwara, K., Seki, M., Shinozaki, K., and Yamaguchi‐Shinozaki, K. (2003). Monitoring expression profiles of rice gene under cold, drought, and high‐salinity stresses and abscisic acid application using cDNA microarray and RNA gel‐blot analyses. Plant Physiol. 133, 1755–1767. Rabinowitz, D. (1997). A transmission disequilibrium test for quantitative trait loci. Hum. Hered. 47, 342–350. Rabinowicz, P. D., Schulz, K., Dedhia, N., Yordan, C., Parnemm, L. D., Parnell, L. D., Stein, L., McCombie, R., and Martienssen, R. A. (1999). DiVerential methylation of genes and retrotransposons facilitates shot gun sequencing of maize genome. Nat. Genet. 23, 305–308. Radovanovic, N., and Cloutier, S. (2003). Gene‐assisted selection for high molecular weight glutenin subunits in wheat double haploid breeding programs. Mol. Breed. 12, 51–59. Rafalski, A., and Morgante, M. (2004). Corns and humans: Recombination and linkage disequilibrium in two genomes of similar size. Trends Genet. 20, 103–111. Ragagnin, V. A., Sanglard, D. A., de Souza, T. L. P. O., Moreira, M. A., and de Barros, E. G. (2003). Simultaneous transfer of resistance genes for rust, anthracnose, and angular leaf spot to cultivar Perola assisted by molecular markers. Annu. Rep. Bean Improv. Coop. 46, 159–160. Ragot, M., Biasiolli, M., Delbut, M. F., Dell’Orco, A., Malgarini, L., Thevenin, P., Vernoy, J., Vivant, J., Zimmermann, R., and Gay, G. (1995). Marker‐assisted backcrossing: A practical example. In ‘‘Techniques et utilisations des marqueurs mole´culaires’’ (A. Berville´ and M. Tersac, Eds.), pp. 45–56. Les Colloques, No. 72, INRA, Paris, France. Ragot, M., Gay, G., Muller, J.‐P., and Durovray, J. (2000). EYcient selection for the adaptation to the environment through QTL mapping and manipulation in maize. In ‘‘Molecular Approaches for the Genetic Improvement of Cereals for Stable Production in Water‐ Limited Environments’’ (J.‐M. Ribaut and D. Poland, Eds.), pp. 128–130. CIMMYT, Me´xico, DF. Raman, H., and Read, B. J. (2000). Molecular breeding for resistance against Russian wheat aphid in Australian barley. J. Agric. Genomics 5, 1–5. Raman, H., Karakousis, A., Moroni, S., Raman, R., and Read, B. (2001). Development, validation and allele diversity of microsatellite markers closely linked with the Alp locus with aluminium tolerance in barley. (http://www.regional.org.au/au/abts/2001/t4/raman2.htm) Raman, H., Platz, G. J., Chalmer, K. J., Raman, R., Read, B. J., Barr, A. R., and Moody, D. B. (2003). Mapping of genomic regions associated with net form of net blotch resistance in barley. Aust. J. Agric. Res. 54, 1359–1367. Ramalingam, J., Basharat, H. S., and Zhang, G. (2002). STS and microsatellite marker‐assisted selection for bacterial blight resistance and waxy gene in rice, Oryza sativa L. Euphytica 127, 255–260. Ramsay, L., Macaulay, M., degli Ivanissevich, S., MacLean, K., Cardle, L., Fuller, J., Edwards, K. J., Tuvesson, S., Morgante, M., Massari, A., Maestri, E., Marmiroli, N., et al. (2000). A simple sequence repeat‐based linkage map of barley. Genetics 156, 1997–2005. Rao, K. E. P., and Rao, V. R. (1995). The use of characterization data in developing a core collection of sorghum. In ‘‘Core Collections of Plant Genetic Resources’’ (T. Hodgkin, A. H. D. Brown, Th. J. L. van Hintum, and E. A. V. Morales, Eds.), pp. 109–115. John Wiley & Sons, Chichester, West Sussex, United Kingdom. Ravel, C., Praud, S., Murigneux, A., Linossier, L., Dardevet, M., Balfourier, F., Dufour, P., Brumel, D., and Charmet, G. (2006). Identification of Glu‐B1‐1 as a candidate gene for the
APPLIED CROP GENOMICS
305
quantity of high‐molecular‐weight glutenin in bread wheat (Triticum aestivum L.) by means of an association study. Theor. Appl. Genet. 112, 738–743. Redon˜a, E. D., and Mackill, D. J. (1996). Mapping quantitative trait loci for seedling vigor in rice using RFLPs. Theor. Appl. Genet. 92, 395–402. Reinke, R. F., Welsh, L. A., Reece, J. E., Lewin, L. G., and Blakenet, A. B. (1991). Procedures for quality selection in aromatic rice varieties. Int. Rice Res. Newsl. 16, 10–11. Ren, Z. H., Gao, J. P., Li, L. G., Cai, X. L., Huang, W., Chao, D. Y., Zhu, M. Z., Wang, Z. Y., Luan, S., and Lin, H. X. (2005). A rice quantitative trait locus for salt tolerance encodes a sodium transporter. Nat. Genet. 37, 1141–1146. Rensink, W., Hart, A., Liu, J., Ouyang, S., Zismann, V., and Buell, C. R. (2005). Analyzing the potato abiotic stress transcriptome using expressed sequence tags. Genome 48, 598–605. Reymond, M., Muller, B., Leonardi, A., Charcosset, A., and Tardieu, F. (2003). Combining quantitative trait loci analysis and an ecophysiological model to analyze the genetic variability of the responses of maize leaf growth to temperature and water deficit. Plant Physiol. 131, 664–675. Ribaut, J. M., and Betra´n, J. (2000). Single large‐scale marker‐assisted selection (SLS‐MAS). Mol. Breed. 5, 531–541. Ribaut, J. M., and Hoisington, D. (1998). Marker‐assisted selection: New tools and strategies. Trends Plant Sci. 3, 236–239. Ribaut, J. M., Hoisington, D. A., Deutsch, J. A., Jiang, C., and Gonza´lez‐de‐Leo´n, D. (1996). Identification of quantitative trait loci under drought conditions in tropical maize. I. Flowering parameters and the anthesis‐silking interval. Theor. Appl. Genet. 92, 905–914. Ribaut, J.‐M., Jiang, C., Gonza´lez‐de‐Leo´n, D., Edmeades, G. O., and Hoisington, D. A. (1997a). Identification of quantitative trait loci under drought conditions in tropical maize. II. Yield components and marker‐assisted selection strategies. Theor. Appl. Genet. 94, 887–896. Ribaut, J. M., Huu, X., Hoisington, D., and Gonzales de Leon, D. (1997b). Use of STSs and SSRs as rapid and reliable preselection tools in marker‐assisted selection backcross scheme. Plant Mol. Biol. Rep. 15, 156–164. Ribaut, J. M., Jiang, C., and Hoisington, D. (2002). Simulation experiments on eYciencies of gene introgression by backcrossing. Crop Sci. 42, 557–565. Richardson, K. L., Vales, M. I., Kling, J. G., Mundt, C. C., and Hayes, P. M. (2006). Pyramiding and dissecting disease resistance QTL to barley stripe rust. Theor. Appl. Genet. 113, 485–495. Rick, C. M. (1974). High soluble‐solids content in large‐fruited tomato lines derived from a wild green‐fruited species. Hilgardia 42, 493–510. Ritter, E., Debener, T., Barone, A., Salamini, F., and Gebhardt, C. (1991). RFLP mapping on potato chromosomes of two genes controlling extreme resistance to potato virus X (PVX). Mol. Gen. Genet. 227, 81–85. Rockman, M. V., and Kruglyak, L. (2006). Genetics of global gene expression. Nat. Rev. Genet. 7, 862–872. Ro¨der, M. S., Korzum, V., Wendehake, K., Plaschke, J., Tixier, M.‐H., Leroy, P., and Ganal, M. W. (1998). A microsatellite map of wheat. Genetics 149, 2007–2023. Romagosa, I., Han, F., Ullrich, S. E., Hays, P. M., and Wesenberg, D. M. (1999). Verification of yield QTL through realized molecular marker‐assisted selection responses in a barley cross. Mol. Breed. 5, 143–152. Ronning, C. M., Stegalkina, S. S., Ascenzi, R. A., Bougri, O., Hart, A. L., Utterbach, T. R., Vanaken, S. E., Riedmuller, S. B., White, J. A., Cho, J., Pertea, G. M., Lee, Y., et al. (2003). Comparative analysis of potato expressed sequence tag libraries. Plant Physiol. 131, 419–429. Rosegrant, M. W., Agcaoili‐Sombilla, M., and Perez, N. D. (1995). Global food projection to 2020: Implications for investment. Food, Agriculture, and the Environment Discussion Paper 5. International Food Policy Research Institute, Washington DC, October 1995, 54pp.
306
S. L. DWIVEDI ET AL.
Rostoks, N., Mudie, S., Cardle, L., Russell, J., Ramsay, L., Booth, A., Svensson, J. T., Wanamaker, S. I., Walia, H., Rodriguez, E. M., Hedley, P. E., Liu, H., et al. (2005). Genome‐wide SNP discovery and linkage analysis in barley based on genes responsive to abiotic stress. Mol. Gen. Genom. 274, 515–527. Roy, J. K., Bandopadhyay, R., Rustgi, S., Balyan, H. S., and Gupta, P. K. (2006). Association analysis of agronomically important traits using SSR, SAMPL and AFLP markers in bread wheat. Curr. Sci. 90, 683–689. Rudd, S. (2003). Expressed sequence tags: Alternative or complement to whole genome sequencing. Trends Plant Sci. 8, 321–329. Saiki, R. K., Gelfand, D. H., StoVel, S., Scharf, S., Higuchi, R. H., Horn, G. T., Mullis, K. B., and Erlich, H. A. (1988). Primer‐directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239, 487–491. Saito, K., Miura, K., Nagano, K., Hayano‐Saito, Y., Araki, H., and Kato, A. (2001). Identification of two closely linked quantitative trait loci for cold tolerance on chromosome 4 of rice and their association with anther length. Theor. Appl. Genet. 103, 862–868. Salvi, S., Tuberosa, R., Chiapparino, E., Maccaferri, M., Veillet, S., van Beuningen, L., Isaac, P., Edwards, K., and Phillips, R. L. (2002). Toward positional cloning of Vgt1, a QTL controlling the transition from the vegetative to the reproductive phase in maize. Plant Mol. Biol. 48, 601–613. Sanchez, A. C., Brar, D. S., Huang, N., Li, Z., and Khush, G. S. (2000). Sequence tagged site marker‐ assisted selection for three bacterial blight resistance genes in rice. Crop Sci. 40, 792–797. Santos, C. M. R., Martins, N. F., Ho¨rberg, H. M., de Almeida, E. R. P., Coelho, M. C. F., Togawa, R. C., da Silva, F. R., Caetano, A. R., Miller, R. N. G., and Souza, M. T., Jr. (2005). Analysis of expressed sequence tags from Musa acuminata ssp. burmannicoides, var. Calcutta 4 (AA) leaves submitted to temperature stresses. Theor. Appl. Genet. 110, 1517–1522. Sawkins, M. C., Farmer, A. D., Hoisington, D., Sullivan, J., Tolopko, A., Jiang, Z., and Ribaut, J.‐M. (2004). Comparative map and trait viewer (CMTV): An integrated bioinformatic tool to construct consensus maps and compare QTL and functional genomics data across genomes and experiments. Plant Mol. Biol. 56, 465–480. Scarcelli, N., Dainou, O., Agbangla, C., Tostain, S., and Pham, J.‐L. (2005). Segregation patterns of isozyme loci and microsatellite markers show the diploidy of African yam Dioscorea rotundata (2n ¼ 40). Theor. Appl. Genet. 111, 226–232. Schachermayr, G., Sielder, H., Gale, M. D., Winzeler, H., Winzeler, M., and Keller, B. (1994). Identification and localization of molecular markers linked to the Lr9 leaf rust resistance gene of wheat. Theor. Appl. Genet. 88, 110–115. Schachermayr, G., Feuillet, C., and Keller, B. (1997). Molecular markers for the detection of the wheat leaf rust resistance gene Lr10 in diverse genetic backgrounds. Mol. Breed. 3, 65–74. Schadt, E. E., Monks, S. A., Drake, T. A., Lusis, A. J., Che, N., Colinayo, V., RuV, T. G., Milligan, S. B., Lamb, J. R., Cavet, G., Linsley, P. S., Mao, M., et al. (2003). Genetics of gene expression surveyed in maize, mouse and man. Nature 422, 297–302. Scheuring, C., Barthelson, R., Gailbraith, D., Betran, J., Cothren, J. T., Zeng, Z.‐B., and Zhang, H.‐B (2006). Preliminary analysis of diVerential gene expression between a maize superior hybrid and its parents using the 57K maize gene‐specific long‐oligonucleotide microarray. In ‘‘48th Annual Maize Genetic Conference,’’ 9–12 March 2006. California. 132pp. Schmierer, D. A., Kandemir, N., Kudrna, D. A., Jones, B. L., Ullrich, S. E., and Kleinhofs, A. (2004). Molecular marker‐assisted selection for enhanced yield in malting barley. Mol. Breed. 14, 463–473. Schneider, K. A., Brothers, M. E., and Kelly, J. D. (1997). Marker‐assisted selection to improve drought resistance in common bean. Crop Sci. 37, 51–60.
APPLIED CROP GENOMICS
307
Schnurbusch, T., Bossolini, E., Messmer, M., and Keller, B. (2004). Tagging and validation of a major quantitative trait locus for leaf rust resistance and leaf tip necrosis in winter wheat cultivar Forno. Phytopathology 94, 1036–1041. Scho¨n, C. C., Utz, H. F., Groh, S., Truberg, B., Openshaw, S., and Melchinger, A. E. (2004). Quantitative trait locus mapping based on resampling in a vast maize testcross experiment and its relevance to quantitative genetics for complex traits. Genetics 167, 485–498. Schwarz, G., Felsenstein, F. G., and Wenzel, G. (2004). Development and validation of a PCR‐ based marker assay for negative selection of the HMW glutenin allele Glu‐B1‐1d (Bx‐6) in wheat. Theor. Appl. Genet. 109, 1064–1069. Seaton, G., Haley, C. S., Knott, S. A., Kearsey, M., and Visscher, P. M. (2002). QTL Express: Mapping quantitative trait loci in simple and complex pedigrees. Bioinformatics 18, 339–340. ˚ ., Skinnes, H., Marøy, A. G., Tarkegne, Y., and William, M. (2006). Semagn, K., Bjørnstad, A Distribution of DArT, AFLP, and SSR markers in a genetic linkage map of a doubled‐ haploid hexaploid wheat population. Genome 49, 545–555. Semel, Y., Nissembaum, J., Menda, N., Zinder, M., Krieger, U., Issman, N., Pleban, T., Lippman, Z., Gur, A., and Zamir, D. (2006). Overdominant quantitative trait loci for yield and fitness in tomato. Proc. Natl. Acad. Sci. USA 103, 12981–12986. Senior, M. L., and Heun, M. (1993). Mapping maize microsatellites and polymerase chain reaction confirmation of the targeted repeats using a CT primer. Genome 36, 884–889. Sethy, N. K., Shokeen, B., Edwards, K. J., and Bhatia, S. (2006). Development of microsatellite markers and analysis of interspecific genetic variability in chickpea (Cicer arietinum L.). Theor. Appl. Genet. 112, 1416–1428. Sharma, P. N., Torii, A., Takumi, S., Mori, N., and Nakamura, C. (2004). Marker‐assisted pyramiding of brown planthopper (Nilaparvata lugens Stal) resistance genes Bph1 and Bph2 on rice chromosome 12. Hereditas 140, 61–69. Sharopova, N., McMullen, M. D., Schultz, L., Schroeder, S., Sanchez‐Villeda, H., Gardiner, J., Bergstrom, D., Houchins, K., Melia‐Hancock, S., Musket, T., Duru, N., Polacco, M., et al. (2002). Development and mapping of SSR markers for maize. Plant Mol. Biol. 48, 463–481. Sharp, P. J., Johnston, S., Brown, G., McIntosh, R. A., Pallotta, M., Carter, M., Bariana, H. S., Khatkar, S., Lagudah, E.S, Singh, R. P., Khairallah, M., Potter, R., et al. (2001). Validation of molecular markers for wheat breeding. Aust. J. Agric. Res. 52, 1357–1366. Shen, L., Courtois, B., McNally, K. L., Robin, S., and Li, Z. (2001). Evaluation of near‐isogenic lines of rice introgressed with QTLs for root depth through marker‐aided selection. Theor. Appl. Genet. 103, 75–83. Shubing, L., Zhou, R., Dong, Y., Li, P., and Jia, J. (2006). Development, utilization of introgression lines using a synthetic wheat as donor. Theor. Appl. Genet. 112, 1360–1373. Siangliw, M., Toojinda, T., Tragoonrung, S., and Vanavichit, A. (2003). Thai Jasmine rice carrying QTLch9 (SubQTL) is submergence tolerant. Ann. Bot. 91, 255–261. Singh, H., Prasad, M., Varshney, R. K., Roy, J. K., Balyan, H. S., Dhaliwal, H. S., and Gupta, P. K. (2001b). STMS marker for grain protein content and their validation using near‐isogenic lines in bread wheat. Plant Breed. 120, 273–278. Singh, R., Datta, D., Priyamvad, Singh, S., and Tiwari, R. (2004). Marker‐assisted selection for leaf rust resistance genes Lr19 and Lr24 in wheat (Triticum aestivum L.). J. Appl. Genet. 45, 399–403. Singh, R. J., Kollipara, K. K., and Hymowtiz, T. (1998). Monosomic alien addition lines derived from Glycine max (L.) Merr. and G. tomentella Hayata: Production, characterization, and breeding behavior. Crop Sci. 38, 1483–1489. Singh, S., Sindhu, J. S., Huang, N., Vikal, Y., Li, Z., Brar, D. S., Dhaliwal, H. S., and Khush, G. S. (2001a). Pyramiding three bacterial blight resistance genes (xa5, xa13, and Xa21) using marker‐assisted selection into indica rice cultivar PR106. Theor. Appl. Genet. 102, 1011–1015.
308
S. L. DWIVEDI ET AL.
Slade, A. J., Fuerstenberg, S. I., LoeZer, D., Steine, M. N., and Facciotti, D. (2005). A reverse genetic, nontransgenic approach to wheat crop improvement by TILLING. Nat. Biotechnol. 23, 75–81. Slikova, S., Gregova, E., Bartos, P., and Kraic, J. (2003). Marker‐assisted selection for leaf rust resistance in wheat by transfer of gene Lr19. Plant Protect. Sci. 39, 13–17. Smits, B. M., Mudde, J., Cuppen, E., and Plasterk, R. H. (2004). Target‐selected mutagenesis of the rat. Genomics 83, 332–334. Sobrizal, K., Ikeda, K., Sanchez, P. L., Doi, K., Angeles, E. R., Khush, G. S., and Yoshimura, A. (1999). Development of Oryza glumaepatulla introgression lines in rice, O. sativa L. Rice Genet. Newsl. 16, 107. Somers, D. J., Kirckpatrick, R., Moniwa, M., and Walsh, A. (2003). Mining single‐nucleotide polymorphisms from hexaploid ESTs. Genome 49, 431–437. Somers, D. J., Thomas, J., DePauw, R., Fox, S., Humphreys, G., and Fedak, G. (2005). Assembling complex genotypes to resist Fusarium in wheat (Triticum aestivum L.). Theor. Appl. Genet. 111, 1623–1631. Song, J., Bradeen, J. M., Naess, S. K., Raasch, J. A., Wielgus, S. M., Haberlach, G. T., Liu, J., Austin‐Phillips, S., Buell, C. R., Helgeson, J. P., and Jiang, J. (2003). Gene RB cloned from Solanum bulbocastanum confers broad spectrum resistance to potato late blight. Proc. Natl. Acad. Sci. USA 100, 9128–9133. Song, Q. J., Shi, J. R., Singh, S., Fickus, E. W., Costa, J. M., Lewis, J., Gill, B. S., Ward, R., and Cregan, P. B. (2005). Development and mapping of microsatellite (SSR) markers in wheat. Theor. Appl. Genet. 110, 550–560. Song, R., and Messing, J. (2003). Gene expression of a gene family in maize based on noncollinear haplotypes. Proc. Natl. Acad. Sci. USA 100, 9055–9060. Spaner, D., Rossnagel, B. G., Legge, W. G., Scoles, G. J., Eckstein, P. E., Penner, G. A., Tinker, N. A., Briggs, K. G., Falk, D. E., Afele, J. C., Hayes, P. M., and Mather, D. E. (1999). Verification of a quantitative trait locus aVecting agronomic traits in two‐row barley. Crop Sci. 39, 248–252. Spielman, R. S., McGinnis, R. E., and Ewens, W. J. (1993). Transmission test for linkage disequilibrium: The insulin gene region and insulin‐dependent diabetes mellitus (IDDM). Am. J. Hum. Genet. 52, 506–516. Spielmeyer, W., Sharp, P. J., and Lagudah, E. S. (2003). Identification and validation of markers linked to broad‐spectrum stem rust resistance gene by Sr2 in wheat (Triticum aestivum L.). Crop Sci. 43, 333–336. Stam, P. (2003). Marker‐assisted introgression: Speed at any cost? In ‘‘EUCARPIA Leafy Vegetables 2003’’ (Th. J. L. van Hintum, A. Lebeda, D. Pink, and J. W. Schut, Eds.), pp. 117–124. EUCARPIA, Valencia, Spain. Stavely, J. R., and McMillan, R. T., Jr. (1992). BARC‐rust resistant bush, fresh‐market green bean germplasm lines. HortScience 27, 1052–1054. Stavely, J. R., and Steinke, J. (1990). BARC‐rust resistant bush waxy bean germplasm lines. HortScience 25, 1451–1452. Stavely, J. R., Kelly, J. D., and Grafton, K. F. (1994). BelMiDak rust resistant navy dry beans germplasm lines. HortScience 29, 709–710. Stavely, J. R., Kelly, J. D., Grafton, K. F., Mullins, C. A., Straw, A., McMillan, R. T., Jr., Beaver, J. S., Miklas, P. N., Steinke, J., Steadman, J. R., Coyne, D. P., Lindgren, D. T., et al. (1997). Rust resistant bean germplasm releases, 1994–96. Annu. Rep. Bean Improv. Coop. 40, 120–121. Stavely, J. R., McMillan, R. T., Beaver, J. S., and Miklas, P. N. (2001). Release of three McCaslan type, indeterminate, rust and golden mosaic resistant snap bean germplasm lines BelDade RGMR 4, 5, and 6. Annu. Rep. Bean Improv. Coop. 44, 197–198.
APPLIED CROP GENOMICS
309
Steele, K. A., Price, A. H., Shashidhar, H. E., and Witcombe, J. R. (2006). Marker‐assisted selection to introgress rice QTL controlling root traits into an Indian upland rice variety. Theor. Appl. Genet. 112, 208–221. Stemple, D. L. (2004). TILLING—a high‐throughput harvest for functional genomics. Nat. Rev. Genet. 5, 145–150. Stepien, L., Chelkowski, J., Wenzel, G., and Mohler, V. (2004). Combined use of linked markers for genotyping the Pm1 locus in common wheat. Cell. Mol. Biol. Lett. 9, 819–827. Stich, B., Maurer, H. P., Melchinger, A. E., Frisch, M., Heckenberger, M., van der Voort, J. R., Peleman, J., Sørensen, A. P., and Reif, J. C. (2006). Comparison of linkage disequilibrium in elite European maize inbred lines using AFLP and SSR markers. Mol. Breed. 17, 217–226. Stratton, D. A. (1998). Reaction norm functions and QTL‐environments for flowering time in Arabidopsis thaliana. Heredity 81, 144–155. Stromberg, L. D., Dudley, J. W., and Rufener, G. K. (1994). Comparing conventional early generation selection with molecular marker assisted selection in maize. Crop Sci. 34, 1221–1225. Stuber, C. W., Lincoln, S. E., WolV, D. W., Helentjaris, T., and Lander, E. S. (1992). Identification of genetic factors contributing to heterosis in a hybrid from two elite maize inbred lines using molecular markers. Genetics 132, 823–839. Stuber, C. W., Polacco, M., and Senior, M. L. (1999). Strategy of empirical breeding, marker‐ assisted selection, and genomics to increase crop yield potential. Crop Sci. 39, 1571–1583. Stuper, R., and Springer, N. (2006). Analysis of allelic variation in gene expression in B73 and Mo17 and implications for hybrid expression patterns. In ‘‘48th Annual Maize Genetic Conference,’’ 9–12 March 2006. California. p. 115. Sun, D. J., He, Z. H., Xia, X. C., Zhang, L. P., Morris, C., Appels, R., Ma, W., and Wang, H. (2005). A novel STS marker for polyphenol oxidase activities in bread wheat. Mol. Breed. 16, 209–218. Sun, Q. X., Ni, Z. F., and Liu, Z. Y. (1999). DiVerential gene expression between wheat hybrids and their parental inbreds in seedling leaves. Euphytica 106, 117–123. Sun, Q. X., Wu, L. M., Ni, Z. F., Meng, F. R., Wang, Z. K., and Lin, Z. (2004). DiVerential gene expression patterns in leaves between hybrids and their parental inbreds are correlated with heterosis in a wheat diallel cross. Plant Sci. 166, 651–657. Swaminathan, M. S. (2006). An evergreen revolution. Crop Sci. 46, 2293–2303. Swamy, P., Panchbhai, A. N., Dodiya, P., Naik, V., Panchbhai, S. D., Zehr, U. B., Azhakanandam, K., and Bharat, R. (2004). Evaluation of bacterial blight resistance in rice lines carrying multiple resistance genes and Xa21 transgenic lines. In ‘‘Proceedings of the Fourth International Crop Science Congress.’’ Australia ( http://www.regional.org.au/ au/cs/2004/poster/3/7/4/251_charbr.htm). Swanson‐Wagner, R. A., Jia, Y., DeCook, R., Borsuk, L. A., Nettleton, D., and Schnable, P. S. (2006). All possible modes of gene action are observed in a global camparison of gene expression in a maize F1 hybrid and its inbred parents. Proc. Natl. Acad. Sci. USA 103, 6805–6810. Swennen, R., and Vuylsteke, D. (1987). Morphological taxonomy of plantains (Musa cultivars AAB) in West Africa. In ‘‘Banana and Plantain Breeding Strategies’’ (G. J. Persley and E. A. de Langhe, Eds.), pp. 165–171. Proc. Workshop held at Cairns, Australia. ACIAR Proc. No. 26, pp. 13–17. Syed, N. H., and Chen, Z. J. (2005). Molecular marker genotypes, heterozygosity and genetic interactions explain heterosis in Arabidopsis thaliana. Heredity 94, 295–304. Szalma, S. J., Buckler, E. S., IV, Snook, M. E., and McMullen, M. D. (2005). Association analysis of candidate genes for maysin and chlorogenic acid accumulation in maize silks. Theor. Appl. Genet. 110, 1324–1333.
310
S. L. DWIVEDI ET AL.
Taba, S., Diaz, J., Franco, J., and Cross, J. (1998). Evalution of Caribbean maize accessions to develop a core subset. Crop Sci. 38, 1378–1386. Takahashi, Y., Shomura, A., Sasaki, T., and Yano, M. (2001). Hd6, a rice quantitative trait locus involved in photoperiod sensitivity, encodes the a subunit of protein kinase CK2. Proc. Natl. Acad. Sci. USA 98, 7922–7927. Tan, G., Jin, H., Li, G., He, R., Zhu, L., and He, G. (2005). Production and characterization of a complete set of individual chromosome additions from Oryza sativa using RFLP and GISH analysis. Theor. Appl. Genet. 111, 1585–1595. Tang, J., Xia, H., Li, D., Cao, M., Tao, Y., Tong, W., Zhang, X., Hu, S., Wang, J., Yu, J., Yang, H., and Zhu, L. (2005). Gene expression profiling in rice young panicle and vegetative organs and identification of panicle‐specific genes through known gene functions. Mol. Gen. Genom. 274, 467–476. Tang, J., Gao, L., Cao, Y., and Jia, J. (2006). Homologous analysis of SSR‐ESTs and transferability of wheat SSR‐EST markers across barley, rice and maize. Euphytica 151, 87–93. Tan, Y. F., Li, J. X., Yu, S. B., Xing, Y. Z., Xu, C. G., and Zhang, Q. (1999). The three important traits for cooking and eating quality of rice grains are controlled by a single locus in an elite rice hybrid, Shanyou 63. Theor. Appl. Genet. 99, 642–648. Tan, Y. F., Xing, Y. Z., Li, J. X., Yu, S. B., Xu, C. G., and Zhang, Q. (2000). Genetic basis of the appearance quality of rice grains in Shanyou 63, an elite rice hybrid. Theor. Appl. Genet. 101, 823–829. Tanhuanpa¨a¨, P., Kalendar, R., Laurila, J., Schulman, A. H., Manninen, O., and Kiviharju, E. (2006). Generation of SNP markers for short straw in oat (Avena sativa L.). Genome 49, 282–287. Tanksley, S. D. (1993). Mapping polygenes. Annu. Rev. Genet. 27, 205–233. Tanksley, S. D., and Nelson, J. C. (1996). Advanced backcross QTL analysis: A method for the simultaneous discovery and transfer of valuable QTLs from unadapted germplasm into elite breeding lines. Theor. Appl. Genet. 92, 191–203. Tanksley, S. D., Young, N. D., Paterson, A. H., and Bonierbale, M. W. (1989). RFLP mapping in plant breeding: New tools for an old science. Biotechnology 7, 257–264. Tanksley, S. D., Ganal, M. W., Prince, J. P., de Vincente, M. C., Bonierbale, M. W., Broun, P., Fulton, T. M., Giovannoni, J. J., Grandillo, S., Martin, G. B., Messeguer, R., Miller, J. C., et al. (1992). High density molecular linkage maps of the tomato and potato genomes. Genetics 132, 1141–1160. Tanksley, S. D., Ganal, M. W., and Martin, G. B. (1995). Chromosome landing: A paradigm for map based gene cloning in plants with large genomes. Trends Genet. 11, 63–68. Taramino, G., Tarchini, R., Ferrario, S., Lee, M., and Pe, M. E. (1997). Characterization and mapping of simple sequence repeats (SSRs) in Sorghum bicolor. Theor. Appl. Genet. 95, 66–72. Tar’an, B., Michaels, T. E., and Pauls, K. P. (1998). Stability of association of molecular markers with common bacterial blight resistance in common bean (Phaseolus vulgaris L.). Plant Breed. 117, 553–558. Tar’an, B., Michaels, T. E., and Pauls, K. P. (2003). Marker‐assisted selection for complex trait in common bean (Phaseolus vulgaris L.) using QTL‐based index. Euphytica 130, 423–432. Tardieu, F. (2003). Virtual plants: Modeling as a tool for the genomics of tolerance to water deficit. Trends Plant Sci. 8, 9–14. Tekeoglu, M., Rajesh, P. N., and Muehlbauer, F. J. (2002). Integration of sequence tagged microsatellites to the chickpea genetic map. Theor. Appl. Genet. 105, 847–854. Thiel, T., Michalek, W., Varshney, R. K., and Graner, A. (2003). Exploiting EST data bases for the development and characterization of gene‐derived SSR‐markers in barley (Hordeum vulgare L.). Theor. Appl. Genet. 106, 411–422. Thomas, W. T. B. (2003). Prospects for molecular breeding of barley. Ann. Appl. Genet. 142, 1–12.
APPLIED CROP GENOMICS
311
Thornsberry, J. M., Goodman, M. M., Doebley, J., Kresovich, S., Nielsen, D., and Buckler, E. S. (2001). Dwarf8 polymorphisms associate with variation in flowering time. Nat. Genet. 28, 286–289. Tian, F., Zhu, Z., Zhang, B., Tan, L., Fu, Y., Wang, X., and Sun, C. Q. (2006a). Fine mapping of a quantitative trait locus for grain number per panicle from wild rice (Oryza rufipogon GriV.). Theor. Appl. Genet. 113, 619–629. Tian, F., Li, D. J., Fu, Q., Zhu, Z. F., Fu, Y. C., Wang, X. K., and Sun, C. Q. (2006b). Construction of introgression lines carrying wild rice (Oryza rufipogon GriV.) segments in cultivated rice (Oryza sativa L.) background and characterization of introgressed segments associated with yield‐related traits. Theor. Appl. Genet. 112, 570–580. Till, B. J., Reynolds, S. H., Grean, E. A., Codomo, C. A., Enns, L. C., Johnson, J. E., Burtner, C., Odden, A. R., Young, K., Taylor, N. E., HenikoV, J. G., Comai, L., et al. (2003). Large‐scale discovery of induced point mutations with high throughput TILLING. Genome Res. 13, 524–530. Till, B. J., Reynolds, S. H., Weil, C., Springer, N., Burtner, C., Young, K., Bowers, E., Codomo, C. A., Enns, L. C., Odden, A. R., Greene, E. A., Comai, L., et al. (2004). Discovery of induced point mutations in maize by TILLING. BMC Plant Biol. 4, 12. Timmerman‐Vaughan, G. M., Frew, T. J., Butler, R., Murray, S., Gilpin, M., Falloon, K., Johnston, P., Lakeman, M. B., Russell, A., and Khan, T. (2004). Validation of quantitative trait loci for Ascochyta blight resistance in pea (Pisum sativum L.), using populations from two crosses. Theor. Appl. Genet. 109, 1620–1631. Tinker, N. A., and Mather, D. E. (1993). GREGOR: Software for genetic simulation. J. Hered. 84, 237. Toojinda, T., Baird, E., Booth, A., Broers, L., Hayes, P., Powell, W., Thomas, W., Vivar, H., and Young, G. (1998). Introgression of quantitative trait loci (QTL) determining stripe rust resistance in barley: An example of marker‐assisted line development. Theor. Appl. Genet. 96, 123–131. Torada, A., Ikeguchi, S., and Koike, M. (2005). Mapping and validation of PCR‐based markers associated with a major QTL for seed dormancy in wheat. Euphytica 143, 251–255. Torada, A., Koike, M., Mochida, K., and Ogihara, Y. (2006). SSR‐based linkage map with new markers using an intraspecific population of common wheat. Theor. Appl. Genet. 112, 1042–1051. Toth, B., Francia, E., Rizza, F., Stanca, A. M., Galiba, G., and Pecchioni, N. (2004). Development of PCR‐based markers on chromosome 5H for assisted selection of frost‐tolerant genotypes in barley. Mol. Breed. 14, 265–273. Tranquilli, G. E., and Dubcovsky, J. (1999). Epistatic interaction between vernalization genes Vrn‐Am1 and Vrn‐Am2 in diploid wheat. J. Hered. 91, 304–306. Ueda, A., Shi, W., Nakamura, T., and Takabe, T. (2002). Analysis of salt‐inducible genes in barley roots by diVerential display. J. Plant Res. 115, 119–130. Ueda, A., Kathiresan, A., Inada, M., Narita, Y., Nakamura, T., Shi, W., Takabe, T., and Bennett, J. (2004). Osmotic stress in barley regulates expression of a diVerent set of genes than salt stress does. J. Exp. Bot. 55, 2213–2218. Upadhyaya, H. D., and Ortiz, R. (2001). A mini core subset for capturing diversity and promoting utilization of chickpea genetic resources in crop improvement. Theor. Appl. Genet. 102, 1292–1298. Upadhyaya, H. D., Bramel, P. J., Ortiz, R., and Singh, S. (2002). Developing a mini core of peanut for utilization of genetic resources. Crop Sci. 42, 2150–2156. Upadhyaya, H. D., Furman, B. J., Dwivedi, S. L., Udupa, S. M., Gowda, C. L. L., Baum, M., Crouch, J. H., Buhariwalla, H. K., and Singh, S. (2006a). Development of a composite collection for mining germplasm possessing allelic variation for beneficial traits in chickpea. Plant Genet. Resour. 4, 13–19.
312
S. L. DWIVEDI ET AL.
Upadhyaya, H. D., Gowda, C. L. L., Pundir, R. P. S., Reddy, V. G., and Singh, S. (2006b). Development of core subset of fingermillet germplasm using geographical origin and data on 14 quantitative traits. Genet. Resour. Crop Evol. 53, 679–685. Upadhyaya, H. D., Reddy, L. J., Gowda, C. L. L., Reddy, K. N., and Singh, S. (2006c). Development of a mini core subset for enhanced and diversified utilization of pigeonpea germplasm resources. Crop Sci. 46, 2127–2132. Utz, H. F., and Melchinger, A. E. (1996). PLABQTL: A program for composite interval mapping of QTL. J. Quant. Trait Loci 2, 1–5. Vales, M. I., Schon, C. C., Capettini, F., Chen, X. M., Corey, A. E., Mather, D. E., Mundt, C. C., Richardson, K. L., Sandoval‐Islas, J. S., Utz, H. F., and Hayes, P. M. (2005). EVect of population size on the estimation of QTL: A test using resistance to barley stripe rust. Theor. Appl. Genet. 111, 1260–1270. van Berloo, R. (1999). GGT: Software for the display of graphical genotypes. J. Hered. 90, 328–329. van Berloo, R., Aalbers, H., Werkman, A., and Niks, R. E. (2001). Resistance QTL confirmed through development of QTL‐NILs for barley leaf rust resistance. Mol. Breed. 8, 187–195. Van Ooijen, A. J., and Voorrips, R. E. (2001). ‘‘JoinMap (tm) 3.0: Software for the Calculation of Genetic Linkage Maps,’’ Plant Research International, Wageningen, the Netherlands. van Os, H., Andrzejewski, S., Bakker, E., Barrena, I., Bryan, G. J., Caromel, B., Ghareeb, B., Ishidore, E., de Jong, W., van Koert, P., Lefebvre, V., Milbourne, D., et al. (2006). Construction of a 10,000‐marker ultradense genetic recombination map of potato: Providing a framework for accelerated gene isolation and a genomewide physical map. Genetics 173, 1075–1087. van Soest, L. J. M., Schober, B., and Tazelaar, M. F. (1984). Resistance to Phytophthora infestans in tuber bearing species of Solanum and its geographic distribution. Potato Res. 27, 393–411. Vargas, M., van Eeuwijk, F. A., Crossa, J., and Ribaut, J.‐M. (2006). Mapping QTLs and QTL environment interaction for CIMMYT maize drought stress program using factorial regression and partial least squares methods. Theor. Appl. Genet. 112, 1009–1023. Varshney, R. K., Graner, A., and Sorrells, M. E. (2005a). Genomics‐assisted breeding for crop improvement. Trends Plant Sci. 10, 621–630. Varshney, R. K., Graner, A., and Sorrells, M. E. (2005b). Genic micorsatellite markers in plants: Features and applications. Trends Biotechnol. 23, 48–55. Venter, J. C., Adams, M. D., Myers, E. W., Li, P. W., Mural, R. J., Sutton, G. G., Smith, H. O., Yandell, M., Evans, C. A., Holt, R. A., Gocayne, J. D., Amanatides, P., et al. (2001). The sequence of the human genome. Science 291, 1304–1351. Visscher, P. M., Thompson, R., and Haley, C. S. (1996). Confidence intervals in QTL mapping by bootstrapping. Genetics 143, 1013–1020. Vision, T. J., Brown, D. G., Shmoys, D. B., Durrett, R. T., and Tanksley, S. D. (2000). Selective mapping: A strategy for optimizing the construction of high‐density linkage maps. Genetics 155, 407–420. Vladutu, C., McLaughlin, J., and Phillips, R. L. (1999). Fine mapping and characterization of linked quantitative trait loci involved in the transition of the maize apical meristem from vegetative to generative structures. Genetics 153, 993–1007. Vodkin, L. O., Khanna, A., Shealy, R., Clough, S. J., Gonzalez, D. O., Philip, R., Zabala, G., Thibaud‐Nissen, F., Sidarous, M., Stro¨mvik, M. V., Shoop, E., et al. (2004). Microarrays for global expression constructed with a low redundancy set of 27,500 sequenced cDNAs representing an array of developmental stages and physiological conditions of the soybean plant. BMC Genomics 4, 73.
APPLIED CROP GENOMICS
313
von KorV, M., Wang, H., Le´on, J., and Pillen, K. (2004). Development of candidate introgression lines using an exotic barley accession (Hordeum vulgare ssp. spontaneum) as donor. Theor. Appl. Genet. 109, 1736–1745. Waldron, B. L., Moreno, S. B., Anderson, J. A., Stack, R. W., and Frohberg, R. C. (1999). RFLP mapping of QTL for Fusarium head blight resistance in wheat. Crop Sci. 39, 805–811. Walia, H., Wilson, C., Wahid, A., Condamine, P., Cui, X., and Close, T. J. (2006). Expression analysis of barley (Hordeum vulgare L.) during salinity stress. Funct. Integr. Genomics 6, 143–156. Walker, D., Boerma, H. R., All, J., and Parrott, W. (2002). Combining cry1Ac with QTL alleles from PI 229358 to improve soybean resistance to lepidopteran pests. Mol. Breed. 9, 43–51. Walker, D. R., Narvel, J. M., Boerma, H. R., All, J. N., and Parrott, W. A. (2004). A QTL that enhances and broadens Bt insect resistance in soybean. Theor. Appl. Genet. 109, 1051–1957. Wang, D., Arelli, P. R., Shoemaker, R. C., and Diers, B. W. (2001b). Loci underlying resistance to race 3 of soybean cyst nematode in Glycine soja plant introduction 468916. Theor. Appl. Genet. 103, 561–566. Wang, E., Robertson, M. J., Hammer, G. L., Carberry, P. S., Holzworth, D., Meinke, H., Chapman, S. C., Hargreaves, J. N. G., Huth, N. I., and McLean, G. (2002). Development of a generic crop model template in the cropping system model APSIM. Eur. J. Agron. 18, 121–140. Wang, H., Nussbaum‐Wagler, T., Li, B., Zhao, Q., Vigouroux, Y., Faller, M., Bomblies, K., Lukens, L., and Doebley, J. F. (2005b). The origin of naked grains of maize. Nature 436, 714–719. Wang, J., van Ginkel, M., Trethowan, R., Ye, G., DeLacy, I., Podlich, D., and Cooper, M. (2004). Simulating the eVects of dominance and epistasis on selection response in the CIMMYT wheat breeding program using QuCim. Crop Sci. 44, 2006–2018. Wang, J., Eagles, H. A., Trethowan, R., and van Ginkel, M. (2005c). Using computer simulation of the selection process and known gene information to assist in parental selection in wheat quality breeding. Aust. J. Agric. Res. 56, 465–473. Wang, R. L., Stec, A., Hey, J., Lukens, L., and Doebley, J. (1999). The limits of selection during maize domestication. Nature 398, 236–239. Wang, X. Y., Chen, P. D., and Zhang, S. Z. (2001a). Pyramiding and marker‐assisted selection for powdery mildew resistance genes in common wheat. Acta Genetica Sinica 28, 640–646 (in Chinese; summary in English). Wang, Y. H., Liu, S. J., Ji, S. L., Zhang, W. W., Wang, C. M., Jiang, L., and Wan, J. M. (2005a). Fine mapping and marker‐assisted selection (MAS) of a low glutelin content gene in rice. Cell Res. 15, 622–630. Warkentin, T., Tar’an, B., Zhang, C., Bett, K., Tullu, A., Somers, D., and Vandenberg, A. (2004). Marker‐assisted selection for lodging resistance in pea. In ‘‘Fifth European Conference on Grain Legumes and Second International Conference on Legume Genomics and Genetics,’’ pp. 149–150. Dijon, France. Wasaki, J., Shinano, T., Onishi, K., Yonetani, R., Yazaki, J., Fujii, F., Shimbo, K., Ishikawa, M., Shimatani, Z., Nagata, Y., Hashimoto, A., Ohta, T., et al. (2006). Transcriptomic analysis indicates putative metabolic changes caused by manipulation of phosphorus availability in rice leaves. J. Exp. Bot. 57, 2049–2059. Wayne, M. L., and Mclntyre, L. M. (2002). Combining mapping and arraying: An approach to candidate gene identification. Proc. Natl. Acad. Sci. USA 99, 14903–14906. Webb, B. D. (1980). Rice quality and grades. In ‘‘Rice: Production and Utilization’’ (B. S. Luh, Ed.), pp. 543–565. Avi Publishing Co., Inc., Westport, CT. Wenzl, P., Carling, J., Kudrna, D., Jaccoud, D., Huttner, E., Kleinhofs, A., and Kilian, A. (2004). Diversity array technology (DArT) for whole‐genome profiling of barley. Proc. Natl. Acad. Sci. USA 101, 9915–9920. Werner, K., Friedt, W., and Ordon, F. (2005). Strategies for pyramiding resistance genes against the barley yellow mosaic virus complex (BaMMV, BaYMV, BaYMV‐2). Mol. Breed. 16, 45–55.
314
S. L. DWIVEDI ET AL.
Whitelaw, C. A., Barbazuk, W. B., Pertea, G., Chan, A. P., Cheung, F., Lee, Y., Zheng, L., van Heeringen, S., Karamycheva, S., Bennetzen, J. L., SanMiguel, P., Lakey, N., et al. (2003). Enrichment of gene‐coding sequences in maize by genome filtration. Science 302, 2118–2120. Wienholds, E., van Eden, F., Kosters, M., Mudde, J., Plasterk, R. H., and Cuppen, E. (2003). EYcient target‐selected mutagenesis in Zebrafish. Genome Res. 13, 2700–2707. Willcox, M. C., Khairallah, M. M., Bergvinson, D., Crossa, J., Deutsch, J. A., Edmeades, G. O., Gonzalez‐de‐Leon, D., Jiang, C., Jewell, D. C., Mihm, J. A., Williams, W. P., and Hoisington, D. (2002). Selection for resistance to Southwestern corn borer using marker‐ assisted and conventional backcrossing. Crop Sci. 42, 1516–1528. Williams, K., Bogacki, P., Scott, L., Karakousis, A., and Wallwork, H. (2001). Mapping of a gene for leaf scald resistance in barley line ‘B87/14’ and validation of microsatellite and RFLP markers for marker‐assisted selection. Plant Breed. 120, 301–304. Wilson, I. D., Barker, G. L. A., Beswick, R. W., Shepherd, S. K., Lu, C., Coghill, J. A., Edwards, D., Owen, P., Lyons, R., Parker, J. S., Lenton, J. R., Holdsworth, M. J., et al. (2004). A transcriptomics resource for wheat functional genomics. Plant Biotechnol. J. 2, 495–506. Winzeler, W., Winzeler, H., and Keller, B. (1995). Endopeptidase polymorphism and linkage of the Ep‐D1c null allele with the Lr19 leaf rust resistance gene in hexaploid wheat. Plant Breed. 114, 24–28. Wittaker, J. C., Curnow, R. N., Haley, C. S., and Thompson, R. (1995). Using marker‐maps in marker‐assisted selection. Genet. Res. 66, 255–265. Wu, R., and Zeng, Z.‐B (2001). Joint linkage and linkage disequilibrium mapping in natural populations. Genetics 157, 899–909. Wu, J., Maehara, T., Shimokawa, T., Yamamoto, S., Harada, C., Takazaki, Y., Ono, N., Mukai, Y., Koike, K., Yazaki, J., Fujii, F., Shomura, A., et al. (2002b). A comprehensive rice transcript map containing 6591 expressed sequence tag sites. Plant Cell 14, 525–535. Wu, L. M., Ni, Z. F., Meng, F. R., Lin, Z., and Sun, Q. X. (2003). Cloning and characterization of leaf cDNAs that are diVerentially expressed between wheat hybrids and their parents. Mol. Gen. Genet. 270, 281–286. Wu, R., Ma, C.‐S., and Casella, G. (2002a). Joint linkage and linkage disequilibrium mapping of qualtitative trait loci in natural mapping populations. Genetics 160, 779–792. Xia, L., Peng, K., Yang, S., Wenzl, P., de Vincente, M. C., Fregene, M., and Kilian, A. (2005). DarT for high‐throughput genotyping of cassava (Manihot esculenta) and its wild relatives. Theor. Appl. Genet. 110, 1092–1098. Xiao, J., Li, J., Yuan, L., and Tanksley, S. D. (1995). Dominance is the major genetic basis of heterosis in rice as revealed by QTL analysis using molecular markers. Genetics 140, 745–754. Xiao, Y. H., Liu, L. L., Jiang, L., Lu, C. G., Yu, C. Y., Zhang, W. W., Zuo, J. S., Zhai, H. Q., and Wan, J. M. (2005). Development of chromosome segment substitution lines (CSSLs) derived from a hybrid rice cross, Pai64S/9311 with super high yield potential. Rice Genet. Newsl. 22, 17–19. Xie, C., and Xu, S. (1998). EYciency of multistage marker‐assisted selection in the improvement of multiple quantitative traits. Heredity 80, 489–498. Xie, X., Song, M.‐H., Jin, F., Ahn, S.‐N., Suh, J.‐P., Hwang, H.‐G., and McCouch, S. R. (2006). Fine mapping of a grain weight quantitative trait locus on rice chromosome 8 using near‐ isogenic lines derived from a cross between Oryza sativa and Oryza rufipogon. Theor. Appl. Genet. 113, 885–894. Xu, Y. (2003). Developing marker‐assisted selection strategies for breeding hybrid rice. Plant Breed. Rev. 23, 73–174. Xu, K., and Mackill, D. J. (1996). A major locus for submergence tolerance mapped on rice chromosome 9. Mol. Breed. 2, 219–224.
APPLIED CROP GENOMICS
315
Xu, K., Xu, X., Ronald, P. C., and Mackill, D. J. (2000). A high resolution genetic linkage map in the vicinity of the rice submergence tolerance locus Sub1. Mol. Gen. Genet. 263, 681–689. Xu, K., Deb, R., and Mackill, D. J. (2004b). A microsatellite marker and a codominant PCR‐ based marker for marker‐assisted selection of submergence tolerance in rice. Crop Sci. 44, 248–253. Xu, K., Xia, X., Fukao, T., Canlas, P., Maghirang‐Rodriguez, R., Heuer, S., Ismail, A. I., Bailey‐Serres, J., Ronald, P. C., and Mackill, D. J. (2006). Sub1A is an ethylene response factor‐like gene that confers submergence tolerance to rice. Nature 442, 705–708. Xu, Y., Ishii, T., and McCouch, S. R. (2003). Marker‐assisted evaluation of germplasm resources for plant breeding. In ‘‘Rice Science: Innovations and Impact for Livelihood’’ (T. W. Mew, D. S. Brar, S. Peng, and B. Hardy, Eds.), pp. 213–229. Proceedings of the 24th International Rice Res. Conference, 16–19 September 2002, Beijing, China. IRRI, CAE, and CAAS, Manila, Philippines. Xu, Y., Beachell, H., and McCouch, S. R. (2004a). A marker‐based approach to broadening the genetic base of rice (Oryza sativa L.) in the U.S. Crop Sci. 44, 1947–1959. Yadav, R. S., Hash, C. T., Bidinger, F. R., Cavan, G. P., and Howarth, C. J. (2002). Quantitative trait loci associated with traits determining grain and stover yield in pearlmillet under terminal drought stress conditions. Theor. Appl. Genet. 104, 67–83. Yamamoto, T., Kuboki, Y., Lin, S. Y., Sasaki, T., and Yano, M. (1998). Fine mapping of quantitative trait loci Hd‐1, Hd‐2 and Hd‐3, controlling heading date of rice, as single Mendelian factors. Theor. Appl. Genet. 97, 37–44. Yan, L., Loukoianov, A., Tranquilli, G., Helguera, M., Fahima, T., and Dubcovsky, J. (2003). Positional cloning of the wheat vernalization gene VRN1. Proc. Natl. Acad. Sci. USA 100, 6263–6268. Yan, L., Loukoianov, A., Blechl, A., Tranquilli, G., Ramakrishna, W., SanMiguel, P., Bennetzen, J. L., Echenique, V., and Dubcovsky, J. (2004a). The wheat VRN2 gene is a flowering repressor down‐regulated by vernalization. Science 303, 1640–1644. Yan, W., Rutger, J. N., Bockelman, H. E., and Tai, T. (2004b). Development of a core collection from the USDA rice germplasm collection. In ‘‘B. R. Wells Rice Research Studies 2003’’ (R. J. Norman, R. J. Meullenet, and J.‐F. Moldenhauer, Eds.), pp. 88–96. Arkansas Agric. Expt. Res. St. Series 517 ( http://www.uark.edu/depts/agripub/publications/research). Yang, L., Zheng, B., Mao, C., Qi, X., Liu, F., and Wu, P. (2004). Analysis of transcript that are diVerentially expressed in three sectors of the rice root system under water deficit. Mol. Gen. Genom. 272, 433–442. Yang, X., Rupe, M., Bickel, D., Arthur, L., Smith, O., and Guo, M. (2006). EVects of cis‐trans‐ regulation on allele‐specific transcript expression in the meristems of maize hybrids. In ‘‘48th Annual Maize Genetic Conference,’’ 9–12 March 2006. California. p. 118. Yang, Z. P., Gilbert, J., Somers, D. J., Fedak, G., Procunier, J. D., and McKenzie, I. H. (2003). Marker‐assisted selection of fusarium head blight resistance genes in two double haploid populations of wheat. Mol. Breed. 12, 309–317. Yano, M., Katayose, Y., Ashikari, M., Yamanouchi, U., Monna, L., Fuse, T., Baba, T., Yamamoto, K., Umehara, Y., Nagamura, Y., and Sasaki, T. (2000). Hd1, a major photoperiod sensitivity quantitative trait locus in rice, is closely related to the Arabidopsis flowering time gene CONSTANS. Plant Cell 12, 2473–2483. Yao, Y., Ni, Z., Zhang, Y., Chen, Y., Ding, Y., Han, Z., Liu, Z., and Sun, Q. (2005). Identification of diVerentially expressed genes in leaf and root between wheat hybrid and its parental inbreds using PCR‐based cDNA subtraction. Plant Mol. Biol. 58, 367–384. Yim, Y.‐S., Davis, G. L., Duru, N. A., Musket, T. A., Linton, E. W., Messing, J. W., McMullen, M. D., Soderlund, C. A., Polacco, M. L., Gardiner, J. M., and Coe, E. H., Jr. (2002). Characterization of three maize bacterial artificial chromosome libraries toward
316
S. L. DWIVEDI ET AL.
anchoring of the physical map to the genetic map using high‐density bacterial artificial chromosome filter hybridization. Plant Physiol. 130, 1686–1696. Yin, X., KropV, M. J., and Stam, P. (1999). The role of ecophysiological models in QTL analysis: The example of specific leaf area in barley. Heredity 82, 415–421. Yin, X., Stam, P., KropV, M. J., and Schapendonk, A. H. C. M. (2003). Crop modeling, QTL mapping, and their complementary role in plant breeding. Agron. J. 95, 90–98. Yin, X., Struik, P. C., and KropV, M. J. (2004). Role of crop physiology in predicting gene‐to‐ phenotype relationships. Trends Plant Sci. 9, 426–432. Yin, X., Struik, P. C., Tang, J., Qi, C., and Liu, T. (2005). Model analysis of flowering phenology in recombinant inbred lines of barley. J. Exp. Bot. 56, 959–965. Yoon, D.‐B., Kang, K.‐H., Kim, H.‐J., Ju, H.‐G., Kwon, S.‐J., Suh, J.‐P., Jeong, O.‐Y., and Ahu, S.‐N. (2006). Mapping quantitative trait loci for yield components and morphological traits in an advanced backcross population between Oryza grandiglumis and the O. japonica cultivar Hwaseongbyeo. Theor. Appl. Genet. 112, 1052–1062. Yoshida, S., and Hasegawa, S. (1982). The rice root system: Its development and function. In ‘‘Drought Resistance in Crops with Emphasis on Rice,’’ pp. 97–114. International Rice Research Institute, Los Ban˜os, Laguna, Philippines. Yoshihashi, T. (2002). Quantitative analysis on 2‐acetyl‐1‐pyrroline of an aromatic rice by stable isotope dilution method and on model studies on its formation during cooking. J. Food Sci. 67, 619–622. Yoshimura, S., Yoshimura, A., Iwata, N., McCouch, S. R., Abenes, M. L., Baraoidan, M. R., Mew, T. W., and Nelson, R. J. (1995). Tagging and combining bacterial blight resistance genes in rice using RAPD and RFLP markers. Mol. Breed. 1, 375–387. Young, N. D. (1999). A cautiously optimistic vision for marker assisted breeding. Mol. Breed. 5, 505–510. Yousef, G. G., and Juvik, J. A. (2001a). Comparison of phenotypic and marker‐assisted selection for quantitative traits in sweet corn. Crop Sci. 41, 645–655. Yousef, G. G., and Juvik, J. A. (2001b). Evaluation of breeding utility of a chromosomal segment from Lycopersicon chmielewskii that enhances cultivated tomato soluble solids. Theor. Appl. Genet. 103, 1022–1027. Yousef, G. G., and Juvik, J. A. (2002). Enhancement of seedling resistance in sweet corn by marker‐assisted backcrossing of beneficial QTL. Crop Sci. 42, 96–104. Yu, J., Hu, S., Wang, J., Wong, G. K.‐S., Li, S., Liu, B., Deng, Y., Dai, L., Zhou, Y., Zhang, X., Cao, M., Liu, J., et al. (2002). A draft sequence of the rice genome (Oryza sativa L. ssp. indica). Science 296, 79–92. Yu, J., Wang, J., Lin, W., Li, S., Li, H., Zhou, J., Ni, P., Dong, W., Hu, S., Zeng, C., Zhang, J., Zhang, Y., et al. (2005). The genome of Oryza sativa: A history of duplications. PLoS Biol. 3, 0266–0281. Yu, K., Park, S. J., and Poysa, V. (2000). Marker‐assisted selection of common beans for resistance to common bacterial blight: EYciency and economics. Plant Breed. 119, 411–415. Yu, L.‐X., and Setter, T. L. (2003). Comparative transcriptional profiling of placenta and endosperm in developing maize kernels in response to water deficit. Plant Physiol. 131, 568–582. Yu, S. B., Li, J. X., Tan, Y. F., Gao, Y. J., Li, X. H., Zhang, Q. F., and Maroof, M. A. S. (1997). Importance of epistasis as the genetic basis of heterosis in an elite rice hybrid. Proc. Natl. Acad. Sci. USA 94, 9226–9231. Yuan, Y., SanMiguel, P. J., and Bennetzen, J. L. (2003). High Cot sequence analysis of the maize genome. Plant J. 34, 249–255 (Erratum: Plant J. 36, 430). Yue‐guang, W., Qi‐yun, D., Feng‐shan, L., Quan‐hua, X., Ji‐ming, L., Yue‐dong, X., Shi‐mong, S., Bao‐tai, G., Long‐ping, Y., and Bin, W. (2004). Molecular marker‐assisted selection for yield‐enhancing genes in the progeny of Minghui63 x Oryza rufipogon.
APPLIED CROP GENOMICS
317
Agric. Sci. (in Chinese) 3, 89–93 (http://www.wanfangdata.com.cn/qikan/periodical.articles/ zgnykx‐e/zgny2004/0402/04020). Yun, S. J., Geynis, L., Bossolini, E., Hayes, P. M., Matus, I., Smith, K. P., SteVenson, B. J., Tuberosa, R., and Muehlbauer, G. J. (2006). Validation of quantitative trait loci for multiple disease resistance in barley using advanced backcross lines developed with a wild barley. Crop Sci. 46, 1179–1186. Yuqing, H., Li, X., Zhang, J., Jiang, G., Liu, S., Chen, S., Tu, J., Xu, C., and Zhang, Q. (2004). Gene pyramiding to improve hybrid rice by molecular marker technique. In ‘‘Proceedings of the Fourth International Crop Science Congress’’ (http://www.regional.org.au/au/cs/2004/ poster/3/4/4/1041‐heyuqing.htm). Zale, J. M., Clancy, J. A., Ullrich, S. E., Jones, B. L., Hays, P. M. and the North American Barley Genome Mapping Project. (2000). Summary of barley malting QTL mapped in various mapping populations. Barley Genet. Newsl. 30, 1–4. Zhang, C., Bunyamin, T., Warkentin, T., Tullu, A., Bett, K. E., Vandenberg, B., and Somers, D. J. (2006a). Selection for lodging resistance in early generations of field pea by molecular markers. Crop Sci. 46, 321–329. Zhang, H., Sreenivasulu, N., Weschke, W., Stein, N., Rudd, S., Radchuk, V., Potokina, E., Scholz, U., Schweizer, P., Zeirold, U., Langridge, P., Varshney, R. K., et al. (2004). Large‐ scale analysis of the barley transcriptome based on expressed sequence tags. Plant J. 40, 276–290. Zhang, J., Zheng, H. G., Aarti, A., Pantuwan, G., Nguyen, T. T., Tripathi, J. N., Sarial, A. K., Robin, S., Babu, R. C., Nguyen, B. D., Sarkarung, S., Blum, A., et al. (2001). Locating genomic regions associated with components of drought resistance in rice: Comparative mapping within and across species. Theor. Appl. Genet. 103, 19–29. Zhang, Q., Gao, Y. J., Yang, S. H., Ragab, R. A., Saghai Maroof, M. A., and Li, Z. B. (1994). A diallel analysis of heterosis in elite hybrid rice based on RFLPs and microsatellites. Theor. Appl. Genet. 89, 185–192. Zhang, W., Collins, A., Maniatis, N., Tapper, W., and Morton, N. E. (2002). Properties of linkage disequilibrium LD maps. Proc. Natl. Acad. Sci. USA 99, 17004–17007. Zhang, Z., Bradbury, P. J., Kroon, D. E., Casstevens, T. M., and Buckler, E. S. (2006b). TASSEL 2.0: A software package for association and diversity analyses in plants and animals, Plant & Animal Genomes XIV Conference. Zhao, J. H. (2004). 2LD, GENECOUNTING and HAP: Computer programs for linkage disequilibrium analysis. Bioinformatics 20, 1325–1326. Zhao, H., Zhang, S., Merikangas, K. R., Trixler, M., Wildenauer, D. B., Sun, F., and Kidd, K. K. (2000). Transmission/disequilibrium tests using multiple tightly linked markers. Am. J. Hum. Genet. 67, 936–946. Zhao, Q., Zhang, Y., Cheng, Z., Chen, M., Wang, S., Feng, Q., Huang, Y., Li, Y., Tang, Y., Zhou, B., Chen, Z., Yu, S., et al. (2002). A fine physical map of the rice chromosome 4. Genome Res. 12, 817–823. Zhao, X. C., Batey, I. L., Sharp, P. J., Crosbie, G., Barclay, I., Wilson, R., Morell, M. K., and Appels, R. (1998). A single genetic locus associated with starch granule properties and noodle quality in wheat. J. Cereal Sci. 27, 7–13. Zhou, P. H., Tan, Y. F., He, Y. A., Xu, C. G., and Zhang, A. (2003a). Simultaneous improvement of four quality traits of Zhenshan 97, an elite parent of hybrid rice, by molecular marker‐assisted selection. Theor. Appl. Genet. 106, 326–331. Zhou, W.‐C., Kolb, F. L., Bai, G.‐H., Domier, L. L., Boze, L. K., and Smith, N. J. (2003b). Validation of a major QTL for scab resistance with SSR markers and use of marker‐assisted selection in wheat. Plant Breed. 122, 40–46. Zhu, S., and Kaeppler, H. F. (2003). A genetic linkage map for hexaploid, cultivated oat (Avena sativa L.) based on an intraspecific cross Ogle/MAM17–5. Theor. Appl. Genet. 107, 26–35.
318
S. L. DWIVEDI ET AL.
Zhu, T., Budworth, P., Chen, W., Provart, N., Chang, H.‐S., Guimil, S., Su, W., Estes, B., Zou, G., and Wang, X. (2003). Transcriptional control of nutrient partitioning during rice grain filling. Plant Biotechnol. J. 1, 59–70. Zinselmeier, C., Sun, Y., Helentjaris, T., Beatty, M., Yang, S., Smith, H., and Habben, J. (2002). The use of gene expression profiling to dissect the stress sensitivity of reproductive development in maize. Field Crop Res. 73, 111–121.
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE D. D. Stuthman,1 K. J. Leonard2 and J. Miller‐Garvin1 1
Department of Agronomy and Plant Genetics, University of Minnesota, St. Paul, Minnesota 55108 2 Cereal Disease Laboratory, USDA‐ARS, St. Paul, Minnesota 55108
I. Introduction A. Importance of Reliable Disease Resistance in Major Crops B. Causes of Plant Disease Epidemics (The Disease Triangle) C. Examples of Plant Disease in Natural Populations II. Concepts of Resistance to Plant Disease Useful in Breeding EVorts A. A Pragmatic Approach B. Classification of Types of Resistance from a Breeder’s Perspective III. What Makes Disease Resistance Durable? A. Resistance in Wild Plant Species B. Impact of Agriculture on Resistance C. EVorts to Delay Breakdown of Inherently Transient Resistance D. Examples of Durable Monogenic Resistance E. Durability of Polygenic Resistance IV. Examples of EVective Polygenic Resistance A. Maize B. Wheat C. Barley D. Potato V. Classical Breeding Approaches A. Recurrent Selection B. Pedigree Breeding C. Perennial Species VI. Molecular Approaches A. Marker‐Assisted Selection B. Genetic Transformation VII. Summary References
Durable resistance to disease is a common component of plant defense systems in natural ecosystems and can be found in virtually all cultivated species. Agricultural production practices increase crop vulnerability to most diseases, so higher levels of resistance than occur in natural ecosystems may be needed. Monogenic resistance to highly specialized pathogens is often 319 Advances in Agronomy, Volume 95 Copyright 2007, Elsevier Inc. All rights reserved. 0065-2113/07 $35.00 DOI: 10.1016/S0065-2113(07)95004-X
D. D. STUTHMAN ET AL.
320
highly eVective when first developed but is generally race‐specific and nondurable, especially when characterized by hypersensitive host reactions. Partial resistance conditioned by several to many genes with additive eVects is usually durable, particularly when it involves morphological or developmental changes in the plant. Exceptions to the general rule of nondurable monogenic resistance and durable polygenic resistance are presented. Recurrent selection is commonly used to develop cultivars with durable resistance, but durable resistance has also been achieved through pedigree breeding in small grains. Success in accumulating polygenic durable resistance to multiple diseases of maize and to leaf rust and stripe rust of wheat are described as the mixed record of success in breeding for late blight resistance in potato. The mlo gene for resistance to powdery mildew in barley oVers an intriguing case of highly eVective monogenic resistance that appears to be durable. Modern molecular genetic approaches oVer promise not only for marker‐ assisted selection of partial resistance genes but also for creation of novel forms of resistance to plant diseases. Nevertheless, traditional breeding and field tests will still have an essential role in developing commercial cultivars. # 2007, Elsevier Inc.
I. INTRODUCTION A. IMPORTANCE OF RELIABLE DISEASE RESISTANCE IN MAJOR CROPS The continuing increase of the human population throughout the world demands corresponding increases in food accessibility. This is especially true for the cultivated crops that nearly all humans rely on for the bulk of their nutrition. With most arable land already under cultivation, increased food accessibility must come from increased yield per hectare, reduced losses of harvested crops to pests and spoilage during storage or transit, and better distribution. The enormous gains in wheat and rice yields characterized as the Green Revolution during the last half of the twentieth century staved oV what had once seemed to be an inevitable round of mass starvation in the poorest countries of the world. Today we are faced with the daunting task of repeating the accomplishments of the Green Revolution. World population will likely have increased 30–50% by 2025 (Gould and Cohen, 2000), which means that accessibility of the main food crops of the world must also be improved by 30–50% in the coming decades. Breeding for disease resistance in crops is an essential component in the campaign to maintain adequate food supplies for the world’s growing population. All crops, indeed all plant species, are subject to disease as a natural part of life’s evolution on the earth. Just as the remains of dying plants serve as food for saprophytes that recycle nutrients accumulated during the plants’ growth, living plants are also a rich source of food for those parasites that
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
321
have evolved the means to attack them (Parlevliet, 1995). Thus, the eVorts by plant breeders to increase the productivity of crops for human consumption also increase the potential food supply for plant pathogens. Unless they can be successfully managed, the yield losses to disease may very well erase most gains in crop productivity achieved by plant breeders. Considering this, Shaner (1981) declared that no field of research and development in agriculture has a more favorable input‐to‐return ratio than breeding for resistance when all socioeconomic eVects are taken into account. Despite the best eVorts of plant breeders to produce resistant cultivars and of plant pathologists to develop and implement other forms of disease control, plant diseases cause at least 10% losses in global food production annually (Strange and Scott, 2005). Estimates for worldwide disease losses in six major crops for the period 2001–2003 ranged from 7.2% for cotton to 14.5% for potato (Oerke, 2006; Table I). Losses to diseases caused by fungi tend to be greater in the humid tropics where crops may be grown continuously throughout the year and weather conditions favor infection than in temperate regions of North America and Northwest Europe. Losses to fungal diseases also tend to increase with increased intensity of crop production. Oerke (2006) estimated that current crop production practices reduce combined yield losses to weeds, insects, and diseases worldwide by as much as 40% for maize, 50% for wheat, 75% for potato, and 80% for cotton. In meeting future food needs for the world’s growing population, it is important to improve the eVectiveness of disease management as new higher yielding crop cultivars are developed. For developing countries, in particular, it is essential to strive for long‐term consistency in disease management as well as short‐term eVectiveness. Countries with substantial resources and highly developed commercial and transportation infrastructure can withstand intermittent
Table I Estimated Yield Losses Worldwide to Diseases in Six Major Crops from 2001 Through 2003 (Data from Oerke, 2006) Crop Rice Wheat Maize Potato Soybean Cotton
Lossa (%)
Rangeb (%)
10.8 10.2 8.5 14.5 8.9 7.2
7–16 5–14 5–14 7–24 3–16 5–13
a Actual losses incurred in spite of accepted disease management practices in commercial production. b Range among means of losses from 19 regions of the world defined according to crop production conditions and production intensity.
322
D. D. STUTHMAN ET AL.
epidemics of important crops by drawing on stored reserves and importing produce from other parts of the world. They may resort to emergency implementation of new chemical disease control agents when existing crop cultivars lack suYcient resistance to the epidemic disease or when the pathogen has developed tolerance to previously used pesticides. In developing countries with fewer resources, such swift changes in food production and distribution are not so easily accomplished. Epidemics of major crops in developing countries can mean extreme privation and, at times, even starvation for the poorer members of the population (Simmonds, 1985). Thus, the goal in breeding disease resistant cultivars of major crops should emphasize long‐term reliability of the resistance. Although application of fungicides and nematicides plays an important role in protecting crops from disease, disease resistance is generally regarded as the preferred method of disease management for several reasons. Disease resistance does not entail potential risks to nontarget organisms in the environment. By avoiding the added labor and fuel costs involved in pesticide application, eVective disease resistance is also more economical than spraying pesticides on crop foliage. For these reasons and for safety of farm workers, Khush (1995) declared that relying on chemical control of diseases and pests of major food crops over prolonged periods is not practical in tropical climates. Breeding for resistance is the logical approach to take to help meet future needs for increased food supply. In the following, we will review the methods for identifying disease resistance most likely to be durable in its eVectiveness and incorporating that resistance into improved cultivars of important crops.
B. CAUSES OF PLANT DISEASE EPIDEMICS (THE DISEASE TRIANGLE) Before considering approaches used to evaluate and select disease resistance in crop breeding programs, it is useful to review the key elements of plant disease epidemics that determine their occurrence and severity. Briefly, the three key elements, which are commonly referred to as the disease triangle (Agrios, 2004), are a susceptible host, a virulent pathogen, and an environment conducive to disease development. If each of these three elements is visualized as one side of the disease triangle, the area within the triangle represents the potential severity of the epidemic. 1. Abundance and Susceptibility of the Host For most plant diseases, particularly foliar diseases caused by fungi, there is positive correlation between host density and severity of disease epidemics (Burdon and Chilvers, 1982). The distance between individual host plants
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
323
increases as the population density of host plants declines. The further pathogen spores have to travel from infected source plants, the more dilute the inoculum load will be when it reaches the nearest uninfected plants. In addition, high‐density populations of plants with lush growth produce dense canopies of foliage that increase the humidity within the canopy and thereby favor pathogens that require a film of moisture for their initial stages of infection. Zadoks (1993) listed nine factors that have contributed to the loss of stability of crop productivity in modern agriculture. The first four factors— increase of field aggregation, increase of field size, increase of host plant density (includes use of fertilizer), and increase of crop species uniformity through specialization—relate directly to host density. Two other factors listed by Zadoks—increase of genetic uniformity at the cultivar level, and breeding lapses (too much reliance on race‐specific resistance and neglect of resistance to currently less‐damaging disease)—contribute to greater susceptibility of the host population.
2.
Abundance and Virulence of the Pathogen
Two of the factors listed by Zadoks (1993) that contribute to the loss of crop production stability relate directly to the presence or abundance of pathogens. First, the increase of international exchange of seed and planting stock has contributed to the dispersal of many pathogens into parts of the world where they did not previously exist. Second, the increase of farmer specialization on a few cash crops has led to shorter crop rotations and, thus, greater survival of pathogen inoculum from one crop cycle to the next. Of course, an increase in density of susceptible host plants promotes an increase in the pathogen population density as well. The development of more specialized agriculture with greater field aggregation and larger fields devoted to more uniform crop cultivars with higher planting densities and more use of fertilizer have collectively produced a need for levels of disease resistance much greater than would have suYced for the wild relatives of our crop plants in natural ecosystems.
3.
Contribution of Environmental Factors
For each plant disease there are specific sets of environmental conditions that favor disease development. Many soilborne pathogens are most damaging in wet soil although some are favored by dry conditions. Some foliar diseases, such as wheat stem rust, thrive in hot weather while others, such as wheat stripe rust, require cool temperatures. Nearly all foliar pathogens
D. D. STUTHMAN ET AL.
324
(with the notable exception of powdery mildew fungi) require high moisture for their initial growth on leaf or stem surfaces before they penetrate the interior regions where moisture is not limiting. Typically, breeding for disease resistance involves exposure of breeding populations to induced epidemics or to natural epidemics in known ‘‘hot spots’’ for the disease. In either case, the breeder’s objective is to optimize the pathogen and environmental sides of the disease triangle to provide consistent epidemics of suYcient severity to clearly damage susceptible breeding lines without also destroying lines with useful levels of resistance. In selecting pathogen isolates or populations, disease inoculation procedures, and environmental interventions to enhance disease development, it is necessary to consider not just the three sides of the disease triangle as main eVects but also the interactions between them.
4.
Host Environment Interactions
Seedling blights of wheat and maize are a classic example of the role of host environment interactions in disease development (Dickson, 1923). The soilborne fungi that are most commonly involved in seedling blights are weak pathogens that do not seriously damage the host plants except under conditions in which seedling growth is slow and the development of defensive morphological barriers to root and hypocotyl infection, termed ‘‘hardening oV,’’ is delayed. The same soilborne fungi can cause seedling blight in both wheat and maize, but the diseases occur under diVerent environmental conditions for the two crops. Seedling blight in winter wheat is most severe at soil temperatures of 20–28 C and does not occur below 12 C. In maize, however, seedling blight does not occur at soil temperatures above 24 C and is most severe at 8–20 C. The reason for these diVerences is that wheat is a cool season crop adapted to rapid seedling development in cool soils, whereas maize is a warm season crop adapted to rapid seedling development in warm soils (Dickson, 1923). One way in which plant breeders have reduced seedling blight in early planted maize has been to select for lines with greater cold tolerance and faster germination in cool soils. Susceptibility of upland and lowland ecotypes to rice blast is another example of host environment interaction. High silicon concentration in epidermal cells of rice foliage is related to reduced disease severity, apparently because silicic acid condenses into a glass‐like coating of polymerized SiO2 on epidermal surfaces that impedes pathogen penetration. The highly weathered upland soils in the tropics contain lower silicon concentrations than the rich alluvial soils of the bottom lands. Ecotypes of rice adapted to lowlands are generally more susceptible to rice blast than the upland rice ecotypes when grown side by side in upland plots without supplemental silicon applied
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
325
to the soil. The reason for this is that the upland ecotypes have evolved more eYcient mechanisms of silicon uptake from the silicon‐deficient soils of the uplands and, thus, are able to form a stronger barrier against pathogen penetration (Winslow et al., 1997). Host environment interactions also are prominent in many diseases caused by biotrophic pathogens, such as the rust fungi, that are highly specialized parasites. For example, the partial resistance of the wheat cultivars Gaines and Nugaines to stripe rust caused by Puccinia striiformis is eVective during the summer months in the Pacific Northwest region of the United States. This resistance is referred to as high‐temperature adult plant (HTAP) resistance. Although HTAP resistance does not protect winter wheat plants during cool spring weather, it becomes eVective during early summer in nearly all years before the stripe rust epidemics have progressed far enough to cause significant yield losses. The HTAP resistance of Gaines and Nugaines wheat has remained eVective for more than 40 years (Line, 2002). Many major genes for race‐specific resistance to stem rust and leaf rust in wheat also are temperature sensitive. In most of the cases of temperature‐ sensitive major genes, the resistance is eVective at low temperatures throughout the development of the wheat plants, but the resistance breaks down at high temperatures. However, a few rust resistance genes in wheat are known to be eVective at high temperatures but ineVective at low temperatures (McIntosh et al., 1995).
5.
Pathogen Environment Interactions
Several types of pathogen environment interactions may be encountered in field plot tests. DiVerences in soil moisture due to low spots, and diVerences in water‐holding capacity, and so on, may result in nonuniform distribution of soilborne pathogens in plots. Edge eVects may also be important. For example, vector‐borne viruses may be more prominent at the edges of fields if weeds or adjacent crops harbor either the virus or the insect vectors of the virus.
6.
Host Pathogen Interactions
The above examples of host environment and pathogen environment interactions illustrate some of the reasons why heritability of disease resistance may be less than desired in field tests of breeding lines. Host pathogen interactions, however, have been the most frustrating for plant breeders.
326
D. D. STUTHMAN ET AL.
The most extreme examples occur as gene‐for‐gene relationships between resistance in the host and virulence in the pathogen, first demonstrated by Flor (1955) for flax rust. In typical gene‐for‐gene relationships, dominant genes for resistance in the host provide complete or nearly complete resistance against races of the pathogen that have a dominant avirulence gene that matches the host resistance gene but not against pathogen races lacking the matching avirulence gene. Thus, for each gene for race‐specific resistance in the host, there is a corresponding gene for avirulence in the avirulent races of pathogen. Also typically, for each gene for race‐specific resistance in the host, there is corresponding recessive virulence gene in the virulent races of the pathogen. That is, in a typical gene‐for‐gene relationship, every gene for race‐ specific resistance can be overcome by a virulence gene that is either present or potentially present in some members of the pathogen population. Since Flor’s (1955) discovery of the gene‐for‐gene relationship between resistance and virulence in flax rust, similar gene‐for‐gene relationships have been described for a large number of other plant diseases and plant pest problems. As a general rule, gene‐for‐gene relationships are found with pathogens that are highly specialized biotrophic parasites that reproduce poorly or not at all outside living host plants. Some well‐known examples include the rust and powdery mildew diseases of cereals, apple scab, potato late blight, Hessian fly on wheat, and the golden nematode of potato. In some cases, resistance genes are not always dominant and virulence is not always recessive. Resistance against avirulent races is usually highly eVective, but some cases of gene‐for‐gene correspondence involving genes conferring only partial resistance are known. There is not always a perfect one‐to‐one correspondence between resistance genes and avirulence genes. Still, systems of race‐specific resistance and virulence are generally characterized by the combination of qualitatively inherited resistance in the host that can be overcome completely or nearly completely by qualitatively inherited virulence in the pathogen. The frustration suVered by plant breeders in attempts to use major genes for race‐specific resistance results from the false hope inspired by the discovery of new resistance genes that show nearly complete eVectiveness against all known races of the pathogen at the time of discovery. Indeed, the very eVectiveness of the new resistance genes is the basis for their undoing. Virulent races, which have the necessary matching virulence gene to overcome the resistance, may be too rare to be detectable in routine surveys before the new resistance gene is used widely, but the virulent races may increase rapidly as soon as competition from old races, which lack the necessary virulence gene, are eliminated. Often the useful life span of the new race‐specific resistance is only a few years, and the rise of virulent races to overcome it is evidenced by rampant epidemics and severe yield loss to the disease.
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
C.
327
EXAMPLES OF PLANT DISEASE IN NATURAL POPULATIONS
Plant disease epidemics are much more evident in cultivated crops than in wild plants in natural ecosystems, but devastating epidemics also have occurred in natural ecosystems. Two well known examples within the last 100 years in North America are chestnut blight, which eliminated the American chestnut as a dominant species in the forests of the eastern United States, and Dutch elm disease, which destroyed large numbers of American elms both in natural ecosystems and in urban plantings, where American elm was the favorite shade tree in many American cities and towns. In both cases, the pathogen responsible for the epidemics was introduced into North America from Asia, where the native hosts for the pathogens were relatively resistant. New combinations of host and pathogen in natural ecosystems seem especially prone to devastating epidemics (Burdon, 1987). In long established host–pathogen combinations that have coevolved together, disease epidemics typically occur in more limited spatial range, are shorter in duration, and are of limited severity (Burdon, 1987; Segal et al., 1980). Native populations of wild emmer wheat (Triticum dicoccoides) in the nature preserve at Ammiad, Israel tend to be relatively free of leaf rust infection caused by Puccinia triticina (Anikster, 2001; Dinoor et al., 1991). Leaf rust develops late in the natural stands of T. dicoccoides, and only a few plants were severely attacked in their natural habitat. On the other hand, leaf rust is common on cultivated wheat in Israel. When seeds from T. dicoccoides plants at Ammiad were planted in cultivated nurseries a short distance from the nature preserve, they became heavily infected with leaf rust. Anikster et al. (2005) tested accessions of T. dicoccoides from throughout Israel for adult partial resistance to endemic races of P. triticina in field plot tests. They found that accessions from Galilee and adjacent areas, where the natural populations of T. dicoccoides are most dense and genetically diverse, had the highest levels of adult plant resistance to leaf rust. DiVerences in native disease resistance can be found over relatively small spatial scales when there are marked diVerences in local climatic conditions. Temperature and moisture are the two dominant environmental factors that determine the distribution of disease in natural ecosystems. For example, Wahl (1970) found that resistance to crown rust caused by Puccinia coronata in wild oat (Avena sterilis) populations in Israel varied according to the climate. Resistance was rarely found in collections of A. sterilis from the arid Negev region of southern Israel, where crown rust was rarely seen on the host. However, in hills of northern Israel, which are cooler and receive more rainfall, crown rust occurs quite commonly, and Wahl (1970) found that A. sterilis plants with resistance to crown rust occurred quite frequently among those collections. Similar results were found for resistance in wild barley (Hordeum spontaneum) to leaf rust (Manisterski et al., 1986) and
D. D. STUTHMAN ET AL.
328
powdery mildew (Dinoor and Eshed, 1987; Segal et al., 1980). Collections of H. spontaneum from hot, dry regions of Israel were more susceptible to leaf rust and powdery mildew than were collections from other areas of Israel where rust and mildew were more prevalent. These results are consistent with Harlan’s (1976) conclusion that disease resistance tends to be lost from plant populations when there is no disease pressure to maintain selection for resistance.
II. CONCEPTS OF RESISTANCE TO PLANT DISEASE USEFUL IN BREEDING EFFORTS A. A PRAGMATIC APPROACH In breeding for resistance, it is not necessary to seek either complete resistance or complete understanding of the resistance. Resistance matters only in so far as it protects yield (Simmonds, 1988). According to Simmonds (1988) ‘‘A state of no disease is rarely a realistic objective. Furthermore, it is not necessarily even a desirable one. The yield‐disease intensity curve is rarely linear so that low, even moderate levels of disease are often found to have eVects so small as to be unmeasurable or economically trivial. Therefore, ‘enough resistance is enough’ is a good practical maxim: how much is enough will, of course, vary from place to place, according to disease severity.’’ Yield‐damaging epidemics, of course, are a clear sign that more resistance is needed. Some general concepts may be useful in guiding the approach to selecting for resistance if the objective is to provide not complete resistance, but rather ‘‘adequate levels of resistance that are reliable over the years’’ (Simmonds, 1988). Without fully understanding the nature of all forms of disease resistance, we may still make reasonable judgments about which types of resistance are most likely to remain durable. The following considerations may be helpful.
1.
Disease Escape or Avoidance
Some plants may avoid disease in ways that are not generally considered to be true resistance mechanisms. For example, early maturing plants may complete their growth and seed production mostly before epidemics of foliar diseases have progressed far enough to cause significant damage. Wheat stem rust epidemics in the United States begin with overwintered infections in winter wheat plants near the Gulf Coast. The pathogen, Puccinia graminis, develops best in warm weather, so stem rust epidemics develop slowly during
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
329
the spring. One reason for the decline of stem rust damage in the Central Plains states of Oklahoma and Kansas was the switch to earlier maturing winter wheat cultivars in those states in the 1930s and 1940s (Leonard, 2001). Reduction in the population density of host plants can result in another type of partial disease avoidance. In mixtures of host and nonhost plants, the distance between individual host plants increases as the density of host plants within the mixture declines. This results in a dilution of wind‐borne inoculum available for spread of disease from infected hosts to healthy host plants. In intensive production of crops in monoculture, of course, this type of disease escape is reduced and the chances of plant‐to‐plant spread of disease are correspondingly increased.
2.
Morphological Resistance
Some morphological characteristics of certain plants may enable them to avoid infection. For example, tuber rot in stored potatoes is initiated primarily in wounds to the tubers incurred during harvesting and/or transport to storage. Rapid healing by development of a suberized layer over the wounded tissue is considered the most critical aspect of resistance to decay in potatoes (Stevenson et al., 2001). Floral development is important in the resistance of wheat to ergot caused by Claviceps purpurea, a significant disease problem in rye but not wheat. The pathogen invades the ovules by way of the stigmas of the rye florets. Once the ovules are fertilized, the stigmas are no longer receptive to infection. Because rye is typically cross‐pollinated, the unfertilized stigmas are exposed to ascospores, which are released by the pathogen into the air at the time of host flowering. Wheat, on the other hand, is self‐pollinated with fertilization generally taking place within the florets before they open. Thus, wheat stigmas are no longer receptive to infection by the time they become exposed to pathogen spores in the air (Wiese, 1987). Use of male sterility factors in the development of hybrid wheat cultivars revealed that unfertilized wheat florets are inherently susceptible to ergot. Male sterile lines of wheat tend to be infected as heavily as rye by ergot (Alexander, 1992). The failure of some barley cultivars to extrude anthers from florets after pollination has also been cited as an example of at least partial escape from Fusarium head blight (SteVenson, 2002). One path of entry for the pathogen into the developing grain is through extruded anthers that present a rich growth medium for the pathogen. Breeding for active mechanisms of resistance to the pathogen in other floret tissues will be more eVective when the anthers remain within the florets. Other means by which plants may avoid disease involve hairs or waxes that cover the leaf epidermis and then repel water needed by the pathogen to
D. D. STUTHMAN ET AL.
330
grow on the surface before penetrating the host (Agrios, 2004). On upright foliage with waxy surfaces, the water droplets may roll oV the leaves allowing them to dry rapidly. Also, some pathogens (e.g., viruses) are typically introduced into host plants by specific insect vectors. Plants that are unattractive to potential vectors may escape infection even though they are susceptible to the pathogen. A form of morphological resistance to Magnaporthe grisea, the rice blast pathogen, was mentioned earlier in the subsection on host environment interactions. When adequate levels of silicon are present in the soil, rice plants extract silicon and deposit it as a coating of polymerized SiO2 on epidermal cells of aerial parts of the plants, which renders their surfaces more diYcult for M. grisea to penetrate. The resulting resistance is not complete, but it is useful enough that in rice culture on silicon‐deficient soils (either highly weathered upland soils or organic soils) it can be economical to apply calcium silicate. Calcium silicate applications have been shown to reduce disease severity and increase rice yield (DatnoV et al., 1991). Successful selection of rice cultivars for more eYcient silicon uptake should have a similar eVect. In a classic study of morphological resistance of wheat to stem rust caused by P. graminis, Hart (1931) found that seedlings exhibited no morphological resistance to P. graminis, but in adult plants the pathogen was able to grow only in the chlorophyllous collenchyma bundles of the wheat stems. Cultivars in which the collenchyma bundles occurred as narrow, isolated strands separated by broad bands of sclerenchyma were partially resistant to stem rust in adult plants.
3.
Preformed Chemical Barriers to Pathogen Invasion
Plants produce a wide variety of chemical defenses with general antibiotic properties. These include, among others, tannins, phenolic compounds, dienes, saponins, proteases, and hydrolytic enzymes (Agrios, 2004). Many of these defense compounds are produced independently of the presence of potential pathogens. They serve as a general defense against nonspecialized pathogens. It is to be expected that plants within species will exhibit genetic variation for the concentrations of such defense compounds that accumulate within the plant’s tissues. Seeds typically have higher concentrations of antibiotic compounds than vegetative parts of annual plants because of the critical need for protection of dormant seeds until they resume growth in subsequent growing seasons. It is also clear that specialized pathogens that have evolved to attack particular families or genera of plants must have undergone natural selection for tolerance to the preformed defense compounds of their hosts or for mechanisms to detoxify those compounds.
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
331
In each of the examples of disease escape and morphological resistance cited above, the factor responsible for limiting disease development is expressed independently of the presence or absence of the pathogen. It does not appear that these factors evolved as responses to a particular pathogen species, let alone a specific pathogen race. Therefore, it also seems unlikely that genotypes in the populations of a particular pathogen species will contain significant genetic variation with regard to their abilities to overcome the factor leading to disease escape or morphological resistance. At the same time, it is unlikely that plant breeders will deliberately select for specific disease escape or morphological resistance mechanisms because, with few exceptions, the eVort required to score progeny for the specific mechanism would be too costly. Still, one may assume that methods used to screen progeny for partial resistance would lead to a degree of selection favoring some mechanisms of disease escape or morphological resistance.
4.
Induced Responses to Invasion
Many pathogens of storage organs of plants, such as tubers, gain entry through breaks in the epidermis resulting from insect feeding or mechanical wounding. Plants respond to wounds in a nonspecific production of corky tissue that walls oV the broken area of the epidermis. Similar barriers are produced in the woody stems and roots of trees or other perennial plants in response to canker‐forming pathogens. The speed at which a plant can produce an impermeable barrier to further infection is a measure of its generalized resistance to these types of pathogens (Agrios, 2004). Similarly, cell walls in the epidermis may induce thickenings in response to pathogens attempting to penetrate their outer walls. The wall thickenings often are infused with phenolic substances that are toxic to most potential invading microbes. In wilt diseases caused by fungi, the pathogen typically grows through the xylem and eventually cuts oV water flow to the aerial parts of the plant. Infected plants may eVectively wall oV the pathogen by forming tyloses within xylem cells at and around the site of infection. The tyloses block the flow of water through the aVected xylem elements, but they may also prevent further growth of the pathogen into additional xylem elements. Plants also respond to invasion by activating pathways to produce characteristic antibiotic compounds known as phytoalexins in the cells around the site of attack. If these responses occur soon enough, and are extensive enough, the plant may display enough partial resistance to keep the damage to a low level unless the plant is attacked by an overwhelming amount of inoculum (Agrios, 2004). The rate of response of the host may depend on its ability to recognize that it is being invaded, and the success of the pathogen in gaining entry may depend on its ability to evade the plants detection mechanisms.
332
D. D. STUTHMAN ET AL.
B. CLASSIFICATION OF TYPES OF RESISTANCE FROM A BREEDER’s PERSPECTIVE Plant breeders are directly concerned with three aspects of disease resistance as they relate to breeding strategies: (1) inheritance of the resistance (monogenic or polygenic), (2) eVectiveness (complete or partial), and (3) specificity (race‐ specific or general in its eVectiveness). There has been a tendency to reduce these six characteristics into just two choices: monogeinc, race‐specific resistance that is completely eVective against avirulent races, or polygenic resistance that is partially eVective against all races (Vanderplank, 1968). This two‐way classification was one of convenience, but it is certainly an oversimplification.
1. Inheritance Monogenic resistance typically refers to resistance in which the phenotypic eVect of a single major gene for resistance can be recognized with reasonable certainty and progeny from a cross can be scored as either having or lacking that particular gene. Such resistance does not need to be completely eVective to be recognizable by routine visual inspection. Cultivars with polygenic resistance also may be recognized as distinct from susceptibility in routine tests, but the separate contributions of their individual minor genes may not be so easily discernible. The sometimes invoked distinction of complete resistance versus partial resistance obviously oversimplifies the range of levels of resistance that occur in populations of plants to the diseases that aVect them. First, major genes for resistance do not necessarily render plants immune from infection. For example, several infection types (IT) are described in cereal rust diseases that are regarded as resistant reactions (McIntosh et al., 1995). Only the IT 0 reactions, which show no macroscopic evidence of pathogen attack can be regarded as immunity. Few genes for rust resistance in cereals condition this level of resistance. Reactions designated as fleck (IT;) do not allow any pathogen sporulation, but do show visible evidence of pathogen‐induced death of host tissue. It is common, however, for resistance genes to condition IT 1 (very small pustules surrounded by a ring of necrotic tissue) or IT 2 (moderately small pustules surrounded by a chlorotic halo) reactions. With the gene Lr34 for leaf rust resistance in wheat, the distinction is not so clear. The resistance of Lr34 is not expressed in seedlings, and even in adult plants the visible appearance of the resistance is relatively subtle. Lr34 in adult plants conditions a moderate level of resistance that reduces somewhat the numbers of successful infections and is characterized by smaller pustules with less sporulation than on fully susceptible plants (Rubiales and Niks, 1995).
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
333
While Lr34 is recognized as an important gene for wheat leaf rust resistance for reasons that will be described later, it clearly is not a gene for complete resistance.
2.
Effectiveness
As previously indicated, the distinctions between high and low levels of resistance should be made with respect to individual genes. It is quite possible that some cultivars with an accumulation of many genes for partial resistance may remain almost completely free of disease under conditions in which fully susceptible cultivars are heavily infected. For practical purposes, such a resistant cultivar might be said to show nearly complete resistance within the area of its normal use even if it has no genes for complete resistance.
3.
Specificity
The third distinction, race‐specific versus nonspecific resistance addresses the central issue of durability of disease resistance in crops. Vanderplank (1968) divided resistance into two types, which he termed vertical (race‐ specific) and horizontal (nonspecific). Although others (e.g., Robinson, 1976) equated vertical resistance with monogenic resistance and horizontal resistance with polygenic resistance, Vanderplank’s terms were epidemiological rather than genetic. Vertical resistance reduces initial inoculum in an epidemic by screening out avirulent races in the pathogen population, but vertical resistance has no eVect on the rate of increase of the virulent races in the epidemic. Horizontal resistance, on the other hand, reduces the rate of pathogen increase by being at least partially eVective against all races. What Robinson and others overlooked is that some genes that give nearly complete resistance to certain diseases are race nonspecific, while some genes for partial resistance are race‐specific. Simmonds (1985) expanded the classification of resistance, identifying four types: (1) vertical resistance as defined by Vanderplank, (2) pathogen‐ nonspecific major gene resistance to cover those cases of monogenic resistance that are known to lack race specificity, (3) horizontal polygenic resistance, and (4) interaction resistance, which relies on the use of mixed populations of host lines with diVerent genes for vertical resistance to avoid rapid selection of races virulent to any single resistance gene in the mixture. This fourth type of resistance, which occurs in heterogeneous mixtures of susceptible and resistant host genotypes, is roughly equivalent to what we referred to earlier as a
334
D. D. STUTHMAN ET AL.
form of disease avoidance. That is, the reduced disease severity in the mixture is due to reduced density of susceptible host plants, but it also may have an added component of induced resistance resulting from incipient infections by avirulent races of the pathogen on host lines with race‐specific resistance. The induced resistance, even if its eVects are only local, may substantially reduce the ability of virulent races to infect the same plants. Simmonds (1985) recognized that his classification of resistance is still an oversimplified version of the real diversity of resistance types. For example, he conceded that combinations of weakly expressed genes for vertical resistance would likely resemble horizontal polygenic resistance, at least until the pathogen population adapted by selection of matching virulence genes. Thus, polygenic resistance should not necessarily be assumed to be nonspecific just because it is polygenic. Simmonds’ recognition of pathogen‐nonspecific major gene resistance is less ambiguous. Resistance in maize to Cochliobolus heterostrophus, in oat to Cochliobolus victoriae, and in sorghum to Periconia circinata is each due to a single gene, whose allele is, in eVect, a gene for sensitivity of the host to a specific toxin produced by the pathogen. Pathogen genotypes that lack the ability to produce the toxin are avirulent, and host plants without the gene for toxin sensitivity are resistant to the disease (Leonard, 1984). Presumably, mutations in the pathogen gene for toxin production would cause the gene product to fail to interact with the corresponding host gene for toxin sensitivity and, thus, would not produce race‐specificity of the host resistance. In the case of maize leaf spot caused by Cochliobolus carbonum, the host resistance gene encodes an NADPH‐dependent reductase that inactivates the HC‐toxin (Takken and Joosten, 2000). No alternative form of the HC‐toxin gene has been found that is not inactivated by the reductase. In other cases, single genes for resistance have proven durable, but are not necessarily known to be nonspecific. For example, cultivars of cabbage with a single gene for resistance to cabbage yellows disease, caused by the soilborne pathogen Fusarium oxysporum f. conglutinans, have retained their resistance in fields since their release in 1926 (Parlevliet, 2002). The mechanism of that resistance and the reason for its durability are not known. Simmonds (1991) listed 53 fungal diseases of 25 diVerent crops for which horizontal, polygenic resistance has been described. The crops listed include all of the major cereal crops, potato, cassava, alfalfa, peanut, cotton, tobacco, sugarcane, and various fruits and vegetables. The pathogens included were rusts, smuts, powdery mildews, downy mildews, and species of Phytophthora, Fusarium, Septoria, Drechslera, Colletotrichum, Cercospora, and Diplodia as well as other fungi. He also listed 9 bacterial diseases on 8 crops and 11 virus diseases on 7 crops for which horizontal polygenic resistance has been found. In fact, Simmonds (1988) regarded horizontal
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
335
polygenic resistance as the rule rather than the exception. He suggested that most plants have some level of this type of protection against most, if not all, of their diseases.
4.
Hypersensitive Resistance
If genetic variation for polygenic nonspecific resistance exists for nearly all diseases in all crop species, the challenge may be to recognize the resistance as nonspecific when it occurs in breeding populations. This is not a trivial challenge. Johnson (1978, 1988) took the view that only time will tell. He defined durable resistance as resistance that did not suVer any significant loss of eVectiveness after many years of widespread use in the presence of the pathogen. No doubt, this definition is safe, but it limits the choice of resistance donors to old cultivars that are many years out of date in terms of yield or quality breeding. A more pragmatic approach may be to identify types of resistance that are least likely to be durable and exclude them from breeding populations. The best documented cases of race‐specific resistance are based on a hypersensitive reaction in the host in which initial contacts in incompatible combinations of avirulent pathogen with resistant host trigger rapid death of the contacted host cells with release of metabolites injurious to the pathogen. For diseases caused by biotrophic pathogens, we may assume that hypersensitive resistance will be race‐specific. It is unlikely that biotrophic pathogens could have survived if they had not evolved races that can avoid inducing hypersensitive responses in at least some genotypes of their hosts.
III. WHAT MAKES DISEASE RESISTANCE DURABLE? A. RESISTANCE IN WILD PLANT SPECIES The existence of disease resistance in wild plant species in natural ecosystems is well documented through the use of wild relatives of cultivated crops as sources of resistance in crop breeding programs (Wahl, 2003). Although durability of resistance in wild plant species is not well documented, it is generally observed that breakdown of resistance, if it occurs in natural ecosystems, does not result in ‘‘boom and bust’’ cycles of short periods of nearly complete resistance followed by total collapse of eVectiveness of race‐ specific resistance, which have been seen so often in agricultural crops. Instead, the coevolution of hosts and pathogens in natural ecosystems
336
D. D. STUTHMAN ET AL.
tends to produce a dynamic equilibrium between them (Dinoor and Eshed, 1984; Segal et al., 1980). The most extensively studied host–pathogen associations in natural ecosystems are the rusts and powdery mildews of wild relatives of barley, oat, and wheat in the Fertile Crescent region of southwest Asia. A great diversity of genes for race‐specific resistance has been found in the progenitors of these small grain crops in southwest Asia, particularly in Israel in recent years. Similarly, there is a great diversity of races of rust and powdery mildew pathogens in the natural ecosystems. This diversity is evidence that single pathogen races in natural ecosystems are not able to dominate the pathogen population when the host population is a diverse mixture of diVerent resistance genotypes (Wahl, 2003). In addition to monogenic, race‐specific resistance, polygenic, partial (presumably nonspecific) resistance is also commonly found in wild plant species in their natural habitats (Alexander, 1992; Browning, 1974). Indeed, hypersensitive, race‐specific resistance to leaf rust, P. triticina, which is expressed in both seedlings and adult plants, was found in relatively few accessions of wild emmer wheat, T. dicoccoides, from indigenous populations in Israel, whereas quantitative resistance expressed in the adult plant stage was quite common at several sites in Israel (Anikster et al., 2005). No doubt, nonspecific partial resistance plays an important role in stabilizing the eVectiveness of disease resistance of plants in natural ecosystems.
B. IMPACT OF AGRICULTURE ON RESISTANCE As mentioned earlier, Zadoks (1993) pointed out a number of aspects of modern agriculture that have placed increased disease pressure on cultivated crops relative to the disease pressure experienced by wild progenitors of those crop species. Large fields with lush, pure stands of highly fertilized, well‐ irrigated, genetically uniform crops oVer an unnaturally rich breeding ground for specialized parasites that can attack those crops. Furthermore, the widespread cultivation of just a few dominant crops over broad geographic areas now allows for continental spread of epidemics that would more likely be localized in natural ecosystems. It is not surprising, therefore, that modern agriculture requires a greater level of resistance to protect crops from disease than might suYce to limit disease development in populations of similar plant species in natural ecosystems. Add to that the demands of growers and consumers for unblemished produce, and it becomes easy to understand why plant breeders continued to search for new genes for hypersensitive resistance to disease even after the race‐specificity of such resistance became apparent.
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
337
Table II Characteristics Associated with Extremes for Risks of Pathogen Adaptation to Overcome Race‐Specific Resistance (Adapted from McDonald and Linde, 2002) Higher risk of adaptation High mutation rate Transposons active Large pathogen populations Minimal bottleneck eVect in oV season Minimal genetic drift High gene flow/genotype migration Wind dispersal of inoculum Human mediated long distance transport Mixed reproduction by pathogen Annual sexual cycle with asexual cycles of reproduction during epidemics Major resistance genes deployed continuously over large areas
Lower risk of adaptation Low mutation rate No transposons Small pathogen populations Extinction of local populations in oV season Significant genetic drift Low gene flow/genotype migration Soilborne or splash dispersed inoculum EVective quarantines against movement by humans Asexual reproduction by pathogen Only asexual propagules Major resistance genes deployed in spatial or temporal diversity in crop rotations or mixed host populations
Pathogens that maintain large population sizes, undergo a mixed mode of sexual and asexual reproduction, and have high rates of gene flow across geographical regions are the most likely to cause problems of breakdown of race‐specific resistance in crops (McDonald and Linde, 2002; Wolfe, 1983; Table II). Mutations for virulence are more likely to persist from year to year in large populations of pathogens than in small populations. Thus, the virulent mutants will respond more consistently to selection imposed by the use of cultivars with novel genes for race‐specific resistance. In this regard, the size of the population at its lowest point during the year is more important than at the maximum size reached at the height of an epidemic. Sexual reproduction in the pathogen allows for eYcient recombination of virulence genes needed to meet new combinations of race‐specific resistance genes used by plant breeders in the development of new cultivars. On the other hand, the ability to undergo multiple cycles of asexual reproduction during the epidemic allows for rapid selection of pathogen genotypes with optimal combinations of genes that enhance parasitism on the predominant cultivars of the crop in any period of time. Gene flow is important in spreading new virulent mutants over broad geographical regions where the crop is grown and thereby increasing the eVective size and genetic diversity of the pathogen population. This provides more opportunities for selection of rare virulence alleles. Thus, fungi such as the rusts and powder mildews, which produce airborne spores that can survive long distance dissemination, create the greatest hazards for breakdown of race‐specific resistance. Cycling of the pathogen between fall‐ and
338
D. D. STUTHMAN ET AL.
spring‐sown cultivars of the same crop increases the hazard by eliminating the population bottleneck of oV‐seasons with limited availability of suitable host plants.
C. EFFORTS TO DELAY BREAKDOWN OF INHERENTLY TRANSIENT RESISTANCE The eVectiveness of race‐specific resistance, while still transient, may be extended somewhat in agricultural systems by introducing a level of diversity into production of the crop. For this reason, Wellhausen (as quoted by Borlaug, 1966) recommended that in the tropics and subtropics, maize should be grown as genetically heterogeneous open‐pollinated cultivars rather than as genetically homogeneous hybrids. Diversity in crops may be inter‐ and intraspecific and temporal as well as spatial. Crop rotations have long been used in traditional agriculture as a way to reduce damage from pests and pathogens such as soilborne fungi and nematodes, which have limited ranges of dispersal (Nusbaum and Ferris, 1973). Crop rotation has also been adapted to extend the eVectiveness of race‐specific resistance. Jones et al. (1967) recommended a rotation of susceptible and resistant potato cultivars separated by 2 years of nonhost crops to control the potato cyst nematodes, Globodera rostochiensis and G. pallida. A similar rotation involving soybean cultivars was proposed to extend eVectiveness of resistance to the soybean cyst nematode (Riggs, 1977). An eVect similar to that of cultivar rotation may be obtained for highly mobile pathogens by deployment of diVerent sets of genes for race‐specific resistance in diVerent geographical regions. Vanderplank (1968) emphasized that in continental epidemics such as those of wheat stem rust, genes for race‐ specific resistance may remain eVective for many years in some regions if those resistance genes are carefully excluded from the crop in the region where the epidemics are initiated each year. He cited the durability of resistance provided by Sr6, a gene for race‐specific resistance to stem rust, in spring wheat cultivars in North America as a clear example of this principal. Sr6 has not been used in winter wheat cultivars, so it remains eVective against races of P. graminis that overwinter in the Southern Plains winter wheat area even now, nearly 30 years after Vanderplank (1968) publicized its potential durability in spring wheat cultivars. A similar gene deployment scheme for oats and oat crown rust in the United States was proposed by Browning et al. (1969), but it was not successful for several reasons. First, Rhamnus cathartica, the alternate host of the oat crown rust fungus, P. coronata, is abundant throughout much of the Northern Plains, so virulent races of P. coronata that are selected on resistant oat cultivars in the Northern Plains during the summer can survive
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
339
the winter as teliospores, cycle through the alternate host in the spring, and spread back to oat to undergo further selection on the same resistant cultivars each year. This results in a successive annual increase in frequencies of races capable of overcoming race‐specific crown rust resistance genes that are widely used in the region. Interspecific diversity of crops within fields undoubtedly served a useful purpose in reducing epidemics and delaying breakdown of race‐specific resistance in traditional agriculture in developing nations (Aiyel, 1949), but it is not generally used in modern agriculture. Mixtures of wheat, barley, and oat, or even oat and peas were grown in North America (Warburton, 1915), but marketing problems likely limited mixed grain production to on‐farm use as livestock feed. For example, oat millers in the United States will not accept mixtures of barley and oat because of problems in separating barley and barley hulls from oats. On the other hand, barley and oat mixtures are still widely grown in Poland and mixtures of cereals with legumes are becoming increasingly popular, largely through farmers’ initiatives without directed research support (Czembor and Gacek, 1996). Jensen (1952) and Borlaug (1953) proposed the development of multiline cultivars as a method of introducing intravarietal diversity in cereal crops in a way that would be compatible with modern agriculture. Multiline cultivars are composed of individual lines that are phenotypically similar but diVer in genes for race‐specific resistance to an important pathogen. Multiline cultivars have been used successfully against wheat stem rust and stripe rust in Columbia and against oat crown rust in Iowa (Browning and Frey, 1969). One disadvantage of multiline cultivars is the cumbersome breeding program required to produce genetically similar lines with diVerent genes for race‐ specific resistance. Advances in yield or quality may be delayed by the heavy commitment required to produce a minimum of 8–10 backcross lines, each with a diVerent resistance gene. More rapid progress in accumulating durable resistance might be made by selecting for increased levels of polygenic, partial resistance in the breeding program. An alternative to multiline cultivars is the use of mixtures of separately developed cultivars of a crop that have complementary qualities including disease resistance. Mixtures of barley cultivars have proven eVective against powdery mildew (Blumeria graminis) in England (Wolfe and Barrett, 1980). Interest in research on the use of mixtures of barley cultivars to enhance durability of race‐specific resistance to powdery mildew in Europe has increased in recent years (Limpert et al., 1996). Pyramiding genes for race‐specific resistance has been a common method for attempts to increase durability of race‐specific resistance without resorting to increased genetic diversity in crops. The theory is that the combined resistance will be more durable if several simultaneous mutations to virulence in the pathogen are necessary to overcome the pyramided resistance
340
D. D. STUTHMAN ET AL.
combination. Of course special methods must be used to track the presence of multiple resistance genes in breeding lines if each new gene by itself gives near complete resistance to the pathogen population in the field. Selecting lines with multiple resistance genes might require considerable preliminary work to obtain suitable molecular markers that are closely linked to each new resistance gene before it can be included in the pyramiding program. When the resistance genes to be pyramided come from unadapted exotic lines, extra eVort is needed to avoid introducing agronomically undesirable genes into the recurrent parent with each new resistance gene that is added to the pyramid. Unfortunately, pyramiding race‐specific resistance genes has had limited success against biotrophic pathogens such as rust and powdery mildew fungi (Pink, 2002). Breeders rarely, if ever, have suYcient control over breeding lines to prevent the spread of individual resistance genes to other breeding programs in which they may be separated from the intended pyramided gene combination. As the resistance genes are used separately, pathogens are able to overcome all of the resistance in steps one gene at a time. Eventually all of the pyramided genes may be overcome, especially if the pathogen undergoes genetic recombination in a sexual cycle of reproduction. Pyramiding resistance genes should be more eVective against soilborne pathogens because their populations tend to be much more localized with minimal mixing with pathogen isolates exposed to other crop cultivars in other regions. Leach et al. (2001) pointed out that for some diseases caused by bacteria, it may be possible to design pyramids of genes for race‐specific resistance that will have long‐term durability. For example, mutants of Xanthomonas oryzae pv. oryzae that lack avirulence genes avrXa7 and avrXa5 are virulent on rice cultivars with race‐specific resistance genes Xa7 and xa5, but their capacity to induce disease lesions and multiply in the lesions is significantly reduced. That is, these avirulence genes perform important functions in parasitism by the pathogen, and loss of those functions carries a substantial fitness cost for the pathogen. Therefore, combining Xa7 and xa5 in a single rice cultivar should provide a reasonably high level of resistance that will be durable. Similar examples of fitness costs from loss of avirulence alleles are known for other bacterial diseases of plants, but not for diseases caused by fungi (Leach et al., 2001).
D. EXAMPLES OF DURABLE MONOGENIC RESISTANCE Takken and Joosten (2000) listed four fundamentally diVerent mechanisms by which monogenic resistance to diseases caused by fungi may be expressed: (1) The product of the R gene may inactivate a host‐specific toxin essential for pathogenicity. The Hm1 gene in maize which encodes an
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
341
NADPH‐dependent reductase that inactivates the HC‐toxin of C. carbonum race 1 is an example of this type of resistance. (2) A specific host gene product encodes a pathogenicity target. Absence of the target (i.e., presence of a resistance gene in place of the gene encoding the toxin‐sensitive target) provides resistance. The mitochondrial gene T‐urf 13 in maize, which confers male sterility also causes sensitivity to the T‐toxin of C. heterostrophus. Plants lacking T‐urf 13 are resistant. (3) The R gene may prime the plant defense response to make it more sensitive to pathogen attack. The only confirmed R gene of this type is the mlo gene in barley for resistance to powdery mildew caused by B. graminis f. sp. hordei. (4) The R gene may mediate recognition of a pathogen through specific interaction with a matching avirulence (Avr) gene product. The first three types of resistance are likely to be durable, but the fourth mechanism, which is characteristic of gene‐for‐ gene interactions involving hypersensitive host responses, is well known for lack of durability unless the pathogen has minimal genetic variability and limited gene flow. The mlo resistance of barley (mechanism 3) may be either a fortunate happenstance or a portent of more valuable but yet undiscovered resistance genes. The wild‐type Mlo allele is a negative regulator of defenses against powdery mildews that include localized cell death (Brown, 2002). Loss of Mlo (i.e., homozygosity for the recessive mlo allele) results in non‐race‐ specific deregulated cell death in response to invasion and leads to nearly complete resistance to powdery mildew. The homozygous condition also causes plants to suVer from spontaneous necrotic flecking of leaves, apparently without any known external cause. The deregulated cell death also appears to be responsible for the greater susceptibility of mlo resistant barley cultivars to facultative parasites such as M. grisea (blast) and Cochliobolus sativus (spot blotch) (Kumar et al., 2001). Also, mlo has been shown to reduce yield of barley lines by about 5% in the absence of disease. A locus in wheat that may be similar to the mlo locus in barley conditions leaf tip necrosis in adult plants. Partial resistance to leaf rust caused by P. triticina (designated Lr34) and to stripe rust caused by P. striiformia (designated Yr18) may actually be due to pleiotropic eVects of the same gene (Johnson, 2000; Spielmeyer et al., 2005). Both the leaf rust and stem rust resistance have proven to be durable and are apparently non‐race‐specific. Others reported that the same locus, or tightly linked loci, provide partial resistance to powdery mildew (B. graminis f. sp. tritici), stem rust (P. graminis f. sp. tritici), and the barley yellow dwarf virus (Kerber and Aung, 1999; Spielmeyer et al., 2005). Lr34 also permits expression of three or more recessives gene for stem rust resistance in the presence of a suppressor gene on chromosome 7D, which normally inhibits their activity (Kerber and Aung, 1999). Neither Lr34 nor Yr18 by themselves provide adequate resistance to rust, but they have been combined with other genes for partial resistance to achieve enhanced
342
D. D. STUTHMAN ET AL.
resistance (Singh et al., 2005). In another possibly related case, Hulbert et al. (2001) reported that recombination within the complex Rp1 locus for resistance to Puccinia sorghi (common maize rust) sometimes resulted in lesion mimic mutants that confer a spontaneous necrotic spotting phenotype in leaves of seedling and adult maize plants. The recombinant lesion mimics also conferred a high level of resistance to all known races of P. sorghi in a manner that may be analogous to the durable resistance of mlo to powdery mildew in barley. In general, resistance to virus diseases in plants is more durable than resistance to diseases caused by biotrophic fungi. More than 200 genes for resistance to virus diseases have been reported in plants, and more than 80% of the cases of resistance to viruses are monogenic (Kang et al., 2005). Several resistance genes are eVective against multiple viruses. For example, the I gene in Phaseolus vulgaris appears to control resistance to 10 diVerent related potyviruses. Many of the forms of monogenic resistance to viruses have remained eVective for several decades (Kang et al. (2005). The mechanisms that determine durability of resistance to viruses remain largely unknown, but it appears that recessive genes for resistance to viruses are more durable than dominant resistance genes (Frasier, 1990).
E. DURABILITY OF POLYGENIC RESISTANCE Durable polygenic resistance to disease is almost ubiquitous among plants. Simmonds (1991) listed 25 crops reported with polygenic resistance to airborne fungal diseases and 17 crops with polygenic resistance to 25 soilborne fungal diseases. There are many possible explanations for the durability of polygenic disease resistance in plants. Resistance to diseases caused by unspecialized, necrotrophic pathogens such as Pythium root rot of sugarcane, Pythium seedling blight of maize, and Stenocarpella and Gibberella stalk rots of maize results from morphological and developmental changes in the host (Bruehl, 1983). Apparently, it is beyond the capacity of the unspecialized pathogens to adapt and overcome those host changes. The same could be said for polygenic resistance in tobacco to root rot caused by Thielaviopsis basicola, which has been used since 1910 with no evidence of pathogen adaptation (Walker, 1959). It seems equally unlikely that a highly specialized pathogen such as P. graminis could overcome the partial resistance in wheat provided by broad bands of sclerenchyma separating isolated strands of susceptible collenchyma bundles in some wheat cultivars (Hart, 1931). Another explanation for the durability of polygenic resistance is that each of the resistance genes contributes only a small additive eVect to the overall resistance, so any virulence gene that might overcome that eVect will have
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
343
only a small selective advantage in the pathogen (Parlevliet, 1995). This should slow down adaptation of the pathogen by requiring multiple mutations to virulence, each of which would increase slowly by selection. Johnson (2000) pointed out the inadequacy of this explanation. If individual resistance genes in the host caused so little reduction in fitness of pathogen genotypes lacking corresponding virulence genes to overcome the resistance, there would be very little incentive for plant breeders to use those resistance genes. Breeders will certainly prefer resistance controlled by relatively few (say three to five) genes that each condition a substantial increase in the level of resistance. A more satisfactory explanation for durability of polygenic resistance with physiological mechanisms against specialized biotrophic pathogens would be that even if specific virulence genes exist in the pathogen against these resistance genes, the fitness cost to the pathogen would limit adaptation. This explanation seems to fit the results of Kolmer and Leonard (1986) who showed that it was possible in the laboratory by recurrent selection to gradually increase the ability of isolates of Cochliobolus heterostrophus to cause disease in leaves of an inbred maize line with polygenic resistance. Some of the increase was due to a general increase in pathogenic ability that aVected all maize lines tested, but a part of the increased pathogenicity was specific to the maize line on which the recurrent selection for increased lesion size was performed. Despite the fact the pathogen was able to accumulate genes for increased aggressiveness under experimental conditions, there was no evidence of erosion of the same resistance under field conditions (Leonard, 1993). It must be emphasized that not all polygenic resistance is beyond the capacity of the pathogen for adaptation. Caldwell (1968) reported a gradual loss of resistance to leaf rust in wheat cultivars derived from cv. Chinese Spring in the United States. According to Caldwell, the resistance was of the hypersensitive type despite being polygenic and expressed only in adult plants. There appeared to be at least four genes conditioning the resistance. The leaf rust population overcame the resistance over a period of 5 years of extensive growth of cultivars derived from cv. Chinese Spring.
IV.
EXAMPLES OF EFFECTIVE POLYGENIC RESISTANCE
Plant breeders have had diVering degrees of success in producing commercial cultivars with good levels of durable resistance to the most important diseases that aVect those crops. At least some of the diVerences can be attributed to the types of crops involved. For that reason, we have included examples
344
D. D. STUTHMAN ET AL.
from maize (cross‐pollinated), wheat and barley (self‐pollinated), and potato (cross‐pollinated, clonally propagated) to illustrate how the diVerences in crop types may aVect the success of breeding for durable resistance.
A.
MAIZE
Before the development of hybrid cultivars, maize was grown as genetically heterogeneous, open‐pollinated varieties. Even though relatively little research was done on resistance to maize diseases, gradual progress was made in improving resistance in the open‐pollinated varieties by selecting seed from the most healthy and most productive plants for the next year’s crop (Ullstrup, 1978). Maize was generally considered a healthy crop, but several disease problems emerged with the development of hybrid maize cultivars and more intensive cultivation. The two most important diseases requiring resistance breeding programs for maize in the United States were the stalk rot complex and northern leaf blight. Stalk rot had long been a chronic problem that typically occurs at the onset of maturity when unspecialized fungal pathogens such as Stenocarpella maydis, Gibberella zea, or Fusarium moniliforme begin to invade senescing pith tissue inside maize stalks. Rotting of the internal stalk tissues leads to early lodging and inability to recover much of the grain by mechanical harvesting. Breeding eVorts have been successful in developing high‐yielding hybrids that are highly resistant to stalk breakage. A variety of breeding schemes were used including mass selection and family selection schemes, including half‐sib, full‐sib, and S1 line methods involving recurrent selection (Renfro, 1985). With stalk rot, it is not possible to score the plants for resistance at pollination time, so family structured, replicated progeny tests were more eVective approaches than mass selection. The nonspecific resistance to stalk rots has not been analyzed in great detail genetically, but it has been shown to be inherited in a quantitative manner. At least nine chromosome arms in maize have been reported to carry genes for stalk rot resistance. Stalk rot resistance in maize lines has remained stable. Continued selection is necessary, however, to maintain high frequencies of stalk rot resistance genes in maize breeding programs. Northern leaf blight caused by Setosphaeria turcica (anomorph Exserohilum turcicum) has been one of the most destructive diseases in maize. In the 1940s and 1950s, during the rapid expansion of hybrid cultivars in maize production, yield losses of up to 50% were reported in severely aVected fields in the mid‐ western United States (Leonard, 1993). Northern leaf blight also caused serious losses in maize in tropical regions at elevations of 900–1600 m (Welz and Geiger, 2000). Both monogenic and polygenic resistance to northern leaf blight has been identified, but maize breeders have been reluctant to use the
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
345
monogenic resistance, because of its race specificity (Leonard, 1993). Polygenic resistance is characterized by fewer and usually smaller lesions as well as an extended latent period between inoculation and the appearance of sporulating lesions than is typical for susceptible maize lines (Ullstrup, 1970). Jenkins et al. (1954) initiated a program of recurrent selection for quantitative resistance to northern leaf blight in 1945 starting with nine crosses, each involving an inbred line with good resistance crossed with a susceptible inbred line currently being used in commercial hybrid production. Populations for selection included a maximum of 250 plants that were artificially inoculated multiple times with S. turcica during the growing season. In each selection generation, pollen from the 10 most resistant plants was collected, mixed, and the pollen mixture was used to pollinate ears of the same 10 plants. Seeds from the hand‐ pollinated plants were mixed in equal proportions for producing the next generation. Jenkins et al. (1954) concluded that two or three generations of recurrent selection were suYcient to produce a population with a high enough frequency of resistant plants to permit selection among them for desired agronomic traits. There has been no evidence of erosion of quantitative resistance to northern leaf blight in normal agricultural use since the initial progress in recurrent selection (Hughes and Hooker, 1971; Welz and Geiger, 2000). Ceballos et al. (1991) reported very rapid resistance gains of 19% per cycle of full‐sib S1 recurrent selection for resistance to northern leaf blight over four cycles in eight subtropical populations. Early research indicated the presence of genes for partial resistance to northern leaf blight on 12 of the 20 chromosome arms of maize (Welz and Geiger, 2000). Among various resistant maize lines, resistance was found most consistently on chromosome arms 3L, 5L, and 7S. Welz and Geiger (2000) identified quantitative trait loci (QTLs) for northern leaf blight resistance in three mapping populations. Population A represented US Corn Belt germplasm with a moderate level of resistance, population B consisted of F3 families from a cross between a US Corn Belt derived European line and a highly resistant tropical African line, and population C was derived from a cross between two moderately resistant early maturing European lines. Twelve QTLs were found that explained 50–70% of the phenotypic variance for northern leaf blight resistance and 60–80% of the genotypic variance. Gene action was additive to partly dominant. In each population, gene eVects of the QTLs were of similar magnitude, and all three populations had QTLs in identical locations on chromosomes 3, 5, and 8. Several disease‐mimic‐ mutations, resistance gene analogues, and genes encoding pathogenesis‐ related proteins also mapped to regions with northern leaf blight resistance QTLs (Welz and Geiger, 2000). Common maize rust caused by P. sorghi occurs nearly everywhere in the world where maize is grown. Although P. sorghi has the potential to develop in epidemic proportions, it has generally not been considered an economic
346
D. D. STUTHMAN ET AL.
problem for field corn in the United States (Hooker, 1969; Ullstrup, 1978). Although most maize plants are susceptible to common rust in the seedling stage, there is suYcient polygenic resistance to protect the crop in commercial production even though some common rust is present nearly every year. Oxalis species, which are the alternate hosts for P. sorghi, become heavily rusted each year in Mexico where the fungus overwinters. Common rust of maize spreads north through the eastern United States every year much as wheat stem rust does. In contrast to wheat stem rust, no severe epidemics of common maize rust have occurred on field corn. Hooker (1969) reported that heritability for rust resistance estimates averaged above 80% for 64 maize crosses studied, and he indicated that it is obvious that a high degree of this type of resistance has been fixed in many inbred maize lines. Kim and Brewbaker (1977) obtained similar estimates. Maize breeders in the United States generally do not maintain significant testing programs for common rust resistance in field corn. Apparently, the practice of discarding any lines with unusually high amounts of rust development in field plots has been suYcient to maintain a high level of polygenic resistance. On sweet corn, however, common rust has become a major concern in many regions of the United States (Randle et al., 1984). The greater susceptibility of sweet corn to common rust undoubtedly reflects the much more restricted genetic diversity in the breeding populations of sweet corn than in those of field corn. Randle et al. (1984) showed that levels of polygenic resistance in sweet corn breeding populations could be increased rapidly by recurrent selection. They obtained realized heritability estimates as high as 45% in two sweet corn composites.
B. WHEAT The history of breeding for resistance to rust diseases of wheat, a self‐ pollinated crop, diVers dramatically from that for resistance to rust diseases in maize. In maize, genes for hypersensitive, race‐specific resistance to rust were largely ignored by breeders who relied instead on recurrent selection to improve levels of quantitative resistance, which proved to be durable. Wheat breeders relied primarily on monogenic resistance especially for stem rust and leaf rust. For stem rust in North America, Europe, and Australia, judicious use of combinations of major genes for race‐specific resistance combined with extensive surveys of virulence in the pathogen population has been quite successful in largely eliminating disease losses (Leonard, 2001). Much of the success in controlling stem rust in the United States has been due to eradication of the alternate host for P. graminis, the limited range of overwinter survival of P. graminis in the absence of barberry, and the use of unique combinations of race‐specific resistance genes in the spring wheat
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
347
region of the northern plains states but not in winter wheat cultivars in the overwintering regions along the Gulf Coast. P. triticina (wheat leaf rust) is able to survive in infected winter wheat plants through much greater extremes of cold weather than P. graminis. It also survives well in late summers on volunteer wheat seedlings between wheat crops. Consequently, pathogen population sizes remain relatively high and virulent mutants tend to persist and increase when selected by the use of race‐ specific resistance in prevailing wheat cultivars. Thus, race‐specific resistance has been much less eVective in the control of leaf rust than with stem rust of wheat. Losses of 5–10% or more continue to occur periodically in the leading winter wheat state of Kansas in the central Great Plains (Long, 2006). Ironically, emphasis on selection for maximal presence of green leaves in field plots of winter wheat breeding programs, such as the very important Kansas breeding program, may have acted against the possible inclusion of Lr34, a gene for durable partial resistance, in the winter wheat cultivars of the United States. For one thing, attempting to eliminate leaf rust completely in breeding lines promoted overreliance on new genes for race‐specific, hypersensitive resistance. For another, selection against evidence of any leaf necrosis in winter wheat lines must have resulted in elimination of lines with Lr34 and, perhaps, other resistance genes that may also be associated with leaf tip necrosis as a pleiotrophic eVect. The CIMMYT wheat breeding program has emphasized selection for slow rusting resistance for more than 25 years with the goal of achieving eVective levels of durable resistance to leaf and stripe rust (Singh et al., 2005). In studies at CIMMYT, wheat cultivars with Lr34 and three or four additional genes for slow rusting have shown stable resistance that is highly eVective. Final leaf rust severities on those cultivars were less than 5% infection even under high leaf rust pressure (Singh et al., 2005). Further studies indicated that there are at least 10–12 diVerent slow rusting genes in the CIMMYT germplasm. One of these genes, Lr46, was identified in cv. Pavon 76 and was shown to be either closely linked or identical to Yr29, which confers moderate resistance to stripe rust in adult plants. In addition, several QTLs contributed increased levels of resistance to both leaf rust and stripe rust (Singh et al., 2005). Schnurbusch et al. (2004) found eight resistance QTLs in a Swiss winter wheat cultivar that has shown a high level of durable resistance to leaf rust since 1986. One QTL with major eVects in the Swiss cultivar may be Lr34. Another QTL with major eVects was not associated with leaf tip necrosis. Two minor QTLs for resistance were found to be associated with leaf tip necrosis that was distinct from the type of leaf tip necrosis characteristic of Lr34. Breeding for durable resistance to wheat stripe rust (Puccinia striiformis) has a longer history than breeding for durable resistance to leaf rust. In the United States, stripe rust was not considered an important disease until
348
D. D. STUTHMAN ET AL.
the 1960s when major epidemics occurred in the Pacific Northwest (Line, 2002). Fortunately, Vogel, who developed the cv. Gaines, the first commercial semidwarf wheat cultivar, had included parental lines with moderate levels of adult plant resistance to stripe rust in the breeding program. Vogel’s breeding nursery was exposed to heavy stripe rust pressure, and he was careful to discard all highly susceptible lines during selection. Consequently, cv. Gaines had more eVective resistance to stripe rust than any other cultivar grown in the Pacific Northwest at that time. Gaines and the improved cultivar Nugaines were the predominant wheat cultivars grown in the Pacific Northwest from the mid‐1960s to early 1980s. During that time there was no significant loss in eVectiveness of their stripe rust resistance (Line, 2002). That resistance, which is now referred to as HTAP resistance, is conditioned by several genes with additive eVects that are expressed in adult plants at temperatures typical for summers in the Pacific Northwest. Since the 1960s, wheat breeders in the Pacific Northwest have continued to select for improved HTAP resistance so that now more than 90% of the wheat cultivars in that region have this type of stripe rust resistance. Most cultivars have a combination of at least five genes for HTAP resistance (Line, 2002), whereas cultivars Gaines and Nugaines probably have only two or three genes for resistance (Johnson, 1992). Wheat breeders in the United Kingdom and in Europe experienced a number of failures of race‐specific resistance to stripe rust from the 1950s to the 1990s (Johnson, 1992). This was true even for genes Yr11, Yr13, and Yr14 for partial resistance expressed in adult plants but not in seedlings. This evidence, that apparently slow rusting resistance may be race‐specific, led Johnson to propose that the only assurance of durability in resistance to stripe rust is actual performance of the resistance over many years of widespread use. For example, the cultivar Capelle Desprez occupied up to 80% of the United Kingdom wheat area for more than 10 years but showed no reduction in its level of resistance to stripe rust. Johnson (1992) reported that the American wheat cultivars Gaines and Nugaines were a little more susceptible in the United Kingdom than in the Pacific Northwest of the United States, probably because of lower summer temperatures in the United Kingdom. Johnson considered the level of resistance of Capelle Desprez to be minimally adequate under United Kingdom conditions. Even with that level of resistance, it has been necessary to apply fungicide to protect from stripe rust on some cultivars in the United Kingdom. According to Johnson (1992), the best approach in breeding for durable resistance to stripe rust is to start with parental cultivars that already have a good record of durability of their resistance. In addition, individual genes with a record of durability such as Yr18, which is either closely linked to or the same as Lr34, may be transferred to new cultivars by backcrossing using leaf tip necrosis and leaf rust resistance as markers for the presence of Yr18.
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
349
Mallard et al. (2005) identified two QTLs for race‐specific resistance and four for apparently nonspecific adult plant resistance in the French wheat cultivar Camp Re´my, which has shown durable resistance to stripe rust since 1980. Two of the QTLs were probably inherited from Capelle Desprez. A study of wheat cultivars grown in the Netherlands showed that cultivars released before 1930 had durable resistance to stripe rust, but most cultivars released after 1930 had nondurable resistance (van Dijk et al., 1988). This was attributed to a change in the methods of selecting for resistance. Before 1930, selection was done under moderate disease pressure, which eliminated the most susceptible genotypes and selected those with moderate levels of partial resistance. In later years, selection was done with artificial selection and high disease pressure that tended to overwhelm genotypes with only moderate resistance. Under those conditions, only the genotypes with hypersensitive race‐specific were selected and genotypes with partial resistance tended to be discarded.
C. BARLEY It has been shown that good levels of partial resistance to barley leaf rust and powdery mildew can be obtained by combining genes with additive eVects. The details are similar to what has been shown in wheat for resistance to leaf rust and stripe rust (Qi et al., 1998). The most striking case of durable resistance to powdery mildew in barley, however, is due primarily to a single gene mlo, although two other genes, Ror1 and Ror2, are required for full expression of the mlo resistance (Lyngkjær et al., 2000). Although sporadic, transient outbreaks of powdery mildew occur occasionally in cultivars with mlo, its resistance is generally highly eVective. Powdery mildew outbreaks on mlo barley are generally associated with unusual conditions such as reduced eVectiveness of mlo during recovery of plants from water stress or from cold stress. Since 1979, barley cultivars with mlo resistance have been widely grown in Europe, increasing to 70% of the spring barley area in the United Kingdom and Germany by 1994 (Lyngkjær et al., 2000). In spite of such extensive use, mlo resistance has remained highly eVective in Europe. It has been recommended that mlo should not be used in winter barley cultivars because that could provide a green bridge for survival of virulent mutants if they should occur in spring barley. Resistance conditioned by mlo appears to operate through papilla formation and other cell wall modifications at the point of attempted penetration of leaf epidermal cell walls by the pathogen (Lyngkjær et al., 2000). In plants with mlo, papillae are larger and form more quickly than in plants with the wild‐type allele Mlo. Deposition of callose in the papillae is important in limiting successful penetration of leaves of plants with mlo resistance.
350
D. D. STUTHMAN ET AL.
In addition, early deposition and cross‐linking of phenolic compounds in papillae and surrounding areas of the cell wall at the point of pathogen attack may also be involved in mlo resistance. Compaction of papillae by cross‐ linking occurs at least 2 h earlier in mlo plants than in susceptible plants. Failure of pathogen spore germ tubes to penetrate the leaf epidermis and produce a haustorium within the epidermal cell stops the infection process. It is thought that the wild‐type Mlo gene codes for a membrane protein that regulates one or more of the early cell defense responses to pathogen attack. In mlo plants, this regulatory protein is missing, and the cell defenses are activated earlier and more strongly than in susceptible barley plants. Barley plants with mlo tend to produce necrotic fleck even in the absence of pathogen attack, which suggests that wild‐type Mlo may be involved in modulating a cell death response (Lyngkjær et al., 2000). Although mlo resistance has remained eVective during widespread use in the field in Europe, there is some concern about its continued durability. Two isolates of B. graminis with partial virulence to mlo barley have been identified (Lyngkjær et al., 2000). One occurred naturally in the field in Japan in the 1950s. The other resulted from a laboratory program of recurrent selection for increased numbers of powdery mildew colonies on barley seedling leaves. The most virulent isolate selected after 37 successive conidial generations in the laboratory produced between 10% and 60% as many colonies on mlo barley as on susceptible barley. The increased virulence in the selected isolate appeared to be due to the combined eVects of one gene with major eVects and two or three minor genes.
D. POTATO No plant disease has impacted human history as profoundly as potato late blight caused by the oomycete Phytophthora infestans, which burst into prominence as the cause of the Great Irish Potato Famine of the 1840s. It is not surprising that potato late blight was among the first plant diseases for which conscious, systematic breeding approaches were directed toward developing resistant cultivars. The breeding eVorts of James Torbitt of Belfast came to the attention of Charles Darwin (1888), who described them in a letter to T. H. Farrer in 1878, ‘‘Mr. Torbitt’s plan of overcoming the potato‐disease seems to me by far the best which has ever been suggested. It consists, as you know from his printed letter of rearing a vast number of seedlings from cross‐fertilized parents, exposing them to infection, ruthlessly destroying all that suVer, saving those which resist best, and repeating the process in successive seminal generations.’’ Darwin went on to say ‘‘Considering the whole subject, it appears to me that it would be a national misfortune if the cross‐fertilized seeds in Mr. Torbitt’s possession produced by
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
351
parents which have already shown some power of resisting the disease, are not utilized by the Government or some public body, and the process of selection continued during several more generations.’’ Although Darwin was frustrated in his eVorts to obtain government support for the work, Torbitt persisted and succeeded in developing ‘‘varieties possessing well‐marked powers of resisting disease.’’ Later Torbitt noted that unfortunately ‘‘this immunity was not permanent and after some years the varieties became liable to attacks of the fungus.’’ In retrospect, it seems likely that the loss of resistance noted by Torbitt was not due to any flaw in his breeding approach. It is now known that accumulation of virus infections in tuber‐propagated potato cultivars can cause loss of resistance to late blight (Thurston, 1971; Turkensteen, 1993). Torbitt’s recurrent selection methods were repeated by other potato breeders in Europe, and gradually the highly susceptible cultivars of the 1840s were replaced by less susceptible cultivars. By the end of the 1800s, cultivars that were moderately resistant to late blight were known in Europe (Turkensteen, 1993). The resistance tended to be stable. For example, the cultivar Champion was widely grown in Ireland from 1877 through 1926, when it still occupied 22% of the potato acreage. Freed of virus infection in 1928, Champion still retained its original level of resistance to late blight (Thurston, 1971). Durable resistance to late blight, however, was not carried over into modern potato cultivars because of two shifts in potato production. The first was the introduction of eVective fungicides, which made it possible to grow more susceptible cultivars (Turkensteen, 1993). The second was the discovery of major genes for race‐specific resistance that essentially eliminated infection by avirulent races. For several decades nearly all potato breeders dropped their work on nonspecific quantitative resistance and concentrated on improving yield and quality in cultivars protected by fungicides and race‐specific resistance (Thurston, 1971). It soon became obvious that race‐specific major genes for resistance cannot provide consistent protection against late blight in potato. Cultivars with single genes for race‐specific resistance lasted no more than 5–10 years in the Netherlands (Turkensteen, 1993). Since 1958, most breeding for late blight resistance in Europe and North America had switched from race‐specific resistance to quantitative nonspecific resistance (Black, 1970; Hawkes, 1978). There has been a concerted eVort to remove major genes for race‐specific resistance from potato breeding populations because the presence of major genes for resistance interferes with the ability to select for polygenic resistance (Turkensteen, 1993). Nevertheless, a survey of potato cultivars in commercial production at the time suggests that resistance to late blight has not received adequate attention by potato breeders and producers (Westie, 1991). Of the 23 potato cultivars recommended to growers in the
D. D. STUTHMAN ET AL.
352
United Kingdom, only 6 scored moderately resistant to foliage blight and 8 scored moderately resistant to tuber blight. Potato cultivars in production tend to have a slow rate of turnover because of consumer preference for familiar named cultivars and because of the slow rate of increase of new potato clones (Turkensteen, 1993). In 1994, more than 60% of the area planted to potatoes in the United States was devoted to just three cultivars, Russet Burbank, Russet Norkotah, and Shepody; resistance to late blight was not regarded as a strong determinant in the choice of cultivars to grow (Inglis et al., 1996). The appearance of highly aggressive P. infestans strains resistant to the systemic fungicide metalaxyl rapidly changed that perspective (Fry and Goodwin, 1997), and the importance of polygenic resistance is again increasingly being recognized. With increased emphasis on the importance of polygenic resistance, there has been interest in changing selection programs. In traditional breeding, selection in the early generations was mainly against low‐yielding progeny. In intermediate generations, lines were selected for yield and specific gravity, processing ability, internal quality, and external appearance of tubers as well as for disease resistance. Lines were often not selected for late blight resistance until the fifth or sixth generation when up to 99% of the genetic variation had been eliminated by selection for other traits (Posch, 2003). It is important to avoid contaminating tubers with late blight infection during early generations of selection so that the pathogen is not perpetuated in tubers used for other tests in subsequent generations (Plaisted et al., 1984). It is possible, however, to produce a large number of clones from each first generation seedling so that some plants could be sacrificed for resistance screening and others can be kept disease free and used to continue selection for other traits, but that may require too great an expenditure of resources and record keeping to be practical. The identification of QTLs for partial resistance to late blight in several potato breeding populations should allow for marker‐assisted breeding in early generations of selection without the risk of infecting tubers to be used in later cycles of selection. Oberhagemann et al. (1999) identified 12 segments on 10 chromosomes that contributed consistently to foliar resistance and one major QTL for tuber resistance to late blight.
V.
CLASSICAL BREEDING APPROACHES A. RECURRENT SELECTION
As seen with the examples of disease resistance in maize, recurrent selection is highly eVective when applied to normally outcrossing species for increasing levels of quantitatively inherited resistance. The obvious advantage of recurrent
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
353
selection is that it allows the breeder to increase the frequencies of desirable alleles for partial resistance in the breeding population before beginning to evaluate individual lines for their potential as commercial cultivars. If a trait such as polygenic resistance is controlled by five or more genes, only a very low proportion of the progeny from a cross between a susceptible, agronomically superior cultivar and a disease resistance donor parent will have the desired number of resistance genes. Several rounds of recurrent selection for both agronomic traits and the desired disease resistance, however, will increase the chances of extracting lines from the population that combine both good agronomic traits and good levels of quantitative resistance. As Eberhart (1990) observed ‘‘population improvement is the foundation of a maize breeding program seeking to maximize long‐term genetic gain per year.’’ He recommended choosing two maize populations for selection and practicing reciprocal recurrent selection using several tester inbred lines from each population to improve the variety cross performance and increase heterosis. Thus, selection for specific hybrid combinations is initiated in the population improvement cycle. Experimental S2 lines from each population are crossed to inbred tester lines from the reciprocal population and performance of the resulting F1 hybrids is evaluated. Then the selected lines with superior performance are recombined in all possible combinations to form the next cycle of the population. The tester lines will change with advancing cycles of selection as improved inbred lines are developed (Eberhart, 1990). Selection intensity can be increased by increasing the number of experimental lines evaluated in testcrosses or by decreasing the number of lines selected for recombining to form the next cycle of the population. According to Eberhart (1990), 250–400 S2 lines should be evaluated to select 16–20 lines for recombination for the next cycle. This will achieve a selection intensity of 4–8%. In the early stages, it is important to select a limited number of the most important traits required for superior yield and resistance to the primary pest and stress problems that exist in farmers’ fields. Attempting to select for too many traits simultaneously will result in slow gains from selection for any individual trait. Eberhart (1990) recommended multistage selection to achieve multitrait improvements. For example, S0 plants that will be selfed in each population can be screened to eliminate undesirable plants based on highly heritable traits such as maturity dates and resistance to certain diseases. When possible these selections should be done before anthesis to reduce the numbers of plants to be self‐pollinated. In the following season, S1 plants can be subjected to selection for less heritable traits such as insect resistance and resistance to lodging. Selection among testcrosses should be based primarily on yield with some additional selection for resistance to root rot and stalk rot. Recurrent selection schemes for population improvement are naturally suited to crops that are predominantly outcrossers, but a number of recurrent selection schemes have been tested with self‐pollinated crops. The use of
354
D. D. STUTHMAN ET AL.
male sterility traits was proposed to facilitate recurrent selection in barley (Gilmore, 1964), sorghum (Doggett and Eberhart, 1968), and soybean (Brim and Stuber, 1973). Matzinger and Wernsman (1968) showed that consistent gains in leaf yield can be obtained by repeated mass selection during cycles of random matings among selected lines in a heterogeneous synthetic variety of tobacco, which is normally self‐pollinated. Jensen (1970) proposed a set of diallel selective mating procedures with recurrent selection for population improvement in populations of small grains to serve as a supplement to conventional pedigree breeding programs. Diaz‐Lago et al. (2002) demonstrated that such a program, with early generation selection for partial resistance to crown rust in oat, led to a total of 42% gain in level of resistance after four cycles of selection for crown rust resistance in an oat population that previously had undergone eight cycles of recurrent selection for grain yield. They found, however, that concurrent selection for early flowering date would be necessary because some of the additional resistance to crown rust was associated with late maturity.
B. PEDIGREE BREEDING Pedigree breeding is widely practiced with self‐pollinated crops because it permits breeders the most opportunities to exercise their skills in selection (Allard, 1960). The main disadvantage is that it limits the amount of material a single breeder can handle. Pedigree breeding usually involves crosses between one parent chosen for its proven agronomic performance and another parent chosen because it complements specific weaknesses (e.g., lack of disease resistance) in the first parent. Superior types are selected in successive selfed generations, and records are maintained of all parent–progeny relationships. The information on pedigrees is useful in avoiding selection of closely related individuals in the same lineage whose probable worth is nearly identical. Selection may begin in the F2 generation. Through the F3 and F4 generations, selection is generally practiced on the best plants in the best families because of the level of heterozygosity that persists in these generations. By the F5 and F6 generations, emphasis shifts to selection among families, often planted as single rows or as replicated hill plots in breeding nurseries. Ultimately, lines that are candidates for new cultivar release are tested in replicated field plots at several locations (Allard, 1960). Pedigree breeding is also used with open‐pollinated crops such as maize for recycling lines that have known strengths and weaknesses for specific traits (Agrawal, 1998). Pedigree selection may start with progenies developed in open‐pollinated varieties, germplasm composites, synthetics, backcross populations, as well as F2 populations. For maize, the object is to select inbred lines with superior combining ability in the production of high‐yielding F1
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
355
hybrids capable of withstanding disease and other stresses characteristic of the area in which the hybrids are to be grown. In crops for which quality standards for cultivar acceptance are exacting, backcrossing is commonly used to transfer disease resistance from agronomically inferior donors to high‐yielding, high‐quality cultivars that have been developed through standard pedigree breeding approaches. The CIMMYT wheat breeding program includes a single backcross approach to incorporate additive genes for durable resistance into high‐yielding cultivars that have shown unacceptable levels of susceptibility to leaf rust and stripe rust (Singh et al., 2005). Each cultivar is crossed with a group of about 10 resistance donors that have been characterized for partial resistance to either leaf rust or stripe rust. Each donor has expressed high levels of rust resistance in field tests and may have up to four or five genes for partial resistance. In the case of Lr34/Yr28, the resistance gene may have a proven record of durability in extensive use in earlier cultivars that have been widely grown. In other cases, such as with Lr46, preliminary evidence indicates a good probability that the resistance will be durable. And in still other cases, little is known about the resistance genes except that they condition partial resistance with additive eVects. From each resistant donor susceptible cultivar cross, 20 spikes from F1 plants are backcrossed to the agronomically superior cultivar to obtain 400–500 seeds. Selection under high rust pressure is practiced from the BC1 generation onward for resistance and other agronomic traits. Plants with low to moderate disease severity in early generations and with low disease severity in later generations are retained. Some of the derived lines from this approach not only showed high levels of resistance to leaf rust or stripe rust but also had 5–15% greater yield potential than the original cultivar (Singh et al., 2005). Of course, a second round of crosses among derived lines with resistance from diVerent donors could provide even greater levels of resistance if more resistance is needed in some environments. To improve the likelihood that the resistance obtained in methods similar to the CIMMYT single backcross approach, Johnson (1978) urged that the resistance donor parents be chosen from lists of older cultivars that have established records of durability of their partial resistance in the past. Johnson (1988) listed 22 European and American cultivars of wheat that have stripe rust resistance that is probably durable based on records of disease severity on the cultivars over many years of commercial use.
C. PERENNIAL SPECIES Both monogenic, race‐specific resistance and polygenic, presumably nonspecific, resistance to diseases occur in species of trees [e.g., resistance to scab in apple (Williams and Kuc, 1969) and resistance to blister rust in sugar pine
356
D. D. STUTHMAN ET AL.
(Carson and Carson, 1989)]. Nevertheless, polygenic resistance has been favored in tree breeding because of its expected greater durability and because of the need for long‐lasting resistance in long‐lived perennial species. In the case of apple scab, historical records of performance of old apple cultivars have enabled breeders to identify cultivars that showed good quantitative resistance in the past. These resistant cultivars served as donors for quantitative resistance in more recent breeding programs (Kellerhals and Farrer, 1994). In forest species bred for lumber or pulp production, the aim has been to produce genetically improved populations rather than pure varieties or clones (Carson and Carson, 1989). Forest tree breeding programs typically produce commercial seed lots from open‐pollinated seed orchards that contain many selected parents (Carson and Carson, 1989). In pines, it is advisable to have at least 15 parent trees in the seed orchard to avoid unacceptable levels of inbreeding. Tree breeders rely heavily on recurrent selection for incremental improvement in successive breeding cycles. Breeders may maintain and select in two types of populations: a breeding population with a broad genetic base and a wood production population with a relatively narrow genetic base (Carson and Carson, 1989). Selection for disease resistance and productivity may be conducted in both populations. In the breeding population, selection intensity is constrained by the need for large population sizes to avoid depleting the genetic diversity of base populations. In wood production populations, selection intensity is less constrained, but selection is limited by the cost of evaluating large numbers of genotypes and the capacity to multiply selected progeny families. For species of pines, selections are commonly made in half‐sib families derived from selected female parents exposed to natural, random pollination within the breeding orchard. Resistance in pine species to diseases such as Dothiostroma needle blight and fusiform rust have been shown to be quantitatively inherited with phenotype determined by the additive eVects of many genes each contributing small additive eVects. Heritability of resistance to fusiform rust ranged from 24% to 30% (Carson and Carson, 1989). Total immunity has not been achieved in breeding for resistance to these diseases, but substantial progress has been made in reducing disease severity in selected populations. So far, there has been no evidence that quantitative resistance has eroded over time in commercial production of selected populations. Because of the long time between planting and harvest of trees for wood production, forest tree breeders emphasize genetically heterogeneous populations in commercial production. Diversity is thought to enhance the durability of resistance in the commercial production populations. Also, the risk of unrecognized vulnerability of any specific progeny family to a new disease or pest problem is buVered in the mixed populations. In recent years, it has been possible to propagate some forest tree species vegetatively on a
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
357
commercial scale. Breeders recognize the risk of planting single clones over large areas where they will grow for several decades before harvest. In some countries (e.g., Germany), diversity is legally mandated. Authorities may insist that each stand of trees for commercial production must have as many 1000 or more clones to maintain genetic diversity.
VI.
MOLECULAR APPROACHES A. MARKER‐ASSISTED SELECTION
The development of multiple types of DNA markers made it possible to identify chromosome regions that contain genes that contribute relatively small incremental eVects in quantitative traits. Once identified, these regions, termed QTLs, can be treated as Mendelian factors of inheritance for breeding purposes (Keller et al., 2000; Lindhout, 2002; Stuber, 1995; Young, 1996). To control genotype environment interactions, QTL mapping is usually done with homozygous populations derived as doubled haploids, recombinant inbred lines, or backcross inbred lines, which allow replicated evaluations of the quantitative trait under consideration. Consequently, most QTL analyses have been performed with only one population, which means that the number of QTLs identified is limited to the numbers of genes that diVer between the two parents from which the population was derived. Furthermore, not all loci contributing to the quantitative trait are identified in QTL analyses. The amount of phenotypic variance explained by all QTLs collectively is generally smaller than the amount that is explained by genotypic variance estimates in calculations of heritability of the trait. Populations must be quite large to reveal QTLs with minor contributions (Keller et al., 2000; Young, 1996). Nevertheless, QTL analyses can be very useful in identifying additive genes that make relatively large contributions to the desired trait. For example, many QTL analyses of quantitative disease resistance have revealed one or two loci that contribute 20–30% or more of the total phenotypic variance for resistance in the population. Incorporating a few genes with eVects of that magnitude into a cultivar could dramatically reduce disease losses. As yet, however, there have been few reports of the application of marker‐assisted selection for disease resistance because inexpensive, high‐throughput genotyping technologies have not developed as quickly as anticipated (Michelmore, 2003). One of the disadvantages associated with marker‐assisted selection is that it imposes rather high costs in resources required for data acquisition and record keeping. In pedigree breeding, decisions must be made among hundreds or thousands of individuals in segregating generations. Most
358
D. D. STUTHMAN ET AL.
judgments by the breeder must be based on quick visual evaluation rather than precise measurements (Allard, 1960). Collecting and labeling samples, submitting them for DNA analyses, and recording presence or absence of QTLs for quantitative resistance among progeny may divert too much of the breeder’s time and resources from the main goals of selecting for high yield, quality, and other necessary agronomic traits for a commercial cultivar. Where the anticipated benefits of introgression of new resistance genes are high, the advanced backcross QTL analysis developed by Tanksley and Nelson (1996) can shorten the time needed to identify and transfer major QTLs to advanced breeding lines. In this method, QTL analysis is delayed until the BC2 or BC3 generations from a cross between an elite breeding line and a resistance donor parent. During these early generations, deleterious traits from the donor parent are selected against. QTL near‐isogenic lines can be developed with enhanced resistance in one or two additional generations and evaluated for potential commercial use. A less intrusive use of QTL analyses in the resistance breeding program would be in the selection of resistance donors. For example, in breeding for durable resistance to leaf rust in wheat, it would be desirable to combine genes for partial resistance from several diVerent donor parents. Because Lr34 has been widely used in wheat, it would not be unusual for several potential donors of good partial resistance to leaf rust to have Lr34 as one of the important contributors of their resistance. Using a DNA marker closely linked to Lr34 will allow breeders to screen for Lr34 in potential parent lines to avoid duplication of the same resistance gene in diVerent parents. As QTL markers are identified for other additive resistance genes, the selection of resistance donors can be made more eYcient. In the single backcross breeding program for leaf and stripe rust resistance at CIMMYT, a group of 10 resistance donors were crossed to the susceptible but agronomically superior parent cultivar, the progeny were backcrossed to the cultivar, and lines were selected for resistance, yield, and so on for successive selfed generations (Singh et al., 2005). Combining QTL analyses with this single backcross program should allow breeders to compare selected progeny from diVerent donor parents to determine which of the selected lines have genetically distinct rust resistance that could then be combined to achieve even more eVective resistance in a second round of crossing and selection.
B. GENETIC TRANSFORMATION Development of techniques to genetically transform plants has opened new possibilities for crop protection. Transgenic crop plants were grown on an estimated 44 million hectares throughout the world in 2000. As much as 36% of sorghum, 16% of cotton, 11% of maize, and 7% of canola grown
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
359
worldwide in 2000 was transgenic. Most of the transgenic cultivars contain herbicide and insect resistance (De Boer, 2003). As yet, commercial releases of cultivars transgenic for resistance to diseases are limited to a relatively few with transgenic resistance to viruses. Virus resistance has been obtained by transforming plants with coat protein genes or other viral genes. The mechanism believed to be responsible for most, if not all, virus resistance in transgenic plants is posttranscriptional gene silencing (Garr and Rushton, 2005; Martelli, 2001). Papaya ring spot has been successfully controlled by the use of transgenic papaya cultivars to suppress spread of the virus, PRSV, in Hawaii and Taiwan after the disease had nearly destroyed the industry. New virus resistant transgenic cultivars of rice, tomato, potato, and stone fruits also show good promise for commercial use (Martelli, 2001). Several approaches are being used to develop transgenic disease resistance in crops (Cornelissen and Melchers, 1993; Garr and Rushton, 2005; Lamb, 1998; Melchers and Stuiver, 2000; Stuiver and Custers, 2001). One of the oldest strategies is to introduce genes for antifungal proteins with promoters that cause overexpression of the genes. Melchers and Stuiver (2000) listed 20 such genes that have produced increased resistance to 17 diVerent pathogenic fungi in seven diVerent crop species. So far, the resistance has been incomplete and has not been confirmed in rigorous field tests. In some cases, it was necessary to transform the host plant with more than one resistance gene to achieve significantly enhanced resistance. For example, tobacco transformed with both a rice chitinase and an alfalfa glucanase were substantially more resistant to the fungus Cercospora nicotianae than plants transformed with either gene alone. A newer strategy involves a resistance pathway‐modulating approach (Stuiver and Custers, 2001). Plant defense pathways against pathogens are characterized by the signaling molecules that regulate the expressions of a whole range of defense proteins. The best known signaling molecule is salicylic acid, which is involved in the systemic acquired resistance pathway in plants. In Arabidopsis, overexpression of the NIM1/NPR gene leads to enhanced resistance against several diVerent pathogens. The disadvantages of this approach are: (1) that constitutive overexpression of the defense pathway tends to reduce yield or plant vigor and (2) that overexpression of one defense pathway may competitively suppress other defense pathways and lead to increased susceptibility to nontarget pathogens. Mutants with constitutive induction of the systemic acquired resistance pathway often develop localized cell death throughout the plant as well as dwarfism (Oldroyd and Staskawicz, 1998). A third approach to genetically engineering plant disease resistance involves induction of a hypersensitive response by transforming a plant that has a race‐specific resistance gene with the specific elicitor gene from an avirulent race of the pathogen (Cornelissen and Melchers, 1993;
360
D. D. STUTHMAN ET AL.
Stuiver and Custers, 2001). This approach requires that the promoter combined with the elicitor gene must give tight regulation of the pathogen‐ inducible promoter. Leakiness of the promoter would produce a lesion‐ mimic‐mutant similar to those described by Hulbert et al. (2001), which undoubtedly would cause yield reduction. Two research groups have succeeded in developing nonleaky transgenic resistant tomato or tobacco plants based on a race‐specific R gene and the pathogen elicitor gene. The transgenic plants did not appear to lack fitness under controlled growth conditions (Stuiver and Custers, 2001). It is not clear whether the resistance will be equally successful under field conditions. In the field, plants are exposed to an abundant community of saprophytic fungi that decompose dead or senescent leaves (Dickinson, 1976). If the pathogen‐inducible promoter also responds to common saprophytes as well as to a wide range of pathogens, the promoter could cause unnecessary hypersensitive responses and, perhaps, render the plants more susceptible to several necrotrophic pathogens (Wubben et al., 1997). Concerns about the use of commercial cultivars transgenic for resistance to fungal and bacterial pathogens are likely to be alleviated to a large extent by current use of transgenic resistance to insects and viruses. Breeders will need to be aware of the potential for pathogens to adapt to overcome the resistance, just as they are in the case of insect resistant cultivars that are transgenic for Bt toxin. Other concerns include possible environmental eVects of horizontal transfer of resistance genes and the potential toxicity or allergenicity of transgenes or the selectable marker genes used in transformation. These concerns have been dealt with successfully in relation to cultivars transgenic for virus resistance, so they should not cause insurmountable diYculties in transgenic resistance to fungi and bacteria (Martelli, 2001). Nevertheless, extensive testing will be necessary with each novel transgene for resistance to ensure that it is as eVective under field conditions as it is under controlled conditions in the laboratory or greenhouse.
VII. SUMMARY Examples of plant diseases included in this chapter were chosen to illustrate several points: (1) durable resistance exists in a variety of forms, often depending on the type of pathogen involved; (2) resistance may be made more or less durable depending on the manner in which it is deployed in the agroecosystem; (3) durable resistance exists in some form in virtually all plant species and against virtually all types of pathogens; and (4) attaining durable resistance to major diseases in important crops is feasible, if not always easy. Indeed, for some diseases such as seedling blights, root rots, and stem (stalk)
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
361
rots caused by unspecialized pathogens with broad host ranges, quantitatively inherited durable resistance may be the only form of resistance available to the breeder. For highly specialized parasites with narrow host ranges, monogenic race‐specific resistance has been a seductive but generally an unreliable form of resistance. It is now abundantly clear that moderate steady gains in resistance are preferable to boom and bust cycles of temporary race‐specific immunity followed by failure due to race shifts in pathogen populations, just as consistent gains in world food supply are preferable to wild fluctuations from year to year and region to region. Durable resistance and sustainable agriculture are mutually supportive concepts (Mundt et al., 2002; Stuthman, 2002). As levels of durable resistance increase in crops, fungicide use will decline because risk‐averse growers will have less incentive to over use fungicides as insurance against crop failure. On the other hand, moderate use of fungicides along with crop rotation, varietal mixtures, and so on, can supplement the impact of partial resistance in the short‐term as well as prolong its durability. Of course, integrating durable resistance into the broad approach of sustainable agriculture entails some costs (Brown, 2002). Initially, breeders must be willing to divert significant resources to selecting for increasing levels of partial resistance at the expense of simply breeding to maximize yield and quality in disease‐free or pesticide‐protected environments. It is to be expected that durable resistance will involve some trade‐oVs in terms of maximum attainable yield. On the other hand, once genes for durable resistance have been accumulated in crop breeding populations, the resistance can be maintained through routine selection to remove breeding lines with the lowest levels of resistance. Perhaps most importantly, durable resistance helps assure greater economic security and food accessibility throughout the world by reducing year‐to‐year and region‐by‐region variation in crop yields.
REFERENCES Agrawal, R. L. (1998). ‘‘Fundamentals of Plant Breeding and Hybrid Seed Production.’’ Science Publishers, Enfield, New Hampshire, USA. Agrios, G. N. (2004). ‘‘Plant Pathology,’’ 5th ed. Elsevier, Academic Press, Amsterdam. Aiyel, A. K. Y. N. (1949). Mixed cropping in India. Indian J. Agri. Sci. 19, 439–453. Alexander, H. M. (1992). Evolution of disease resistance in natural plant populations. In ‘‘Plant Resistance to Herbivores and Pathogens’’ (P. S. Fritz and E. L. Simms, Eds.), pp. 326–344. University Chicago Press, Chicago. Allard, R. W. (1960). ‘‘Principles of Plant Breeding.’’ John Wiley & Sons, New York. Anikster, Y. (2001). Host‐parasite relations in a population of wild emmer in eastern Galilee. Isr. J. Plant Sci. 49, S37–S41. Anikster, Y., Manisterski, J., Long, D. L., and Leonard, K. J. (2005). Leaf rust and stem rust resistance in Triticum dicoccoides populations in Israel. Plant Dis. 89, 55–62.
362
D. D. STUTHMAN ET AL.
Black, W. (1970). The nature and inheritance of field resistance to late blight (Phytophthora infestans) in potatoes. Am. Potato J. 47, 279–288. Borlaug, N. E. (1953). New approach to the breeding of wheat varieties resistant to Puccinia graminis tritici. (Abstr.) Phytopathology 43, 467. Borlaug, N. E. (1966). Basic concepts which influence the choice of methods for use in breeding for disease resistance in cross‐pollinated and self‐pollinated crop plants. In ‘‘Breeding Disease Resistant Trees’’ (H. D. Gerhold, E. J. Schreiner, R. E. McDermott, and J. A. Winieski, Eds.), pp. 327–348. Proceedings of the NATO and NSFU Symposium, Pennsylvania State University, Pergamon Press, New York. Brim, C. A., and Stuber, C. W. (1973). Application of genetic male sterility to recurrent selection schemes in soybean. Crop Sci. 13, 528–530. Brown, J. K. M. (2002). Yield penalties of disease resistance in crops. Curr. Opin. Plant Biol. 5, 339–344. Browning, J. A. (1974). Relevance of knowledge about natural ecosystems to development of pest management programs for agro‐ecosystems. Proc. Am. Phytopathol. Soc. 1, 191–199. Browning, J. A., and Frey, K. (1969). Multiline cultivars as a means of disease control. Annu. Rev. Phytopathol. 7, 355–382. Browning, J. A., Simons, M. D., Frey, K. J., and Murphy, H. C. (1969). Regional deployment for conservation of oat crown rust resistance genes. In ‘‘Disease Consequences of Intensive and Extensive Culture of Field Crops’’ (J. A. Browning, Ed.), pp. 49–56. Iowa Agricultural and Home Economics Experimental StationSpecial Report 64, 56 pp. Bruehl, G. W. (1983). Non‐specific genetic resistance to soilborne fungi. Phytopathology 73, 948–951. Burdon, J. J. (1987). ‘‘Diseases and Plant Population Biology.’’ Cambridge University Press, London. Burdon, J. J., and Chilvers, G. A. (1982). Host density as a factor in plant disease ecology. Annu. Rev. Phytopathol. 20, 143–166. Caldwell, R. M. (1968). Breeding for general and/or specific disease resistance. In ‘‘Proceedings of the Third International Wheat Genetics Symposium’’ (K. W. Finley and K. W. Shepherd, Eds.), pp. 263–272. Plennum Press, New York. Carson, S. D., and Carson, M. J. (1989). Breeding for resistance in forest trees: A quantitative approach. Annu. Rev. Phytopathol. 27, 373–395. Ceballos, H., Deutsch, J. A., and Gutie´rrez, H. (1991). Recurrent selection for resistance to Eserohilum turcicum in eight subtropical maize populations. Crop Sci. 31, 964–971. Cornelissen, B. J. C., and Melchers, L. S. (1993). Strategies for control of fungal diseases with transgenic plants. Plant Physiol. 101, 709–712. Czembor, H. J., and Gacek, E. S. (1996). The use of cultivar and species mixtures to control diseases and for yield improvement in cereals in Poland. In ‘‘Integrated Control of Cereal Mildews and Rusts: Towards Coordination of Research across Europe’’ (E. Limpert, M. R. Finckh, and M. S. Wolfe, Eds.), pp. 177–184. European Commission Directorate‐ General XII, Science Research and Development, Brussels. Darwin, C. R. (1888). ‘‘The Life and Letters of Charles Darwin’’ (F. Darwin, Ed.), pp. 519–522. Appleton, New York. DatnoV, L. E., Rand, R. N., Snyder, G. F., and Jones, D. B. (1991). EVect of calcium silicate on blast and brown spot intensities and yields of rice. Plant Dis. 75, 729–732. De Boer, S. H. (2003). Perspective on genetic engineering of agricultural crops for resistance to disease. Can. J. Plant Pathol. 35, 10–20. Diaz‐Lago, J. E., Stuthman, D. D., and Abadie, T. E. (2002). Recurrent selection for partial resistance to crown rust in oat. Crop Sci. 42, 1475–1482.
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
363
Dickinson, C. H. (1976). Fungi on the aerial surfaces of higher plants. In ‘‘Microbiology of Aerial Plant Surfaces’’ (C. H. Dickinson and T. F. Preece, Eds.), pp. 293–324. Academic Press, London. Dickson, J. G. (1923). Influence of soil temperature and moisture on the development of the seedling blight of wheat and corn caused by Gibberella saubinetii. J. Agric. Res. 23, 837–870. Dinoor, A., and Eshed, N. (1984). The role and importance of pathogens in natural plant communities. Annu. Rev. Phytopathol. 22, 443–466. Dinoor, A., and Eshed, N. (1987). The analysis of host and pathogen populations in natural ecosystems. In ‘‘Populations of Plant Pathogens: Their Dynamics and Genetics’’ (M. S. Wolfe and C. E. Caten, Eds.), pp. 75–90. Blackwell Scientific Publications, Oxford, UK. Dinoor, A., Eshed, N., Ecker, R., Gerechter‐Amitai, Z., Solel, Z., Manisterski, J., and Anikster, Y. (1991). Fungal diseases of wild tetraploid wheat in a natural stand in northern Israel. Isr. J. Bot. 40, 481–500. Doggett, H., and Eberhart, S. A. (1968). Recurrent selection in sorghum. Crop Sci. 8, 119–121. Eberhart, S. A. (1990). A comprehensive breeding system for developing improved maize hybrids. In ‘‘Maize Improvement, Production and Protection in Eastern and Southern Africa’’ (B. Gebrekidan, Ed.). AMREF Printing Department, Nairobi, Kenya. Flor, H. H. (1955). Host‐parasite interaction in flax rust—its genetics and other implications. Phytopathology 45, 680–685. Frasier, R. S. S. (1990). The genetics of resistance to plant viruses. Annu. Rev. Phytopathol. 28, 179–200. Fry, W. E., and Goodwin, S. B. (1997). Re‐emergence of potato and tomato late blight in the United States. Plant Dis. 81, 1349–1357. Garr, S. J., and Rushton, P. J. (2005). Engineering plants with increased disease resistance: What are we going to express? Trends Biotechnol. 23, 275–282. Gilmore, E. C., Jr. (1964). Suggested method of using reciprocal recurrent selection in some naturally self‐pollinated species. Crop Sci. 4, 323–325. Gould, F., and Cohen, M. B. (2000). Sustainable use of genetically modified crops in developing countries. In ‘‘Agricultural Biotechnology and the Poor’’ (G. Persley and M. Lantin, Eds.), pp. 139–146. Proceedings of an International Conference, Washington, DC, USA, 21–22 October 1999. Harlan, J. R. (1976). Disease as a factor in plant evolution. Annu. Rev. Phytopathol. 14, 31–51. Hart, H. (1931). ‘‘Morphologic and Physiologic Studies of Stem‐rust Resistance in Cereals.’’ U. S. Department of Agriculture Technical Bulletin, No. 206. Hawkes, J. G. (1978). Genetic poverty of the potato in Europe. In ‘‘Broadening the Genetic Base of Crops’’ (A. C. Zeven and H. M. van Harten, Eds.), pp. 19–27. Centre for Agricultural Publishing and Documentation, Wagengingen, The Netherlands. Hooker, A. L. (1969). Widely based resistance to rust in corn. In ‘‘Disease Consequences of Intensive and Extensive Culture of Field Crops’’ (J. A. Browning, Ed.), pp. 28–34. Iowa Agricultural Home Economic Experimental Station, Special Report No. 64. Hughes, G. R., and Hooker, A. L. (1971). Gene action conditioning resistance to northern leaf blight in maize. Crop Sci. 11, 180–184. Hulbert, S. H., Webb, C. A., Smith, S. M., and Sun, Q. (2001). Resistance gene complexes: Evolution and utilization. Annu. Rev. Phytopathol. 39, 285–312. Inglis, D. A., Johnson, D. A., Legard, D. E., Fry, W. E., and Hamm, P. B. (1996). Relative resistance of potato clones in response to new and old populations of Phytophthora infestans. Plant Dis. 80, 575–578. Jenkins, M. T., Robert, A. L., and Findley, W. R., Jr. (1954). Recurrent selection as a method for concentrating genes for resistance to Helminthosporium turcicum leaf blight in corn. Agron. J. 46, 89–94.
364
D. D. STUTHMAN ET AL.
Jensen, N. F. (1952). Intra‐varietal diversification in oat breeding. Agron. J. 44, 30–34. Jensen, N. F. (1970). A diallel selective mating system of cereal breeding. Crop Sci. 10, 629–635. Johnson, R. (1978). Practical breeding for durable resistance in rust diseases in self‐pollinated cereals. Euphytica 27, 529–540. Johnson, R. (1988). Durable resistance to yellow (stripe) rust in wheat and its implications in plant breeding. In ‘‘Breeding Strategies for Resistance to the Rusts of Wheat’’ (N. W. Simmonds and S. Rajaram, Eds.), pp. 63–75. CIMMYT, Mexico, DF. Johnson, R. (1992). Past, present and future opportunities in breeding for disease resistance, with examples from wheat. Euphytica 63, 3–22. Johnson, R. (2000). Classical plant breeding for durable resistance to diseases. J. Plant Pathol. 82, 3–7. Jones, F. G. W., Parrott, D. M., and Ross, G. J. S. (1967). The population genetics of the potato cyst nematode, Heterodera rostochiensis: Mathematical models to simulate the eVects of growing eelworm‐resistant potatoes bred from Solanum tuberosum ssp. Andigena. Ann. Appl. Biol. 60, 151–171. Kang, B.‐C., Yeam, I., and Jahn, M. M. (2005). Genetics of plant virus resistance. Annu. Rev. Phytopathol. 43, 581–621. Keller, B., Feuillet, C., and Messmer, M. (2000). Genetics of disease resistance. In ‘‘Mechanisms of Resistance to Plant Diseases’’ (A. J. Slusarenko, R. S. S. Fraser, and L. C. van Loon, Eds.), pp. 101–160. Kluwer Academic Publishers, Dordrecht, The Netherlands. Kellerhals, M., and Farrer, B. (1994). Approaches for breeding apples with durable resistance. Euphytica 77, 31–35. Kerber, E. R., and Aung, T. (1999). Leaf rust resistance gene Lr34 associated with nonsupression of stem rust resistance in the wheat cultivar Canthatch. Phytopathology 89, 518–521. Khush, G. (1995). Modern varieties—their real contribution to food supply and equity. GeoJournal 35(3), 275–284. Kim, S. K., and Brewbaker, J. L. (1977). Inheritance of general resistance in maize to Puccinia sorghi Schw. Crop Sci. 17, 456–461. Kolmer, J. A., and Leonard, K. J. (1986). Genetic selection and adaptation of Cochliobolus heterostrophus to corn with partial resistance. Phytopathology 76, 774–777. Kumar, J., Hu¨ckelhoven, R., Beckhove, U., Nagarajan, S., and Kogel, K.‐H. (2001). A compromised Mlo pathway aVects the response of barley to the necrotrophic fungus Bipolaris sorokiniana (telomorph: Coclilobolus sativus) and its toxins. Phytopathology 91, 127–133. Lamb, C. (1998). Prospects for engineering enhanced durable disease resistance in crops. Pages 183–190, In ‘‘Feeding a World Population of More than Eight Billion people: A Challenge to Science’’ (J. C. Waterlow, D. G. Armstrong, L. Fowden, and R. Riley, Eds.), pp. 183–190. Oxford University Press in association with the Rank Prize Funds, New York. Leach, J. E., Vera Cruz, C. M., Bai, J., and Leung, H. (2001). Pathogen fitness penalty as a predictor of durability of disease resistance genes. Annu. Rev. Phytopathol. 39, 187–224. Leonard, K. J. (1984). Population genetics of gene‐for‐gene interactions between plant host resistance and pathogen virulence. In ‘‘Genetics: New Frontiers, Proceedings of the XV International Congress on Genetics, New Delhi. Vol. IV. Applied Genetics’’ (V. L. Chopra, B. C. Joshi, R. P. Sharma, and H. C. Bansal, Eds.), pp. 131–148. Oxford & IBH Publication Company, New Delhi. Leonard, K. J. (1993). Durable resistance in the pathosystems maize‐northern and southern leaf blights. In ‘‘Durability of Disease Resistance’’ (T. H. Jacobs and J. E. Parlevliet, Eds.), pp. 99–114. Kluwer Academic Publishers, Dordrecht, The Netherlands. Leonard, K. J. (2001). Stem rust—future enemy? In ‘‘Stem Rust of Wheat: From Ancient Enemy to Modern Foe’’ (P. D. Peterson, Ed.), pp. 123–150. APS Press, Eagen, MN, USA.
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
365
Limpert, E., Finckh, M. R., and Wolfe, M. S. (1996). ‘‘Integrated Control of Cereal Mildews and Rusts: Towards Coordination of Research across Europe.’’ European Commission Directorate‐General XII, Science Research and Development, Brussels. Lindhout, P. (2002). The perspectives of polygenic resistance in breeding for durable resistance. Euphytica 124, 217–226. Line, R. F. (2002). Stripe rust of wheat and barley in North America: A retrospective historical review. Annu. Rev. Phytopathol. 40, 75–118. Long, D. L. (2006). Small grain losses due to rust. USDA, ARS, Cereal Disease Laboratory. http://www.ars.usda.gov/Main/docs.htm?docid ¼ 10123 Lyngkjær, M. F., Newton, A. C., Atzema, J. L., and Baker, S. J. (2000). The barley mlo‐gene: An important powdery mildew resistance source. Agronomie 20, 745–756. Mallard, S., Gaudet, D., Aldeia, A., Abelard, C., Besnard, A. L., Sourdille, P., and Dedryver, F. (2005). Genetic analysis of durable resistance to yellow rust in bread wheat. Theor. Appl. Genet. 110, 1401–1409. Manisterski, J., Treeful, L., Tomerlin, J. R., Anikster, Y., Moseman, J. G., Wahl, I., and Wilcoxson, R. D. (1986). Resistance of wild barley accessions from Israel to leaf rust collected in the USA and Israel. Crop Sci. 26, 727–730. Martelli, G. P. (2001). Transgenic resistance to plant pathogens: Benefits and risks. J. Plant Pathol. 83, 37–46. Matzinger, D. F., and Wernsman, E. A. (1968). Four cycles of mass selection in a synthetic variety of an autogamous species Nicotiana tabacum L. Crop Sci. 8, 239–243. McDonald, B. A., and Linde, C. (2002). Pathogen population genetics, evolutionary potential, and durable resistance. Annu. Rev. Phytopathol. 40, 349–379. McIntosh, R. A., Wellings, C. R., and Park, R. F. (1995). ‘‘Wheat Rusts: An Atlas of Resistance Genes.’’ Kluwer Academic Publishers, London. Melchers, L. S., and Stuiver, M. H. (2000). Novel genes for disease‐resistance breeding. Curr. Opin. Plant Biol. 3, 147–152. Michelmore, R. W. (2003). The impact zone: Genomics and breeding for durable disease resistance. Curr. Opin. Plant Biol. 6, 397–404. Mundt, C. C., Cowger, C., and Garrett, K. A. (2002). Relevance of integrated disease management to resistance durability. Euphytica 124, 245–252. Nusbaum, J. J., and Ferris, H. (1973). The role of cropping systems in nematode population management. Annu. Rev. Phytopathol. 11, 423–440. Oberhagemann, P., Chatot‐Balandras, C., Scha¨fer‐Pregl, R., Wegener, D., Palomino, C., Salamini, F., Bonnel, E., and Gebhardt, C. (1999). A genetic analysis of quantitative resistance to late blight in potato: Towards marker‐assisted selection. Mol. Breed. 5, 399–415. Oerke, E. C. (2006). Crop losses to pests. J. Agric. Sci. 144, 31–43. Oldroyd, G. E. D., and Staskawicz, B. J. (1998). Genetically engineered broad‐spectrum disease resistance in tomato. Proc. Natl. Acad. Sci. USA 95, 10300–10305. Parlevliet, J. E. (1995). Present problems in and aspects of breeding for disease resistance. In ‘‘Molecular Methods in Plant Pathology’’ (R. P. Singh and U. S. Singh, Eds.), pp. 273–287. CRC Press. Boca Raton, FL, USA. Parlevliet, J. E. (2002). Durability of resistance against fungal, bacterial and viral pathogens: Present situation. Euphytica 124, 147–156. Pink, D. A. C. (2002). Strategies using genes for non‐durable disease resistance. Euphytica 124, 227–236. Plaisted, R. L., Thurston, H. D., Brodie, B. B., and Hoopes, R. W. (1984). Selecting for resistance to diseases in early generations. Am. Potato J. 61, 395–403. Posch, D. M. (2003). ‘‘Development and evaluation of selection for potato genotypes resistant to phytophthora infestans in the early breeding generations.’’ M.S. Thesis University of Minnesota, St. Paul, MN, USA.
366
D. D. STUTHMAN ET AL.
Qi, X., Niks, R. E., Stam, P., and Lindhout, P. (1998). Identification of QTLs for partial resistance to leaf rust (Puccinia hordei) in barley. Theor. Appl. Genet. 96, 1205–1215. Randle, W. M., Davis, D. W., and Groth, J. V. (1984). Improvement and genetic control of partial resistance in sweet corn to corn leaf rust. J. Amer. Soc. Hort. Sci. 109, 777–781. Renfro, B. L. (1985). Breeding for disease resistance in tropical maize and its genetic control. In ‘‘Breeding Strategies for Maize Production Improvement in the Tropics’’ (A. Brandolini and F. Salamini, Eds.), pp. 341–366. FAO/UN, Instituto Agronomico per l’Otremare, Fireze. Riggs, R. D. (1977). Worldwide distribution of soybean‐cyst nematode and its economic importance. J. Nematol. 9, 34–39. Robinson, R. A. (1976). ‘‘Plant Pathosystems.’’ Springer‐Verlag, Berlin. Rubiales, D., and Niks, R. E. (1995). Characterization of Lr34, a major gene conferring nonhypersensitive resistance to wheat leaf rust. Plant Dis. 79, 1208–1212. Schnurbusch, T., Pailard, S., Schori, A., Messner, M., Schachemayr, G., Winzeler, M., and Keller, B. (2004). Dissection of quantitative and durable leaf rust resistance in Swiss winter wheat reveals a major resistance QTL in the L34 region. Theor. Appl. Genet. 108, 477–484. Segal, A., Manisterski, J., Fischbeck, G., and Wahl, I. (1980). How plant populations defend themselves in natural populations. In ‘‘Plant Disease: An Advanced Treatise, Vol. V. How Plants Defend Themselves’’ (J. G. Horsfal and E. B. Cowling, Eds.), pp. 75–102. Academic Press, New York. Shaner, G. (1981). Genetic resistance for control of plant disease. In ‘‘CRC Handbook of Pest Management in Agriculture: Vol. 1’’ (D. Pimentel, Ed.), pp. 395–444. CRC Press. Boca Raton, FL, USA. Simmonds, N. W. (1985). A plant breeder’s perspective of durable resistance. FAO Plant. Prot. Bull. 33, 13–17. Simmonds, N. W. (1988). Synthesis: The strategy of rust resistance breeding. In ‘‘Breeding Strategies for Resistance to the Rusts of Wheat’’ (N. W. Simmonds and S. Rajaram, Eds.), pp. 119–136. CIMMYT, Mexico, DF. Simmonds, N. W. (1991). Genetics of horizontal resistance to diseases of crops. Biol. Rev. 66, 189–241. Singh, R. P., Huerta‐Espino, J., and William, H. M. (2005). Genetics and breeding for durable resistance to leaf and stripe rusts in wheat. Turk. J. Agric. For. 29, 121–127. Spielmeyer, W., McIntosh, R. A., Kolmer, J., and Lagudah, E. S. (2005). Powdery mildew resistance and Lr34/Yr18 genes for durable resistance to leaf and stripe rust congregate at a locus on the short arm of chromosome 7D of wheat. Theor. Appl. Genet. 111, 731–735. SteVenson, B. J. (2002). Fusarium head blight of barley: Impact, epidemics, management, and strategies for identifying and utilizing genetic resistance. In ‘‘Fusarium Head Blight of Wheat and Barley’’ (K. J. Leonard and W. R. Bushnell, Eds.), pp. 241–295. APS Press, St. Paul, MN, USA. Stevenson, W. R., Loria, R., Franc, G. D., and Weingartner, D. P. (2001). ‘‘Compendium of Potato Diseases, 2nd ed.’’ APS Press, St. Paul, MN. Strange, R. N., and Scott, P. R. (2005). Plant disease: A threat to global food security. Annu. Rev. Phytopathol. 43, 83–116. Stuber, C. W. (1995). Maping and manipulating quantitative traits in maize. Trends Genet. 11, 477–481. Stuiver, M. H., and Custers, J. H. H. V. (2001). Engineering disease resistance in plants. Nature 411, 865–868. Stuthman, D. D. (2002). Contribution of durable disease resistance to sustainable agriculture. Euphytica 124, 253–258. Takken, F. L. W., and Joosten, M. H. A. J. (2000). Plant resistance genes: Their structure, function and evolution. Eur. J. Plant Pathol. 106, 699–713.
BREEDING CROPS FOR DURABLE RESISTANCE TO DISEASE
367
Tanksley, S. D., and Nelson, J. C. (1996). Advanced backcross QTL analysis: A method for the simultaneous discovery and transfer of valuable QTLs from unadapted germplasm into elite breeding lines. Theor. Appl. Genet. 92, 191–203. Thurston, H. D. (1971). Relationships of general resistance: Late blight of potato. Phytopathology 61, 620–626. Turkensteen, L. J. (1993). Durable resistance of potatoes against Phytophthora infestans. In ‘‘Durability of Resistance’’ (T. H. Jacobs and J. E. Parlevliet, Eds.), pp. 115–124. Kluwer Academic Publishers, Dordrecht, The Netherlands. Ullstrup, A. J. (1970). A comparison of monogenic and polygenic resistance to Helminthosporium turcicum in corn. Phytopathology 60, 1597–1599. Ullstrup, A. J. (1978). Evolution and dynamics of corn diseases and insect problems since the advent of hybrid corn. In ‘‘Maize Breeding and Genetics’’ (D. B. Walden, Ed.), pp. 283–297. Wiley Interscience, New York. Vanderplank, J. E. (1968). ‘‘Disease Resistance in Plants.’’ Academic Press, New York. van Dijk, P., Parlevliet, J. E., Kema, G. H. J., Zeven, A. C., and Stubbs, R. W. (1988). Characterization of the durable resistance to yellow rust in old winter wheat cultivars in the Netherlands. Euphytica 38, 149–158. Wahl, I. (1970). Prevalence and geographic distribution of resistance to crown rust in Avena sterilis. Phytopathology 60, 746–749. Wahl, I. (2003). ‘‘Disease Resistance from Crop Progenitors and Other Wild Relatives.’’ Herbert Utz Verlag, Mu¨nchen, Germany. Walker, J. C. (1959). Progress and problems in controlling plant diseases by host resistance. In ‘‘Plant Pathology: Problems and Progress, 1908–1958’’ (C. S. Holton, G. W. Fischer, R. W. Fulton, H. Hart, and S. E. A. McCallam, Eds.), pp. 32–41. University of Wisconsin Press, Madison, WI, USA. Warburton, C. W. (1915). Grain crop mixtures. J. Am. Soc. Agron. 7, 20–29. Westie, R. L. (1991). Breeding for resistance. Adv. Plant Pathol. 7, 193–227. Welz, H. G., and Geiger, H. H. (2000). Genes for resistance to northern corn leaf blight in diverse maize populations. Plant Breed. 119, 1–14. Wiese, M. V. (1987). ‘‘Compendium of Wheat Diseases.’’ APS Press, St. Paul, MN, USA. Williams, E. B., and Kuc, J. (1969). Resistance in Malus to Venturia inaequalis. Annu. Rev. Phytopathol. 7, 223–246. Winslow, M. D., Okada, K., and Correa‐Victoria, F. (1997). Silicon deficiency and the adaptation of tropical rice ecotypes. Plant Soil 188, 239–248. Wolfe, M. S. (1983). Genetic strategies and their value in disease control. In ‘‘Challenging Problems in Plant Health’’ (T. Kommendahl and P. H. Williams, Eds.), pp. 461–473. American Phytopatholgical Society, St. Paul, MN. Wolfe, M. S., and Barrett, J. A. (1980). Can we lead the pathogen astray? Plant Dis. 64, 148–155. Wubben, J. P., Boller, T., Hone´e, G., and De Wit, P. J. G. M. (1997). Molecular and biochemical aspects of host‐pathogen‐interactions; with prospects for molecular resistance breeding. In ‘‘Resistance of Crop Plants against Fungi’’ (H. Hartleb, R. Heitefuss, and H. H. Hoppe, Eds.), pp. 202–237. Gustav Fischer Verlag, Jena. Young, N. D. (1996). QTL mapping and quantitative disease resistance in plants. Annu. Rev. Phytopathol. 34, 479–501. Zadoks, J. C. (1993). The partial past: Comments on the history of thinking about resistance of plants against insects, nematodes, fungi, and other harmful agents. In ‘‘Durability of Disease Resistance’’ (T. H. Jacobs and J. E. Parlevliet, Eds.), pp. 11–22. Kluwer Academic Publishers, Dordrecht, The Netherlands.
Index A ADB. See Asia Development Bank ADOC genes. See Allele diversity at orthologous candidate genes AVymetrix microarray analysis, 91–2 Aflatoxin, 168 AFLPs. See Amplified fragment length polymorphisms Agricultural impact, on disease resistance, 336–7 Agricultural Production System Simulator cropping systems model, 261 Agrobacterium-mediated transformation, 112–14 Agrobacterium tumefaciens, 80, 112 Agronomic traits cloning of QTL, 255–6 linkage mapping, 254 QTL mapping expression, 257–9 and seed quality traits, 228–33 Alfalfa, 83–5 flavonoid profiling in, 98 improvement, case study for, 119 of aluminum tolerance, 120–4 forage quality enhancement, 124–32 molecular development, 132 Alien gene introgression, challenges for, 233–5 Allele diversity at orthologous candidate genes, 175 Allele-specific amplification, 188 Aluminum tolerance, alfalfa improvement and, 120. See also Alfalfa marker-assisted breeding for, 121–2 transgenic approach to, 122–4 Amplified fragment length polymorphisms, 104, 180 Annual water resource, 29 ANR. See Anthocyanidin reductase Anthesis silking interval, 216 Anthocyanidin reductase, 131 Anthocyanin pigment (PAP1) mutant, 85 Anthracnose, in common bean, 168 APSIM cropping systems model. See Agricultural Production System Simulator cropping systems model Arabidopsis, 80, 97, 100, 110, 359 BANYULS gene, 131 CT biosynthesis in, 131
EST and, 101 floral dip method and, 84 Medicago and, 85 microarray technologies in, 91 nontissue culture develpoment and, 115 PAP1 mutant, 101 SSRs in, 106 TILLING project, 178 Arabidopsis thaliana, 78, 107 genome sequencing and, 80–2 Arsenic, in water, 21–2 ASA. See Allele-specific amplification Ascochyta blight, in pea, 168 ASI. See Anthesis silking interval Asia Development Bank, 13 Avena sterilis, 327 Avirulent (Avr) gene, in pathogen, 326, 340 avrXa7 and avrXa5 genes, 340 AWR. See Annual water resource
B Bacillus thuringiensis, 18, 114 Bacterial blight, 168 BaMMV. See Barley mild mosaic virus Banana genome sequencing in, 190–1 world-wide average production, 166–7 Barley, 83, 234, 327 cultivars and hybrids by MAS, 246–7 marker/QTL validation agronomic/seed quality traits, 211 biotic and abiotic stresses in, 207 Barley mild mosaic virus, 168 Barley yellow dwarf virus, 168 BB. See Bacterial blight ‘‘Beavis eVect,’’ genetic variance, 213 Best linear unbiased prediction procedure, 266–7 Biofuel production, 170 Bloat resistance, alfalfa improvement and, 129–32. See also Alfalfa Blue water. See Humanity’s freshwater resource Blumeria graminis, 339 BLUP procedure. See Best linear unbiased prediction procedure 369
370
INDEX
BMV. See Brome mosaic virus BPH. See Brown plant hopper Brassica juncea, 117 Brome mosaic virus-based vector, 118 Brown midrib (bm) mutants in maize, 187 Brown plant hopper (Bph1 and Bph2) in rice, 217 BYDV. See Barley yellow dwarf virus
C CaVeic acid O-methyltransferase, 187 CaVeic acid 3-O-methyltransferase, 127–8 Callus culture, 114 Candidate gene marker. See Gene-targeted markers CAPS. See Cleaved amplified polymorphic sequences Cardamine amar, 188 Cassava mosaic virus, 168 CBB. See Common bacterial blight CCoAOMT. See CoA 3-O-methyltransferase cDNA amplified fragment length polymorphism (cDNA-AFLP), 94, 101 Cellulose-based bioenergy systems, 170 Cercospora nicotianae, 359 Cereal, world-wide average production, 166–7 CGIAR. See Consultative Group on International Agricultural Research Chemical barriers to pathogen invasion, in plants, 330–1 Chestnut blight, in North America, 327 Chip-based technologies, 107 Chlorinated hydrocarbons, 18 Chromosome segment substitution lines, 176, 177 CIM. See Composite interval mapping CIMMYT wheat breeding program, 347–8, 355 CISPs. See Conserved-intron scanning primers Classical plant breeding techniques, 103 Claviceps purpurea, 329 Cleaved amplified polymorphic sequences, 104 marker, 180, 188 Clonal single molecule arrays technologies, 95 CLs. See Contig lines CMTV. See Comparative map and trait viewer CoA 3-O-methyltransferase, 128 Cochliobolus carbonum, 334
Cochliobolus heterostrophus, 334, 343 Cochliobolus sativus, 341 Cochliobolus victoriae, 334 Common bacterial blight, 168 Common bean agronomic/seed quality traits, 232 cultivars and hybrids by MAS, 246 drought tolerance, 227 Comparative map and trait viewer, 272 Composite interval mapping, 268 COMT. See CaVeic acid O-methyltransferase Condensed tannins, 129–31 Conserved-intron scanning primers, 192 CONSTANS-LIKE gene family, 111–12 Consultative Group on International Agricultural Research, 171 gene banks, 172 mandated crops, 175 Contig lines, 176 Conventional core collections, 173 Corn. See Zea mays Crop breeding classical approaches pedigree breeding, 354–5 perennial species, 355–7 recurrent selection, 352–4 molecular approaches genetic transformation, 358–9 marker-assisted selection, 357–8 Crop inherently transient resistance breakdown, 338–40 Crop productivity, factors aVecting stability loss, 321, 323 Crop rotations, 323, 338 Crown rust, in wild oat, 327 CSMATM technologies. See Clonal single molecule arrays technologies CSSLs. See Chromosome segment substitution lines CTs. See Condensed tannins Cyamopsis tetragonoloba (guar), 84
D Dana-Farber Cancer Institute (DFCI), 87 DArT markers spanning 1137-cM barley genome, 180 DArT microarray-based technique, 180 Database of ‘‘omes,’’ 100
INDEX Dead zones, 41 Dengue hemorrhagic fever, 15 De novo genome sequencing, 137–8 Deoxynivalenol, 168 Desalination, 53–4 Developing countries, demand for maize in, 169 DHF. See Dengue hemorrhagic fever DHL. See Double-haploid lines Digestible neutral detergent fiber, 187 Disease avoidance and escape, in plants, 328–9 Disease management, by crop cultivars, 321 Disease resistance pathway, modulating approach in, 359 Disease resistant crops, breeding for, 320–1, 324 Disease triangle, 322 DNA marker, 179 development of, 357 enhanced selection power, 239 technology, 170 DNA-selection breeding, 107 DNDF. See Digestible neutral detergent fiber DOME. See Database of ‘‘omes’’ DON. See Deoxynivalenol Dot-blot-SNP analysis, 107 Dothiostroma, 356 Double-haploid lines, 176 Downy mildew in pearl millet, 217 DpnII, 95 Drip irrigation, 48, 51, 53 Dutch elm disease, in America, 327 Dwarf8 gene of maize, 187
E Ecological footprint, 31–2 EcoTILLING approach, 178 EF. See Ecological footprint Electra parasitic weeds, 168 Environmental water requirement, 41 ESTs. See Expressed sequence tags EST-SSR markers, 106, 121 European water framework directive, 56 EWR. See Environmental water requirement Expressed sequence tags, 78, 80, 101, 182 generation and analysis of, 89–90 for legume species, 87–9 sequence databases, 182 sequencing, 84, 104
371 SSRs development from, 105–6 trichomes and, 92
F Falkenmark stress indicator, 11, 26–9 False discovery rate (FDR), 214 Fecal pollution, 14 FHB. See Fusarium head blight Floral development, in wheat disease resistance, 329 Fluorescence-based technologies, 107 FM. See Functional markers Food crops productivity, abiotic and biotic factors eVects, 168 Forage quality enhancement, alfalfa improvement and See also Alfalfa bloat resistance, 129–32 forage digestibility, improved, 124–9 Forest tree breeding programs, 356 Fourier transform ion cyclotron mass spectrometry, 100, 101 Freshwater annual global consumption, 10 Fritillaria assyriaca, 188 FSI. See Falkenmark stress indicator FT-MS. See Fourier transform ion cyclotron mass spectrometry Functional genomics genetic resources for, 85–6 transgenesis and, 110–12 virus-induced gene silencing (VIGS) for, 117–18 Functional markers, 179, 186–8 Fungal diseases food production loss by, 321–2 monogenic resistance to, mechanisms, 340–1 Fusarium, 329 Fusarium head blight, in wheat and barley, 168 Fusarium moniliforme, 344 Fusarium oxysporum f. conglutinans, 334 Fusarium solani f. sp. glycines, SDS caused by, 133
G Gaines, wheat cultivar, 325, 348 Gall midge Gm-6t in rice, 217 Ganges river issue, 38
372
INDEX
Gas chromatography-mass spectrometry (GC-MS), 97–8, 100–1 GBSS1 gene. See Granule-bound starch synthase 1 gene GCP. See Generation Challenged Program GEI. See Genotype-by-environment interaction GenBank, 136 Genealogy Management System, 271 Gene expression, by eQTL analysis, 258 Gene-for-gene relationships, in plant diseases, 326 Gene knockdown transgenesis, TILING and, 119 Gene pyramiding cereals, 220 in crops, 339–40 legumes, roots and tubers, 221 Generation Challenged Program, 175 Gene-targeted markers, 179, 182–3 Genetically structured trait-based core collections, 173 Genetic libraries development, 234 Genetic linkage map, 192–3 Genetic marker development, for selection genetic linkage map, 192–200 genetic resources, 171–9 genome sequencing, 188–92 genomic resources, 179–88 marker-trait associations, 200–4 Genomic revolution, 175 Genotype-by-environment interaction, 169, 270 Germplasm data evaluation by software packages, 262–5 Gibberella zea, 344 Global food production, breeding challenges and, 165–70 Global germplasm collections, 173 Global International Waters Assessment (GIWA), 23 Global situation, of water chemical contamination and agricultural nutrients, 16–8 agricultural pesticides, 18–20 chemicals emission from industries, 20–1 natural toxics, 21–2 drinking, sanitation and waterborne disease and, 12–6
ecosystems and groundwater overdraft, 24–6 stream flow modification, 23 wetlands loss, 23–4 in scarced and stressed countries, 10–12 use by sectors, 9–10 Globodera pallida, 338 Globodera rostochiensis, 338 Glycine max. See Soybean GMendel and MapDisto, linkage map construction and, 272 GMS. See Genealogy Management System Granule-bound starch synthase 1 gene, 178 Greater Anatolian Project (GAP), 36 Green Revolution, 3–4, 168–9 Green water. See Humanity’s freshwater resource H Helianthus annuus L., SNP markers and, 107 Heterosis molecular basis, 250–4 High performance liquid chromatography, 98 High-temperature adult plant resistance, 325 Hm1 gene, 340–1 Hordeum vulgare. See Barley HPLC. See High performance liquid chromatography HTAP resistance. See High-temperature adult plant resistance Humanity’s freshwater resource, 6–7 I ICIS. See International Crop Information System ILs. See Introgression lines IMAS. See Integrated decision support system for marker-assisted plant breeding Indus river issue, 37–8 Institute for Genome Research (TIGR), 87, 89 Integrated decision support system for marker-assisted plant breeding, 272 Intermittent overdrafting, 25 International Crop Information System, 271 International Drinking Water Supply and Sanitation Decade 1980s, 13 International Wheat Genome Sequencing Consortium, 189
INDEX Intestinal helminths, 15 Introgression lines, 176 IWGSC. See International Wheat Genome Sequencing Consortium
J Johnston plan, 35 Jordan river issue, 34–6
L Late blight disease, in potato, 220 LD mapping, 201–4 Leaf rust in barley, 327 (Lr19, Lr51, and Yr15) in wheat, 217 Legume Information System, 133–4 Legume, world-wide average production, 166–7 Lignin, 125–8 LIS. See Legume Information System Livestock revolution, 169 Lolium temulentum, 115 Lotus japonicus as model species, 83–5 TILLING in, 86 Lr34 and Lr46 genes, in wheat, 332–3, 341, 347, 355
M MABC. See Marker-assisted back crossing MAGE. See Marker-assisted germplasm evaluation Magnaporthe grisea, 330, 341 Maize breeding programs in, 344 drought tolerance, 226 genome sequencing in, 190 marker/QTL validation agronomic/seed quality traits, 230–1 biotic and abiotic stresses in, 207 QuGene/APSIM with QTL data, 261 Maize streak virus, in corn, 168 Malate dehydrogenase, 124 MALDI-TOF MS, 107 MAPMAKER/QTL mapping software, 268
373
MAP MANAGER QT mapping software, 268 ‘‘Mapping As You Go’’ (MAYG) approach, 216 MapPop mapping software, 268 Marker-accelerated backcross breeding, 217 Marker-assisted back crossing, 205 Marker-assisted genetic enhancement impact of, 239–50 in private sector breeding programs application, 235–9 public sector breeding programs in agronomic and seed quality traits, 228–33 alien gene introgression, 233–5 resistance to biotic stresses, 217–20 tolerance to abiotic stresses, 221–8 Marker-assisted germplasm evaluation, 262 Marker-assisted selection, 108–10, 171, 236 modelling and stimulation, 259–61 Marker-dense meiotic linkage maps, 193 Marker for inherited traits, 205–6 Marker validation and refinement, 204 and gene traits, 205–6 QTL markers, 206–16 MARS. See Marker-assisted recurrent selection MAS. See Marker-assisted selection Massively parallel signature sequencing, 78, 95, 253 MCQTL mapping software, 268 MDH. See Malate dehydrogenase Medicago, 86 Arabidopsis and, 85 ‘‘gene expression atlas’’ for, 91 glycosyltransferase gene expression in, 92 Medicago sativa. See Alfalfa Medicago truncatula, 100, 115, 121, 127, 132, 137 arrays generated for, 91 herbicide-resistant, 102 as model species, 83–5 Tnt1 in, 86 orthologues, 93 SSRs in, 106 triterpene saponins accumulation in, 101 WXP1 gene in, 112 MeJA signaling. See Methyl jasmonate signaling Mendelian genetic markers, 103 Mesorhizobium loti, 113
374
INDEX
Metabolome, 79 Metabolomic analysis classification of, 96–7 metabolic fingerprinting, 98–100 nonbiased metabolomics, 100 targeted profiling, 97–8 integrating transcriptomic, metabolomic datasets and, 101–2 ‘‘substantial equivalence’’, profiling technologies and, 102–3 Metabolomic datasets, integrating transcriptomic and, 101–2 Methemeglobinemia, 17–18 Methyl jasmonate signaling, 91, 101 tobacco BY2 cell suspension cultures, 95 MicroRNAs (miRNAs), 92–3 Microsatellites. See Simple sequence repeats Millennium Ecosystem Assessment, 23, 41–3 Mini-core collections, 175 Mixture over markers model, 258 Mlo allele, in powdery mildews, 341, 349 Molecular breeding programs, computational systems in, 261 design and simulation, 270–1 eYciency and scope, 250–61 GEI analysis, 270 genetic map construction, 267 germplasm evaluation, 262 information management and integrated tool, 271–3 management of breeding populations, 266–7 marker-assisted selection, 269 marker trait association, 267–9 Molecular genetic maps, 104–5 Molecularization of Public Crop Improvement, future prospects, 273–8 Molecular markers generation, genomics for simple sequence repeats, 105–6 single nucleotide polymorphisms, 106–7 TILLING, 107–8 marker types, 103–4 metabolomic-based, 108–10 molecular genetic maps, 104–5 MOM model. See Mixture over markers model Morphological resistance, in plants collenchyma bundle morphology, 330 floral development in wheat, 329 hair and waxes in leaf epidermis, 329–30
MPSS. See Massively Parallel Signature Sequencing; Massively parallel signature sequencing MRNA transcript, poly-A tail of, 95 M. truncatula Gene Index (MtGI), 87, 89 Multiline cultivars, in cereal crops, 339
N National pesticide survey, 19 National Water Quality Assessment (NAWQA), 19 ncRNAs. See Nonprotein-coding RNAs Near infrared (NIR) spectroscopy, 98–100 Near isogenic lines, 176 Netherlands National Institute of Public Health and Environmental Protection, 19 Neutral markers. See DNA marker Nicotiana benthamiana, 118 Nile river issue, 36–7 NILs. See Near isogenic lines NIM1/NPR gene, 359 Nitrogen, in soil, 17 NMR. See Nuclear magnetic resonance Nonbiased metabolomics, 100 Nonprotein-coding RNAs, 117 Northern leaf blight in maize, 344–5 Nuclear magnetic resonance, 98–100 Nugaines, wheat cultivar, 348
O Onchocerciasis, 14 Orobanche parasitic weeds, 168 Oryza rufipogon, 234 Overdominant (ODO) eVects, 251
P Parana´ River basin, 20 PCA. See Principle component analysis PCR-based sequence-tagged site (STS) markers, 180 PCR technique. See Polymerase chain reaction technique
INDEX Pea agronomic/seed quality traits, 232–3 marker/QTL validation agronomic/seed quality traits in, 211 biotic and abiotic stresses in, 207 Pearl millet cultivars and hybrids by MAS, 246–7 drought tolerance, 227 Per capita liquid water resource availability (PWR), 11 Periconia circinata, 334 PET. See Potential evapotranspiration PGR. See Plant genetic resources Phaseolus vulgaris, 342 Phenylalanine ammonia-lyase (PAL2) promoter, 128 Phytoalexins, in plant, 331 Phytophthora infestans, 350 PLABQTL mapping software, 268 Plantain crops, world-wide average production, 166–7 Plant breeders breeding strategies in disease resistance, 332–5 eVorts and objectives of, 321, 324 Plant breeding strategies, for disease resistance eVectiveness, 333 hypersensitive resistance, 335 inheritance, 332 specificity, 333–5 Plant disease epidemics, causes of, 328 environmental factors aVecting, 323–4 host abundance and susceptibility, 322–3 host and environment interactions, 324–5 host and pathogen interactions, 325–6 pathogen abundance and virulence, 323 pathogen and environment interactions, 325 Plant disease, natural populations and, 327 Plant disease resistance, in breeding eVorts chemical barriers to pathogen invasion, 330 disease avoidance and escape, 328–9 induced responses to invasion, 331 morphological resistance in plants, 329 Plant genetic resources, 171 Plant hypersensitive response, induction of, 359–60 Plant’s response to disease invasion and wounds cell wall in disease resistance, 331 hairs or waxes in disease resistance, 329–30
375
Polygenic disease resistance durability of, 342–3 examples of barley, 349–50 maize, 344–6 potato, 350–2 wheat, 346–9 Polymerase chain reaction technique, 103 Poplar. See Populus trichocarpa Population growth, food production and, 4–6 trends and water stresses on climate, 42–3 industries and municipality, water consumption by, 32–4 in short and stressed countries, 26–30 threats to ecosystem, 40–2 transboundary issues and, 34–8 in urban states, 30–2 water deficit, projection of, 38–40 Populus trichocarpa, genome sequencing and, 83 Postgenomics technologies, 79 Potato breeding program, for disease resistance, 351–2 gene pyramiding, 222–3 marker/QTL validation, biotic and abiotic stresses in, 207 Potential evapotranspiration, 42, 48 Powdery mildew, in barley and cereal, 326, 328, 349 Powdery mildew (mlo-9) in barley, 217 Principle component analysis, 97, 101 of soluble phenolic compound, 102 ‘‘2010 Program’’, 82 Proteomics, 95–96 Puccinia coronata, 327 Puccinia graminis, 328 Puccinia sorghi, 342 Puccinia striiformis, 325, 341 Puccinia triticina, 327 Pythium root rot of sugarcane, 342 seedling blight of maize, 342
Q QTL. See Quantitative trait loci QTL EXPRESS mapping software for QTL, 268
376
INDEX
Quantitative real-time polymerase chain reaction (qRT-PCR), 92 Quantitative trait loci, 109, 132 analyses in disease resistance, 357–8 analysis, 98, 170 PlabQTL for, 272 QTL Cartographer for, 272 mapping, 104, 108, 121, 206 markers for complex traits, 206–16 in northern leaf blight, 345 QU-GENE software for GEI analysis in crop breeding, 270
genome sequencing, 82–3, 190 maker/QTL validation agronomic/seed auality traits in, 211 biotic and abiotic stresses in, 207 submergence tolerance, 228 RILs. See Recombinant inbred lines Riparian ecosystems, 9 Riparian nations, 36 Root and tuber, world-wide average production, 166–7 Rp1 gene, 342 Russian wheat aphidm, in barley, 168 Rust disease of cereal, 326
R S Race-specific resistance gene, in host, 326 Rainfed agriculture, improvement methods for, 50–1 Random DNA markers, 180 Randomly amplified polymorphic DNA, 103, 180 Random mutation techniques, 178 RAPD. See Random amplified polymorphic DNA rasiRNAs. See Repeat-associated small interfering RNAs RDM. See Random DNA markers Recalcitration of crop species to genetic transformation, prevention strategies for, 114–16 Recombinant inbred lines, 104, 176 ‘‘Regulated deficit irrigation’’ (RDI), 49 Reliable disease resistance in crops, importance of, 320 Repeat-associated small interfering RNAs, 92 Resistance classification, types of, 333–4 Restriction fragment length polymorphisms, 103, 104, 121 RFLPs. See Restriction fragment length polymorphisms Rhamnus cathartica, 338 Rhizobium species, 113 Rice agronomic/seed quality traits, 228–30 cultivars and hybrids by MAS, 246–8 drought tolerance, 221–6 endosperm, starch biosynthesis in, 187 fragrance in, 187–8 gene expression with QTL for osmotic adjustment (OA), 258–9
SAGE. See Serial analysis of gene expression S. almum, 199 SBS. See Sequencing-by-synthesis SCARs. See Sequence characterized amplified regions Schistosomiasis, 15 SCN. See Soybean cyst nematode SDS. See Sudden death syndrome SDS-polyacrylamide gel electrophoresis, 96 Seedling blight, in wheat and maize, 324 Selenium, in water, 22 Semantic web services (SWS) plateform, 137 Sequence characterized amplified regions, 104, 180 Sequenced genomes A. thaliana, 80–2 functional genomics, 85–6 Medicago truncatula and Lotus japonicus, 83–5 poplar, 83 rice, 82–3 transcript profiling approaches for, 87–93 Sequence-tagged microsatellite sites, 104 Sequencing-by-synthesis, 95 Serial analysis of gene expression, 78, 94, 252 Setosphaeria turcica, 344 Silage maize, food for dairy cattle, 187 Silicon concentration in soil, as disease resistance, 330 in upland and lowland ecotypes, 324–5 Simple sequence repeats, 104–6, 180 markers, 175 SINGER. See System-wide Information Network for Genetic Resources
INDEX Single gene introgression, 217–20 Single gene transfer by MAS in legumes, 220 for resistance to biotic stresses, markerassisted selection for crops and, 218–19 Single nucleotide polymorphisms, 104, 106–7 Single strand conformation polymorphisms, 104 Sinorhizobium meliloti, 91, 113 SiRNAs. See Small interfering RNAs Small interfering RNAs, 92–3 SNPs. See Single nucleotide polymorphisms Solanum bulbocastanum, 220 Solanum pennellii, 251 Sorghum bicolor, 83 Sorghum, genome sequencing in, 190 Sorghum halepense, 189, 199 Sorghum propinquum, 199 Soybean, 84 cultivars and hybrids by MAS, 246–8 marker/QTL validation agronomic/seed quality traits in, 211 biotic and abiotic stresses in, 207 Soybean Breeder’s Toolbox, 133 Soybean cyst nematode, 235 Spotted arrays, 90 Sr6 gene, 338 SSCP. See Single strand conformation polymorphisms SSRs. See Simple sequence repeats Stem and leaf rust, in wheat, 325, 327–8 Stenocarpella maydis, 344 STMS. See Sequence-tagged microsatellite sites Stream flow modification, 23 Striga parasitic weeds, 168 Stripe rust (Yr4), 217 in wheat, 355 STSs. See Sequence-tagged sites ‘‘Substantial equivalence’’, profiling technologies and, 102–3 Sudden death syndrome, in soybean, 133–4 Sunflower. See Helianthus annuus L. Surge irrigation, 48 Sustainable management in developing world, 66–7 management strategies and, 62–6 in society, 67–8 unsustainable practices for, 58–62 ‘‘Systems Biology,’’ 79
377
System-wide Information Network for Genetic Resources, 171
T Targeting Induced Local Lesions IN Genome, 86, 107–8, 176 gene knockdown transgenesis and, 119 Target production environment, 270 Target region amplification polymorphisms, 185 T-DNA. See Transferred DNA TDT. See Transmission-disequilibrium test Tentative consensus sequences (TCs), 89 Thielaviopsis basicola, 342 Tigris-Euphrates basin, 36 TILLING. See Targeting Induced Local Lesions IN Genome Tnt1, tobacco retrotransposon, 86 TPE. See Target production environment Trachoma, 14 Trait integration and commercialization, transgenesis for, 116–17 Transcriptomics, approaches for sequenced genomes, 87–93 for species lacking genomics resources, 93–5 Transferred DNA, 85, 86, 113 functional genomics and, 110–11 tagging, 80 Transgenesis functional genomics, as tool for, 110–12 prevention strategies, for recalcitration of crop species to genetic transformation, 114–16 TILLING, 119 trait integration and commercialization, 116–17 transgenic plants generation, approaches to, 112–14 virus-induced gene silencing, 117–18 Transgenic disease resistant crops, development approaches, 359 Transgenic plants generation, approaches, 112–14 Transmission-disequilibrium test, 201 Transpired rainwater. See Humanity’s freshwater resource Transposon-flanking sequences, database development of, 86
378
INDEX
TRAP. See Target region amplification polymorphisms Trichomes, 92, 96 Triticum aestivum. See Wheat Triticum dicoccoides, 327 Trypanosomiasis, 15 Tuber rot, in potatoes, 329 T-urf 13 gene, in maize, 341 Tyloses block, in plant, 331
U United Nations Commission on Sustainable Development, 11 US Geological Survey (USGS), 19
V VIGS. See Virus-induced gene silencing ‘‘Virtual water,’’ 65 Virulent gene, in pathogen, 326 Virulent mutants in crops plants, 337 Virus-induced gene silencing, for functional genomics, 117–18
W Water, global situation of chemical contamination and agricultural nutrients, 16–8 agricultural pesticides, 18–20 chemicals emission from industries, 20–1 natural toxics, 21–2 drinking, sanitation and waterborne disease and, 12–6 ecosystems and groundwater overdraft, 24–6 stream flow modification, 23 wetlands loss, 23–4 in scarced and stressed countries, 10–12 use by sectors, 9–10 Water scarcity dimensions conservation, saving by, 44–5 desalination, 53–4 institutional changes, improvement by, 55–7
irrigation use and eYciency, 46–50 rainfed agriculture, improvement in, 50–1 supplementation of water, economic methods for, 51–3 Water stresses, population trends and on climate, 42–3 industries and municipality, water consumption of, 32–4 in short and stressed countries, 26–30 threats to ecosystem, 40–2 transboundary issues and, 34–8 in urban states, 30–2 water deficit, projection of, 38–40 Wetlands loss, 23–4 Wheat, 83 breeding program, CIMMYT, 347–8, 355 cultivar, 325, 348 cultivars and hybrids by MAS, 246–8 FHB in, 168 leaf rust in, 217 marker/QTL validation agronomic/seed quality traits in, 211 biotic and abiotic stresses in, 207 seedling blight in, 324 Yr11, Yr13, Yr14, Yr18 and Yr28 genes in, 341, 348, 355 Wholegenome ‘‘shotgun’’ sequencing, 81, 83 Whole-plant physiology modeling project, 260 Wild plant species, disease resistance, 335–6 Wind erosion, 6 World Health Organization (WHO), 7
X Xanthomonas oryzae, 109 Xanthomonas oryzee pv. oryzae, 340
Y Yam mosaic virus (YMV), in yam, 168 Yeast artificial chromosomes (YAC), 193 Yellow dwarf virus (Yd2), 217
Z Zea mays, 83