Plant Transformation Technologies E-Book

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

Seitenzahl: 704

Veröffentlichungsjahr: 2011

Beschreibung

Plant Transformation Technologies is a comprehensive, authoritative book focusing on cutting-edge plant biotechnologies, offering in-depth, forward-looking information on methods for controlled and accurate genetic engineering. In response to ever-increasing pressure for precise and efficient integration of transgenes in plants, many new technologies have been developed. With complete coverage of these technologies, Plant Transformation Technologies provides valuable insight on current and future plant transformation technologies. With twenty-five chapters written by international experts on transformation technologies, the book includes new information on Agrobacterium, targeting transgenes into plant genomes, and new vectors and market systems. Including both review chapters and protocols for transformation, Plant Transformation Technologies is vitally important to graduate students, postdoctoral students, and university and industry researchers.

Leseprobe

Contents

Cover

Title Page

Copyright

Contributors

Preface

Section 1: Agrobacterium-Mediated Transformation

Chapter 1: Host Factors Involved in Genetic Transformation of Plant Cells by Agrobacterium

Introduction

Plant Signals Affecting Agrobacterium’s Virulence Machinery

Cell-to-Cell Contact and Passage of T-DNA through Host Cell Barriers

Roles of Plant Factors in Transcytoplasmic Transport and Nuclear Import of the T-Complex

Intranuclear Movement of the T-Complex and Its Uncoating

T-DNA Integration into the Host Genome

Activation and Modulation of the Host Plant Defense Reaction

Concluding Remarks

Acknowledgments

Chapter 2: Genomics of Agrobacterium–Plant Interaction: An Approach to Refine the Plant Transformation Technology

Introduction

Host Gene Expression Profiling in Response to Agrobacterium Infection

Virus-Induced Gene Silencing: A Plant Functional Genomics Tool for Identifying Host Genes Involved in Agrobacterium-Mediated Plant Transformation

Future Prospects

Acknowledgments

Section 2: Other Transformation Technologies

Chapter 3: Particle Bombardment: An Established Weapon in the Arsenal of Plant Biotechnologists

Microprojectiles

Gene Gun Devices

Transient Expression Studies

Stable Transformation by Random Integration

Transformation with Artificial Chromosomes and by Targeted Integration

Chloroplast Transformation

Conclusions

Chapter 4: A Novel Gene Delivery System in Plants with Calcium Alginate Micro-Beads

Introduction

Development of a Novel Transformation Method Using Bioactive Beads

Transformation of Plants, Yeast, and Mammalian Cells Using Bioactive Beads Method

Transformation with Large DNA Fragments Using Bioactive Beads Method

Improvements to Bioactive Beads-Mediated Transformation

Chapter 5: Pollen Transformation Technologies

Introduction

Mature Pollen-Based Transformation

Microspore Maturation-Based Transformation

Microspore and Immature Pollen Embryogenesis-Based Transformation

Conclusions

Chapter 6: Intragenic Vectors and Marker-Free Transformation: Tools for a Greener Biotechnology

Introduction

Genetic Elements

Transformation

Acknowledgments

Chapter 7: Visualizing Transgene Expression

Introduction

History/Evolution of Visual Marker Genes

GFP

Other Fluorescent Proteins

Considerations for Fluorescent Protein Detection

Conclusions

Acknowledgments

Section 3: Vectors, Promoters, and Other Tools for Plant Transformation

Chapter 8: Current State and Perspective of Binary Vectors and Superbinary Vectors

Introduction

Intermediate Vector and Binary Vector

Commonly Used Binary Vectors

Structure of Binary Vectors

Advanced Features of Improved Vectors

Conclusion

Chapter 9: Novel Dual Binary Vectors (pCLEAN) for Plant Transformation

Introduction

Description of the pCLEAN Vector System

Benefits of the pCLEAN Vector System

Conclusion

Acknowledgments

Chapter 10: pORE Modular Vectors for Plant Transformation

Introduction

The pORE Binary Vectors

Enhanced Utilities in Other Modular Vectors

Acknowledgments

Chapter 11: pANIC: A Versatile Set of Gateway-Compatible Vectors for Gene Overexpression and RNAi-Mediated Down-Regulation in Monocots

Why Make a New Vector Set for Grass Transformation?

Features of pANIC

Distribution

Acknowledgments

Section 4: Transgene Integration, Stability, Methylation, Silencing

Chapter 12: Understanding and Avoiding Transgene Silencing

Incidence and Practical Significance of Transgene Silencing

Factors Influencing Transgene Silencing

Mechanisms of Transgene Silencing

Strategies to Avoid Transgene Silencing

Conclusions and Future Prospects

Chapter 13: Site-Specific Recombination for Precise and “Clean” Transgene Integration in Plant Genome

Introduction

Site-Specific Recombination Systems

Generating Target (Founder) Lines

Co-integration and Cassette Exchange Strategies

Mutant Lox Sites

Efficiency of Recovered Events

Co-integration of Random Insertions

Gene Expression from Site-Specific Integration

Possible Factors in Expression-Stability of Site-Specific Transgene

“Clean” Site-Specific Integration Locus

Concluding Remarks

Section 5: Selection Systems, Marker-Free Transformation

Chapter 14: Selectable Marker Genes: Types and Interactions

Introduction

Background

Categories of Selectable Markers and Reporters

Changes in the Plant

Pleiotropic Effects of the Gene

Substantial Equivalence

Position Effects at the Insertion Sites

Effects on Cotransforming Genes

Strategic Vector Design

Conclusions

Acknowledgments

Chapter 15: Transformation Methods for Obtaining Marker-Free Genetically Modified Plants

Introduction

Selectable Markers and Public Concern

Marker-Free Transformation Technology

Transformation without Selectable Marker

Specific Issues Associated with Transformation without Selectable Marker

Generation of Amylose-Free Potato Lines by Transformation without Selectable Marker

Marker Elimination

Conclusion

Acknowledgment

Chapter 16: Intellectual Property Aspects of Plant Transformation

Plant Patents: The Early Years

The Basis of Patents and Other Intellectual Property Rights

Sources of Patent Information

Patents and the Transformation Process

Agrobacterium

Direct Gene Transfer

Transgenic Traits, Genes, and Regulatory Sequences

Patents and Examples of “Second Generation” Traits

Patents and Economic Development

International Perspectives

Sociological and Ethical Aspects

Present and Future Trends

Conclusion

Color Plate

Index

This edition first published 2011 © 2011 by Blackwell Publishing Ltd.

Blackwell Publishing was acquired by John Wiley & Sons in February 2007. Blackwell’s publishing program has been merged with Wiley’s global Scientific, Technical, and Medical business to form Wiley-Blackwell.

Registered office: John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK

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

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

Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted by Blackwell Publishing, provided that the base fee is paid directly to the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923. For those organizations that have been granted a photocopy license by CCC, a separate system of payments has been arranged. The fee codes for users of the Transactional Reporting Service are ISBN-13: 978-0-8138-2195-5/2011.

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

Library of Congress Cataloging-in-Publication Data

Plant transformation technologies / editors, C. Neal Stewart Jr. … [et al.]. p. ; cm. Includes bibliographical references and index. ISBN 978-0-8138-2195-5 (pbk. : alk. paper) 1. Plant genetic engineering. 2. Genetic transformation. 3. Genomics–Methods. 4. Transgenic plants. I. Stewart Jr., C. Neal. [DNLM: 1. Plants, Genetically Modified. 2. Genes, Plant. 3. Genetic Engineering. 4. Transformation, Genetic. SB 123.57]

QK981.5.P586 2011 660.6′5–dc22 2010028091

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

This book is published in the following electronic formats: ePDF 9780470958872; Wiley Online Library 9780470958988; ePub 9780470958940

Contributors

Ashraf Abdeen

Eastern Cereals and Oilseeds Research Centre Agriculture and Agri-Food Canada Ottawa, Ontario, Canada

Laura L. Abercrombie

Department of Plant Sciences University of Tennessee Knoxville, TN, USA

Ajith Anand

DuPont/Pioneer Crop Genetics Research Johnston, IA, USA

Robert G. Birch

Botany Department, BIOL The University of Queensland Brisbane, Australia

Vitaly Citovsky

Department of Biochemistry and Cell Biology State University of New York Stony Brook, NY, USA

Anthony Conner

New Zealand Institute for Crop & Food Research Christchurch, New Zealand

Jim M. Dunwell

School of Biological Sciences University of Reading, UK

John J. Finer

Department of Horticulture and Crop Science Plant Molecular Biology and Biotechnology OARDC/The Ohio State University Wooster, OH, USA

Kiichi Fukui

Department of Biotechnology Graduate School of Engineering Osaka University Osaka, Japan

Michael W. Graham

Botany Department, BIOL The University of Queensland Brisbane, Australia

Loreta Gudynaite-Savitch

Iogen Corp. Ottawa, Ontario, Canada

Zac Hanley

Pastoral Genomics ViaLactia Biosciences Newmarket, Auckland, New Zealand

Dwayne D. Hegedus

Agriculture and Agri-Food Canada Saskatoon, SK, Canada

Ming Hu

Eastern Cereals and Oilseeds Research Centre Agriculture and Agri-Food Canada Ottawa, Ontario, Canada

Douglas A. Johnson

Department of Biology University of Ottawa Ottawa, Ontario, Canada

Shin’ichiro Kajiyama

Department of Biotechnology Graduate School of Engineering Osaka University Osaka, Japan

Naruemon Khemkladngoen

Department of Biotechnology Graduate School of Engineering Osaka University Osaka, Japan

Theodore M. Klein

DuPont Agricultural Biotechnology DuPont Experimental Station Wilmington, DE, USA

Toshihiko Komari

Plant Innovation Center Japan Tobacco Inc. Shizuoka, Japan

Toshiyuki Komori

Plant Innovation Center Japan Tobacco Inc. Shizuoka, Japan

Frans A. Krens

Wageningen UR Plant Breeding Wageningen, The Netherlands

Benoît Lacroix

Department of Biochemistry and Cell Biology State University of New York Stony Brook, NY, USA

Peter R. LaFayette

Department of Crop and Soil Sciences University of Georgia Athens, GA, USA

Phil MacDonald

Biotechnology Environmental Release Assessments Canadian Food Inspection Agency Ottawa, Ontario, Canada

Yuzuki Manabe

Eastern Cereals and Oilseeds Research Centre Agriculture and Agri-Food Canada Ottawa, Ontario, Canada

David G.J. Mann

Department of Plant Sciences University of Tennessee Knoxville, TN, USA

Brian Miki

Eastern Cereals and Oilseeds Research Centre Agriculture and Agri-Food Canada Ottawa, Ontario, Canada

Stephen R. Mudge

Botany Department, BIOL The University of Queensland Brisbane, Australia

Kirankumar S. Mysore

Plant Biology Division The Samuel Roberts Noble Foundation 2510 Sam Noble Pkwy Ardmore, OK, USA

Souad El Ouakfaoui

New Substances Program Biotechnology Section Science and Risk Assessment Directorate Science and Technology Branch Environment Canada Gatineau, Québec, Canada

David W. Ow

South China Botanical Garden Chinese Academy of Sciences Guangzhou, China

Wayne A. Parrott

Department of Crop and Soil Sciences University of Georgia Athens, GA, USA

Tatiana Resch

Max F. Perutz Laboratories University Departments at the Vienna Biocenter Vienna, Austria

Caius M. Rommens

J. R. Simplot Company Simplot Plant Sciences Boise, ID, USA

Jan G. Schaart

Wageningen UR Plant Breeding Wageningen, The Netherlands

Vibha Srivastava

Department of Crop, Soil & Environmental Sciences, and Department of Horticulture University of Arkansas Fayetteville, AR, USA

Peter R. Sternes

Botany Department, BIOL The University of Queensland Brisbane, Australia

C. Neal Stewart, Jr.

Department of Plant Sciences University of Tennessee Knoxville, TN, USA

Alisher Touraev

Max F. Perutz Laboratories University Departments at the Vienna Biocenter Vienna, Austria

Zarir E. Vaghchhipawala

Monsanto Company Middleton, WI, USA

Philippe Vain

John Innes Centre Department of Crop Genetics Norwich Research Park Norwich, United Kingdom

Richard G.F. Visser

Wageningen UR Plant Breeding Wageningen, The Netherlands

Naoki Wada

Department of Biotechnology Graduate School of Engineering Osaka University Osaka, Japan

Anne-Marie A. Wolters

Wageningen UR Plant Breeding Wageningen, The Netherlands

Hua Yan

J. R. Simplot Company Simplot Plant Sciences Boise, ID, USA

Adi Zaltsman

Department of Biochemistry and Cell Biology State University of New York Stony Brook, NY, USA

Preface

From the early 1980s to the present, biotechnologies have yielded, with a great degree of success, the ability to genetically transform a wide variety of plant species. These plant transformation technologies have literally changed the face of agriculture and plant biology. Invariably, whenever any discussion ensues about developments in plant biotechnology with one of the pioneers of the field, especially in industry, we often hear stories of life on the frontier and excitement of breaking new ground in the 1980s. However, “all the really exciting research has already been done,” we are told.

In February 2007, the International Conference on Plant Transformation Technologies was held in the beautiful city of Vienna, Austria. Over 300 participants from 47 countries learned that there was plenty of exciting plant biotechnology research in progress. In the course of the conference, research was presented by groundbreaking researchers, such as Dr. Mary-Dell Chilton, who demonstrated that there was still plenty left to do in plant transformation. Indeed, the editors of this volume were so convinced of this fact that we decided to invite many of the presenters and others to contribute to this volume, which gives a taste of the excitement we felt at the conference.

Thus, borrowing from the title of the Vienna conference, this book covers many topics on the cutting edge of transgenic plants. Of course, we are just really beginning to understand how the original methods of plant transformation, those using Agrobacterium tumefaciens and particle bombardment, truly work. Several chapters are reviews and updates on these technologies, and pertinent genomic interactions among organisms. Several other chapters present new vector systems and describe plasmids that we think will make a huge impact on making plant transformation more accessible for a wider variety of species. Indeed, this theme of accessibility and efficiency is prominent in chapters written by experts in their respective fields. It seems that plant biotechnology is becoming more egalitarian to more types of scientists—from molecular biologists, of course, but now accessible to ecologists and environmental scientists. Transformation technologies have played a huge role in this movement.

The upcoming frontier of challenges for biotechnology deals with issues of transgene precision and regulations. There is now a premium on minimizing the amount of transgenic DNA in plants while maximizing stability of gene expression and trait performance. For regulatory and commercial purposes, characterized integration of transgenes in known locations and precise expression patterns are viewed as helpful to target traits in predictable ways. Indeed, genetic engineering is better than genetic tinkering and transgenic technologies that maximize precision have increasing value in the world marketplace. Chapters were written on methods to gain higher precision of transgene integration and marker-free transformation. However, transgene silencing, either on purpose or by accident, which is one of the key research breakthroughs in recent years, is also important.

As we look to the future of plant biotechnology, we can envision new methods of transformation that could be more efficient and widen the breadth of species and genotypes that can be manipulated. Methods such as those using calcium alginate micro-beads and those targeting alternative cells, such as pollen grains, could be game-changers. In the future, nanotechnologies alone or coupled with established methods such as Agrobacterium-mediated transformation will likely be important contributors to biotechnology as well. The ever-expanding color pallet of fluorescent proteins and pigmented proteins will also likely be very useful as transformation tools and indicators of expression from the subcellular to ecological levels. Looming ahead are potentially greater regulatory hurdles and demands for safety beyond that required of traditional technologies. Thus, plant transformation technologies will be invented to serve dual purposes of increasing trait and crop value as well as biosafety. We believe that the most exciting times lie ahead in plant biotechnology as we plan for the second Plant Transformation Technologies conference held in Vienna February 2011.

We thank all the authors who contributed to the conference and the book by freely sharing their knowledge and expertise. We appreciate the work of Justin Jeffryes and his team at Blackwell for commissioning and working with us during our unpredictable schedules. We also thank Ronald D’Souza for his work at the proofing stage. The editors express special thanks to Ms. Julia Szederkenyi for her great assistance in organizing chapters, formatting and contributing to the final form as well as Jennifer Young Hinds for her help. A volume such as this could never be produced without the work of so many people that we have not acknowledged by name, but we wish to express our heartfelt thanks to each one.

Section 1

Agrobacterium-Mediated Transformation

1

Host Factors Involved in Genetic Transformation of Plant Cells by Agrobacterium

Benoît Lacroix, Adi Zaltsman, and Vitaly Citovsky

Introduction

Agrobacterium tumefaciens and several other species of the Agrobacterium genus possess the unique ability to transfer a DNA segment from a specialized plasmid (tumor inducing or Ti plasmid in the case of A. tumefaciens and hairy root inducing or Ri plasmid for Agrobacterium rhizogenes, the two main species of pathogenic Agrobacterium) into a host plant cell. This feature is widely used in plant biotechnology, and Agrobacterium is, by far, the most important tool employed to produce transgenic plants (Newell 2000). Not surprisingly, the biology of Agrobacterium and its interactions with host plant have been the subject of numerous studies in the past three decades (for recent reviews, see Gelvin 2003; Citovsky et al. 2007; Dafny-Yelin et al. 2008).

In brief, the main steps of host genetic transformation mediated by A. tumefaciens are the following. The induction of Agrobacterium’s virulence machinery results in expression and activation of the virulence genes (vir genes) (Stachel et al. 1985b, 1986; McLean et al. 1994; Turk et al. 1994; Lee et al. 1996). This first step mobilizes a single-stranded DNA segment from the Ti or Ri plasmid. This segment of transferred DNA (T-DNA), delimited by two 25-bp direct repeat sequences known as left and right borders (LB and RB) (Peralta and Ream 1985; Wang et al. 1987), is termed the T-strand, and it represents the substrate of DNA transfer to the host cell. VirD2, associated with VirD1, forms a nuclease able to excise the T-strand by a strand-replacement mechanism, at the completion of which VirD2 remains covalently linked to the 5′-end (RB) of the T-strand (Ward and Barnes 1988; Young and Nester 1988; Durrenberger et al. 1989; Pansegrau et al. 1993; Jasper et al. 1994; Scheiffele et al. 1995; Relic et al. 1998). This VirD2–T-DNA complex is then translocated into the host cell cytoplasm by a mechanism relying on the VirB/VirD4 secretion system (Zupan et al. 1998; Vergunst et al. 2000; Christie 2004). The 11 proteins encoded by the VirB operon together with the VirD4 protein form a type IV secretion system, similar to the system allowing plasmid exchange by conjugation between bacteria. The type IV secretion system consists of a protein complex, spanning Agrobacterium internal membrane, periplasm and external membrane, and of an extracellular appendage, termed the T-pilus, composed mostly of VirB2 molecules forming a hollow channel (Christie et al. 2005). The VirB/VirD4 secretion system mediates the export of the VirD2–T-DNA complex out of the bacterial cytoplasm, and likely plays a role in its entry in the host cell. This secretion system is also required for the export of several Agrobacterium virulence proteins, that is, VirD5, VirE2, VirE3, and VirF, via their C-terminal secretion signals (Vergunst et al. 2000; Schrammeijer et al. 2003; Vergunst et al. 2003; 2005; Lacroix et al. 2005). There, the T-DNA–VirD2 complex is packaged by the single-stranded DNA-binding protein VirE2 (Christie et al. 1988; Citovsky et al. 1989; Sen et al. 1989). The resulting helical structure, called the T-complex, with the help of several bacterial and host proteins, is then imported into the host cell nucleus, targeted to the host chromatin, and ultimately integrated into the host genome (reviewed in Gelvin 2003; Lacroix et al. 2006a; Citovsky et al. 2007). The native T-DNA contains genes encoding enzymes that modify growth regulators and induce uncontrolled cell proliferation, which results in neoplastic cell growths (crown galls), and proteins mediating production and secretion of opines, amino acid, and sugar phosphate derivatives, secreted by the transformed cells and utilized almost exclusively by the Agrobacterium as carbon and nitrogen source (Escobar and Dandekar 2003).

The transfer of T-DNA is not sequence-specific, and any sequence of interest can be inserted between the T-DNA borders. The ability to engineer Agrobacterium to introduce genes of interest for plant genetic transformation is the basis of Agrobacterium’s use in biotechnology. The natural host range of Agrobacterium is very large, including most of the dicotyledonous and gymnosperm families (De Cleene and De Ley 1976). However, although the number of plant species transformable by Agrobacterium under laboratory conditions is always increasing (Newell 2000), in practice, producing transgenic plants efficiently is still a challenge for many plant species. Moreover, even nonplant species can be transformed by Agrobacterium under laboratory conditions (Lacroix et al. 2006b), including yeast (Bundock et al. 1995; Piers et al. 1996), various fungi (de Groot et al. 1998; Michielse et al. 2005), and cultured human cells (Kunik et al. 2001). This chapter focuses on numerous host plant factors that play important roles in the transformation process, from the initial interactions between Agrobacterium and plant cells and the activation of Agrobacterium’s virulence, to the integration of T-DNA into the host genome.

Plant Signals Affecting Agrobacterium’s Virulence Machinery

The rhizosphere is a complex and dynamic environment, where plant-associated bacteria such as Agrobacterium need subtle regulation systems to efficiently induce their virulence machinery (Brencic and Winans 2005). Agrobacterium’s virulence depends mostly on transcriptional activation of a set of virulence (vir) genes; this regulatory system allows the integration of environmental signals to ensure a timely expression of these genes. Moreover, the induction of virulence system obviously represents a high cost in energy for the bacterial cell, and its activation must be tightly regulated to ensure that it occurs only at the proximity of a susceptible host tissue. To this end, Agrobacterium harbors sensors able to recognize signals emitted by its host plants, and to activate the virulence machinery in response to these signals. The induction of vir gene expression in Agrobacterium relies on a two-component regulatory system encoded by the virA and virG genes that respond, directly or indirectly, to different plant and environmental cues (Klee et al. 1983; Stachel and Nester 1986). virA and virG have low basal expression, but their expression is highly inducible by a self-regulated system (Winans et al. 1988). The expression of other vir genes is virtually nonexistent in absence of induction, and it is strongly enhanced when the VirA–VirG system is activated. VirA–VirG represents a two-component regulatory system, in which VirA is the membrane-spanning sensor kinase that responds to external signals and activates the response regulator VirG by phosphorylation. Phosphorylated VirG recognizes and binds to a 12-bp long specific sequence, the vir box, which is present in all vir gene promoters, and serves to activate transcription (Brencic and Winans 2005).

Several signals, from both host plants and the environment, can modulate vir gene expression (Table 1.1); these include phenolic compounds, monosaccharides, low pH, and low phosphate (McCullen and Binns 2006). Among these signals, only phenolics are absolutely required for virulence induction, whereas the other signals render Agrobacterium cells more sensitive to phenolics and/or enhance virulence induction levels.

Table 1.1 Plant and environmental signals that influence Agrobacterium virulence

Phenolic Compounds Activating Agrobacterium’s Virulence

Initially, during the analyses of plant cell exudates, a single phenolic compound, acetosyringone (3,5-dimethoxyacetophenone) was identified. It was present at elevated concentrations and able to induce vir gene expression even in the absence of the plant cells (Stachel et al. 1985a, 1986; Bolton et al. 1986). Since then, more than 80 related phenolics, including glycoside derivatives (Joubert et al. 2004), have been shown to act as vir inducers with variable efficiency (Melchers et al. 1989a; Palmer et al. 2004). These studies revealed that all vir-inducing molecules share common structural features that enable this family of chemicals to interact with bacterial receptors and to act as virulence inducers, suggesting that these molecules are recognized by a unique bacterial receptor (Lee et al. 1992). Whereas direct interaction between radioactively labeled acetosyringone and VirA has not been detected (Lee et al. 1992), genetic studies have demonstrated that phenolic inducers most likely interact directly with the linker domain of VirA, thereby activating VirA’s kinase activity (Lee et al. 1995). Indeed, the specific range of phenolic compounds recognized by different Agrobacterium strains was dependent on the virA locus, and could be transferred from one strain to another via the transfer of virA.

Reducing Monosaccharides

Sugar monomers are involved in vir gene activation in two ways: by enhancing VirA–VirG system sensitivity to phenols and by elevating the saturating concentration of phenols for virulence activation (Cangelosi et al. 1990; Shimoda et al. 1990). In addition, the range of phenolics recognized by the Agrobacterium vir gene induction system increases when monosaccharides are present as they act as coinducers (Peng et al. 1998). Several monosaccharides, such as D-glucose and D-galactose, are coinducers (Ankenbauer and Nester 1990; Shimoda et al. 1990), which share minimal structural features (i.e., the presence of a pyranose ring and acidic groups), also suggesting that they are recognized by a specific receptor. The virulence response to monosaccharides indeed relies on a chromosome-encoded factor, ChvE. This periplasmic sugar-binding protein is believed first to bind monosaccharides, then to interact with the periplasmic domain of VirA, and to enhance the VirA ability to activate vir gene expression (Cangelosi et al. 1990; Lee et al. 1992; Shimoda et al. 1993; Banta et al. 1994).

Low pH and Low Phosphate

Low pH (i.e., ∼5.7) enhances virulence activation, and this effect is mediated by VirA (Melchers et al. 1989b; Chang et al. 1996) as well as ChvE (Gao and Lynn 2005). Low pH and low concentration of phosphate (both are frequently observed in a variety of soils) activate the virG expression (Winans 1990), likely by inducing another two-component regulatory system—also required for vir gene induction—composed of ChvG and ChvI (Charles and Nester 1993).

Production of Virulence Inducers by Plant Tissues

The presence of the vir gene inducers mentioned above can be associated with some characteristics of the plant cell or tissues susceptible to Agrobacterium DNA transfer. It is well known that wounded sites of the plant tissue are particularly susceptible to Agrobacterium infection (Smith and Townsend 1907), and wounding of plant tissue is thus classically used in many Agrobacterium-mediated plant genetic transformation protocols. Consistently, wound repair is usually associated with low pH, high activity of the phenylpropanoid pathway, and presence of monosaccharides involved in cell wall modification and synthesis (Baron and Zambryski 1995), showing that the most vulnerable sites for infection are usually associated with the presence of virulence-inducing signals. Moreover, phenolic compounds are classically secreted by plant roots in the rhizosphere, along with sugars, organic acids, amino acids, and other secondary metabolites (Walker et al. 2003).

Wounding is not absolutely required for infection (Escudero et al. 1995; Brencic et al. 2005); thus, alternative pathways of Agrobacterium infection are possible. Indeed, acetosyringone was first isolated from intact tissues, such as root exudates, and plant cell culture (Stachel et al. 1985a, 1986); thus, intact plant cells may release sufficient amount of phenolic compounds for vir gene induction. In addition, several studies of the modification of plant gene expression in response to Agrobacterium contact and infection have shown that many enzymes of the phenolic metabolism, potentially involved in the production of acetosyringone and other phenolic inducers of Agrobacterium virulence, are induced on interaction with Agrobacterium (Ditt et al. 2001, 2006; Veena et al. 2003). Consistently, the phenolic metabolism is modified in response to Agrobacterium infection (Simoh et al. 2009). Interestingly, phenolic molecules are usually produced by plants as part of defense reaction, and are toxic for many bacterial pathogens; Agrobacterium likely has evolved resistance to these molecules and utilize them as signals for induction of virulence.

In addition to their most important role as vir gene inducers, phenolics and monosaccharides also trigger a chemotactic response in Agrobacterium, directing the bacterial cell to move toward a potential point of infection in the plant tissue. Chemotaxis of Agrobacterium cells toward several vir inducer phenolics is constitutive and does not require vir gene induction (Parke et al. 1987), but relies on a chromosome-encoded cluster of genes (Wright et al. 1998).

Plant-Produced Inhibitors of Bacterial Virulence

Several extracellular plant metabolites are able to inhibit Agrobacterium vir gene expression and might, together with virulence inducers, contribute to the variability of susceptibility to Agrobacterium between plant species and tissues.

Homogenates of corn seedlings have a strong inhibitory effect on both growth and acetosyringone-dependent virulence activation of A. tumefaciens (Sahi et al. 1990). The substance responsible for this inhibitory effect was identified as DIMBOA (2,4-dihydroxy-7-methoxy-2H-1,4-benzoxazin-3(4H)-one). DIMBOA, like indole acetic acid (IAA) and other auxins, is derived from the tryptophan biosynthetic pathway (Melanson et al. 1997). A similar molecule, MDIBOA (2-hydroxy-4,7-dimethoxybenzoxazin-3-one), is present at high concentration (up to 98%) in corn seedling root exudates. MDIBOA is also a potent inhibitor of Agrobacterium virulence, but has limited effect on bacterial growth (Zhang et al. 2000).

The auxin IAA itself inactivates vir gene expression by competing with the inducing phenolic compound acetosyringone for interaction with VirA (Liu and Nester 2006). In natural conditions, IAA is produced at relatively high concentrations by crown galls that develop after transformation, and is likely to inhibit new transformation.

Salicylic acid (SA) is a phenolic compound commonly produced by plants in response to many types of abiotic or biotic stress, and it is the major signal molecule of the systemic acquired resistance (SAR) in plants (Loake and Grant 2007). SA acts as an inhibitor of vir expression; most likely, SA shuts down virA and virG by attenuating the VirA protein kinase activity (Yuan et al. 2007), which would result in inhibition of expression of all vir genes. Arabidopsis mutants deficient in SA accumulation are more sensitive to Agrobacterium infection, whereas mutants overproducing SA are relatively recalcitrant (Yuan et al. 2007). Similar effects of SA on vir gene expression were observed in Nicotiana benthamiana, using either mutant plants altered in SA metabolism or exogenous application of SA (Anand et al. 2008).

The plant gaseous growth regulator ethylene was also suggested to inhibit the virulence of Agrobacterium. Indeed, plants impaired in ethylene production are more sensitive to Agrobacterium, whereas plants overproducing ethylene are more resistant (Nonaka et al. 2008b). Consistently, expression in Agrobacterium of 1-aminocyclopropane-1-carboxylate (ACC) deaminase, an enzyme that degrades ACC, the immediate precursor of ethylene in higher plants, enhances the efficiency of Agrobacterium infection (Nonaka et al. 2008a). However, although these data suggest that ethylene might inhibit Agrobacterium virulence, a direct effect of ethylene on the vir gene regulation system has not been conclusively demonstrated.

Cell-to-Cell Contact and Passage of T-DNA through Host Cell Barriers

A close cell-to-cell contact is necessary for the T-DNA transfer from Agrobacterium to its host cell. Indeed, Agrobacterium mutants impaired in their ability to attach to plant cell generally show a diminished virulence (Matthysse 1987). Putative plant and Agrobacterium proteins that mediate cellular recognition and attachment have been suggested; however, the actual nature of the factors involved remains elusive. By analogy with other plant-associated Rhizobiaceae, a two-step mechanism was proposed (Smit et al. 1992; Rodriguez-Navarro et al. 2007). First, a contact between Agrobacterium and plant cells is initiated by as yet unidentified bacterial and plant extracellular receptors; these cellular interactions are believed to be nonspecific and reversible. Second, the attachment is consolidated by cellulose fibrils synthesized by the bacterial cells (Matthysse et al. 1981; Matthysse 1983).

Initial Cellular Interactions: Is There a Plant Cell Surface Receptor for Agrobacterium?

Plant lectins (proteins that bind reversibly to mono- or oligosaccharides) could play a role in binding bacterial exopolysaccharides (Hirsch 1999), as they do in the case of other Rhizobiaceae. Indeed, A. tumefaciens mutants in chvA, chvB, and exoC (pscA) that encode enzymes involved in the synthesis of an exocellular cyclic glucan (cyclic 1,2-β-D-glucan) were deficient in virulence, likely because of impaired attachment to the plant cell (Cangelosi et al. 1989; de Iannino and Ugalde 1989). However, the specific plant receptors for recognition of exocellular glucan produced by Agrobacterium have not been identified so far.

Rhicadhesin, an extracellular protein initially isolated from Rhizobium, inhibits attachment of Rhizobium and Agrobacterium to the plant cell surface when added exogenously, likely by saturating a putative plant cell surface receptor (Smit et al. 1989). It was, thus, suggested that Agrobacterium also encodes a similar protein, which might be responsible for initial attachment to plant cells, in a Ca2+-dependent manner. However, the gene encoding an Agrobacterium rhicadhesin-like protein has not been identified, even though complete genome sequences have already become available for three Agrobacterium strains (Goodner et al. 2001; Wood et al. 2001; Slater et al. 2009). Several putative plant rhicadhesin-like receptors have been identified (Wagner and Matthysse 1992; Swart et al. 1994), but their actual functionality in Agrobacterium virulence has not been demonstrated. Because exogenous human vitronectin as well as antibodies against vitronectin inhibited binding of Agrobacterium to carrot cells, it was suggested that a vitronectin-like protein on the plant cell surface may bind bacterial rhicadhesin and thereby act as a receptor for initial attachment of Agrobacterium to the plant cell. However, recent data (Clauce-Coupel et al. 2008) demonstrated that whereas a vitronectin-like protein is present in the cell wall of plant tissues susceptible to Agrobacterium, this protein is involved neither in Agrobacterium attachment nor in its virulence. Using a bioassay based on suppression of rhicadhesin activity, a pea cell wall glycoprotein, which shows similarity to germin-like proteins present in many plant species, was also proposed to be a rhicadhesin receptor. Nevertheless, its actual interaction with rhicadhesin and its role in Agrobacterium infection have not been demonstrated.

Another series of putative plant proteins potentially involved in Agrobacterium attachment was identified using Arabidopsis insertional mutants, disrupted in genes encoding cell wall proteins. In a genetic screen for Arabidopsis mutants resistant to Agrobacterium (rat mutants) (Nam et al. 1999), several mutant lines impaired in their ability to allow Agrobacterium attachment were discovered. For example, the rat1 phenotype results from the absence of expression of AtAGP17 (Gaspar et al. 2004). Agrobacterium attachment seems to be reduced in the rat1 mutant, but the effect of the mutation might also be the result of other pathways, such as signaling or carbon allocation. rat4 is deficient in CSLA9, a homolog of cellulose synthase (Zhu et al. 2003), the activity of which could modify the properties of the plant cell surface and influence bacterial attachment.

From the bacterial side, extracellular proteins involved in virulence, such as the components of the type IV secretion system, VirB1*, VirB2, and VirB5, might play a role in initial attachment (Aly and Baron 2007; Backert et al. 2008). A search for potential plant interactors of these proteins could help understand these cellular interactions. However, it remains unknown whether VirB1*, VirB2, and/or VirB5 are required at the earlier infection step of cell–cell recognition and attachment, or they function only later, during the transfer of DNA and proteins into the host cell cytoplasm. So far, the only identified bacterial factors essential both for attachment and for virulence are chvA, chvB, and exoC, which are all involved in exocellular oligosaccharide production. The vir region seems not to be essential for attachment, whereas the att region, located in the pAt linear chromosome and initially considered to be involved in attachment (Matthysse et al. 1996; Matthysse and McMahan 1998), is not required for DNA transfer to plants, but mostly for control of quorum sensing (Nair et al. 2003).

Consolidation of Agrobacterium Attachment to Plant Cells by Cellulose Fibril Synthesis

In a second stage, the Agrobacterium–host cell interaction is consolidated by the production of cellulose fibrils by the bacterial cell, ending in irreversible binding and formation of bacterial aggregates at the plant cell surface. The mutants of Agrobacterium disrupted in the celABCDE operon were unable to form cellulose and showed a weaker attachment to plant cells as compared with wild-type bacteria (Matthysse 1983; Robertson et al. 1988). However, tumorigenicity of these mutants was only slightly reduced, but not completely blocked (Matthysse and McMahan 1998). Thus, this second step of attachment might not be absolutely necessary for T-DNA transfer, but it might be required to allow bacterial cells to remain in the vicinity of the transformed tissue (galls) and to use opines produced by the tumors. There are no known plant factors involved in binding of the bacterial cellulose fibrils on the plant cell surface.

When considering attachment of Agrobacterium cells to the host cell surface, the formation of bacterial biofilm in which bacteria are embedded appears to be essential for Agrobacterium virulence (Matthysse et al. 2005), and more generally for the virulence of many pathogenic bacteria (Danhorn and Fuqua 2007). Consistent with the ability of Agrobacterium to infect many different unrelated hosts, including nonplant species, it is uncertain whether there exists any absolutely required specific receptor(s) on the surface of the host cell; indeed, none of the putative receptors described above have ever been substantiated. Biofilm formation, which relies on the production of exocellular glucans, for example, cyclic 1,2-β-D-glucan and cellulose, could then be sufficient for the Agrobacterium’s attachment and virulence. Structural and chemical properties of the host cell surface could influence the genesis of biofilms.

Translocation of T-DNA and Virulence Proteins across the Plant Cell Wall and Plasma Membrane

The T-DNA and virulence proteins are exported from Agrobacterium via its VirB/VirD4 type IV secretion system (Ding et al. 2003; Christie 2004; Christie et al. 2005). The molecular details of T-DNA interactions with proteins of the VirB/VirD4 secretion system during transport through the bacterial membranes and periplasm were studied by coimmunoprecipitation (Cascales and Christie 2004). This study identified contacts of a T-DNA substrate with several subunits of the VirB/VirD4 system, and, using mutants in different vir genes, suggested the transport pathway for T-DNA substrate. However, this study was performed in bacteria and, thus, it provides information only about the first step of the T-DNA transfer, that is its export out of bacterial cells. The second step of the transfer process, that is, the passage of the translocated macromolecules through the host cell wall and plasma membrane, and the mechanism by which the extracellular proteins of the type IV secretion system, mainly VirB2, VirB5, and VirB7, could be involved in this process remain largely uncharacterized. During this second step of the T-DNA transfer, the T-pilus could act as a hollow needle allowing the injection of these macromolecules directly from the bacterial to the plant cytoplasm (Kado 2000), similar to how protein transport is mediated by type III secretion systems. However, the role of the T-pilus is still debated, and it could also function mainly by mechanically perforating the host cell wall and plasma membrane and allowing entry of macromolecules via another pathway (Llosa et al. 2002). Indeed, T-DNA transfer can occur in absence of detectable levels of T-pilus biogenesis; for example, the inhibition of T-pilus formation by blocking polymerization of VirB2 monomers does not abolish substrate transfer through the VirB/VirD4 type IV secretion system channel (Zhou and Christie 1997; Sagulenko et al. 2001; Jakubowski et al. 2005). Whether the VirD2–T-DNA complex and the exported virulence proteins move through the T-pilus or not, it is possible that plant factors interacting with components of the T-pilus and located in the plant cell wall, plasma membrane, and/or cytoplasm, play a role in this mechanism.

In a search for these putative receptors, four Arabidopsis proteins interacting with the processed C-terminal VirB2—that does not contain the 42-amino acid signal peptide, cleaved before T-pilus biogenesis—were identified (Hwang and Gelvin 2004). Three related proteins of unknown function, termed BTI1, 2, and 3, and a membrane-associated GTPase, AtRAB8, were found. Inhibition of expression of these proteins in Arabidopsis conferred relative resistance to Agrobacterium, whereas overexpression of BTI1 induced a hypersensitive phenotype. Although it is not clear exactly at which step these proteins might play a role, for example, during the initial attachment of Agrobacterium to the plant cell surface or later during the entry of the T-DNA, or virulence proteins into the host cell cytoplasm, they represent good candidates for host cell receptors required in the early Agrobacterium–plant cell interaction and/or macromolecule translocation.

Another possible pathway for translocation of the T-strand–VirD2 complex was suggested by the ability of the VirE2 molecule to form membrane-spanning channels, which allow passage of negatively charged macromolecules, such as oligonucleotides, in artificial lipid bilayers (Dumas et al. 2001). If this VirE2 channel also forms in plant cell membranes during the Agrobacterium–plant interaction, it may allow passage of macromolecules. Furthermore, the cooperative binding of VirE2 to the T-strand molecule during formation of the T-complex in the host cell cytoplasm may actively pull this DNA molecule, for example, out of the VirB/VirD4 and/or VirE2 channels, without the need for external energy sources (Grange et al. 2008). Although these activities of VirE2 have not been demonstrated in vivo so far, they might also involve interactions with plant factors in the cell wall or plasma membrane.

Overall, the host factors involved in macromolecular transfer between the Agrobacterium cell and the host cell cytoplasm are perhaps the least well characterized among all host factors that participate in the infection process. It is noteworthy that Agrobacterium can transfer DNA and proteins to numerous nonplant species (Lacroix et al. 2006b), suggesting a general nature of its macromolecular transfer machinery. That can be explained either by ancestral factors involved in host–pathogen interactions, which are conserved among eukaryotic organisms, or by the Agrobacterium’s ability to transport its macromolecules into host cell cytoplasm via a host-independent pathway, such as the one that does not rely on a specific host cell receptor.

Roles of Plant Factors in Transcytoplasmic Transport and Nuclear Import of the T-Complex

Structure of the T-Complex

The movement of a large DNA molecule, such as a segment of DNA of the typical size of the nopaline-type T-DNA (∼20 kilobases), is limited in the environment of the cytoplasm of a eukaryotic cell. In the cytoplasm, DNA movement could be impaired by molecular crowding and, more importantly, by electrostatic associations. This is because DNA molecules are densely charged polyanions that could interact with many cellular components (Verkman 2002). Thus, large segments of free DNA are unlikely to reach the cell nucleus by simple diffusion. Indeed, studies in mammalian cells have shown that diffusion of circular or linear plasmid DNA molecules is extremely slow in the cytoplasm, and is negatively correlated with molecule size (Leonetti et al. 1991; Lukacs et al. 2000). Moreover, the free T-strand would form a polymeric random coil in the absence of packaging proteins. Typically, a randomly coiled free single-stranded DNA corresponding to a 20-kilobase T-DNA would reach a diameter, that is, the geometric mean of its extended length and its persistence length (Briels 1986), of about 300 nm; molecules of this size are unable to move freely in the cytoplasm and are also much larger than the nuclear pore exclusion limit of about 25 nm (Dworetzky and Feldherr 1988; Forbes 1992). Furthermore, even much larger T-DNA molecules, of up to 150 kilobases, can be transferred into the cells of tobacco (Hamilton et al. 1996) and tomato (Frary and Hamilton 2001) and integrated in their genomes. Packaging into transferable forms more suited for transcytoplasmic traffic and nuclear import is obviously required for such large molecules. Consequently, the T-strand must undergo a specific spatial organization that relies on interactions with packaging proteins in order to travel to, and subsequently enter the host cell nucleus. Indeed, within the host cell, the T-strand is thought to exist in a form of a nucleoprotein complex, the T-complex (Citovsky et al. 1988, 1989; Gelvin 1998).

In the T-complex, two bacterial virulence proteins, VirD2 and VirE2, which are essential for Agrobacterium virulence (Stachel et al. 1985a), directly associate with the T-strand (Young and Nester 1988; Citovsky et al. 1989; Sen et al. 1989). The T-complex is formed in the host cell cytoplasm (Figure 1.1, step 1) after VirE2 and the T-strand with covalently attached VirD2 are translocated independently of each other from Agrobacterium to the host cell (Otten et al. 1984; Citovsky et al. 1992; Gelvin 1998; Vergunst et al. 2000). Structural analyses of artificially reconstituted T-complexes (Citovsky et al. 1997; Abu-Arish et al. 2004; Grange et al. 2008) indicated that its diameter is about 15 nm (Abu-Arish et al. 2004); this is larger than the 9 nm diffusion limit of the nuclear pore (reviewed in Forbes 1992). Overall, the size of the T-complex suggests that its transport through the host cell cytoplasm and subsequent import into the nucleus occur by active mechanisms.

Figure 1.1 Nuclear import of the T-complex and virulence proteins. (1) The exported virulence proteins VirD5, VirE2, VirE3, and VirF and the T-strand covalently attached to VirD2 and enter independently of each other into the host cell cytoplasm via the VirB/VirD2 type IV secretion system. VirE2 then associates with the T-strand, forming the T-complex. Besides packaging the T-strand into a protected form suited for transcytoplasmic and nuclear transport, VirE2 might provide the energy needed for pulling the T-strand into the host cell. (2) Plant proteins interact with the T-complex and facilitate its movement across the cytoplasm toward the nuclear pore. The transcytoplasmic movement is likely mediated by the cytoskeleton and molecular motors; the latter might interact with VIP1, which is bound to VirE2. For nuclear targeting, VirD2 binds directly to importin-α, whereas VirE2 binds to VIP1 (and VirE3), which, in turn, binds to importin-α. (3) The T-complex passes through nuclear pores, likely in a polar manner, and importins are released inside the nucleus, whereas the T-complex is targeted to the host chromatin. The biological function of VirD5 in the process of the T-DNA transfer remains unknown. (For a color version of this figure, see Plate 1.)

Roles of Molecular Motors and the Cytoskeleton in the T-Complex Movement through the Host Cell Cytoplasm

Before nuclear import can begin, the T-complex has to be transported across the cytoplasm from its point of entry and assembly to the cell nucleus. By analogy to many DNA viruses, which depend on dynein motors and microtubule networks for their transport toward the host cell nucleus, the transcytoplasmic transport of the T-complex might also represent an active process. Two lines of evidence support this notion. A plant VirE2-interacting protein 1 (VIP1) (Tzfira et al. 2001), which participates in nuclear import and intranuclear transport of the T-complex (see below), was shown to interact with the dynein-like DLC3 protein of Arabidopsis, suggesting a role for molecular motors in the T-complex movement through the cytoplasm (Tzfira 2006). That this movement might involve cytoskeletal elements is suggested by the observations that active transport of artificial T-complexes in a cell-free system occurs along the microtubule network (Salman et al. 2005). To date, the mechanism of the T-complex movement toward the host cell nucleus remains relatively unexplored and in need of more experimentation.

Nuclear Import of the T-Complex

VirD2 is a nuclear protein when synthesized in eukaryotic cells, and it directly interacts with plant importin-α, which is a part of the cell nuclear import machinery that mediates the nuclear import of VirD2 (Ballas and Citovsky 1997). VirD2 carries two nuclear localization signals (NLSs), a monopartite N-terminal NLS and a bipartite C-terminal NLS (Herrera-Estrella et al. 1990; Howard et al. 1992; Tinland et al. 1992), but only the latter is essential for its nuclear import (Howard et al. 1992; Ziemienowicz et al. 2001). Several other plant VirD2 interactors could play a role in its subcellular localization. For example, Arabidopsis cyclophilins interact with VirD2 and might assist its nuclear targeting (Deng et al. 1998). In addition, VirD2 nuclear import might be regulated by phosphorylation/dephosphorylation of VirD2 itself. An enzymatically active type 2C serine/threonine protein phosphatase from tomato was found to interact with VirD2, and its overexpression resulted in inhibition of the VirD2 nuclear import (Tao et al. 2004).

Unlike VirD2, the nuclear import of VirE2 likely requires a more complex mechanism. VirE2 nuclear import in plant cells is strongly dependent on the presence of VIP1, a plant nuclear protein with a basic leucine zipper (bZIP) motif (Tzfira et al. 2001). VIP1, via its direct interactions with VirE2 and importin-α, likely links between VirE2 and the host nuclear import machinery (Tzfira et al. 2001, 2002). Consistently, Agrobacterium-mediated transformation efficiency is positively correlated with the expression level of VIP1 (Tzfira et al. 2001, 2002).

Interestingly, the VIP1’s own nuclear import depends on its phosphorylation at a specific site (Djamei et al. 2007). This phosphorylation is mediated by the MAP kinase 3 (MPK3), an enzyme expressed as a part of a plant defense reaction that is elicited, among other factors, by Agrobacterium. MAP kinases are key factors in signal transduction during plant responses to many biotic and abiotic signals (Colcombet and Hirt 2008). It has been shown that an Arabidopsis insertional mutant in the MPK3 gene is also resistant to Agrobacterium (Djamei et al. 2007). Thus, Agrobacterium might have evolved mechanisms to subvert the host defense response, that is, induction of MPK3 and phosphorylation of VIP1, to enhance its ability to infect its host (Djamei et al. 2007).

Recently, VirE2 has been shown to interact with some isoforms of plant importin-α, particularly importin-α-4 (Bhattacharjee et al. 2008); however, it is still unclear whether this interaction is functionally important for the VirE2 nuclear import. Generally, there might exist several pathways for nuclear import of VirE2 that Agrobacterium can utilize, depending on the host species and/or physiological conditions.

While it appears that there is redundancy between the roles of VirE2 and VirD2 in mediating T-DNA nuclear import, it is more likely that, in natural conditions, an efficient polar transport of the T-complex requires both factors (Figure 1.1, step 2) (Ziemienowicz et al. 2001). Both VirD2 (Ziemienowicz et al. 1999) and VirE2 (Zupan et al. 1996; Gelvin 1998) can mediate, independently of each other, nuclear import of short single-stranded DNA segments in animal (Ziemienowicz et al. 1999) and plant cells (Zupan et al. 1996; Gelvin 1998). The most likely mechanism, which is consistent with the polar structure of the T-complex, is that VirD2, attached to the 5′-end of the T-strand, directs the T-complex to the nuclear pore, while VirE2 and the associated VIP1, which presumably are distributed along the entire length of the T-strand, assist in its movement first through the cytoplasm (Tzfira 2006) and then through the nuclear pore (Ziemienowicz et al. 2001) (Figure 1.1, steps 2 and 3).

Another Agrobacterium virulence protein translocated to plant cells, VirE3, can interact with VirE2 and importin-α and facilitate the VirE2 nuclear import, thus partially mimicking the VIP1 function (Lacroix et al. 2005). Whereas VirE3 is not essential for plant genetic transformation, it is known to act as a host range factor of Agrobacterium (Hirooka and Kado 1986), potentially compensating for the lack or low amounts of VIP1-like proteins in some plant species. This strategy of Agrobacterium to improve its infection efficiency by exporting an effector protein that mimics functionally a host factor required for infection might represent a general adaptation of infectious microorganisms, including animal pathogens (Nagai and Roy 2003). Additionally, VirE3, as a nuclear protein in plant cells (Lacroix et al. 2005), could be involved in transcriptional regulation of yet unidentified host genes (Garcia-Rodriguez et al. 2006).

Remarkably, some strains of A. rhizogenes do not possess the virE2 gene, yet are able to transfer and integrate DNA into their host genomes. In these strains, the function of VirE2 is likely fulfilled by the GALLS protein (Hodges et al. 2004, 2006, 2009) because virulence of an A. tumefaciens mutant in the virE2 gene was restored by expression of the A. rhizogenes GALLS gene in the mutant bacterial cells (Hodges et al. 2004). Whether GALLS and VirE2 function by the same molecular mechanism remains unclear. On the one hand, the full-length GALLS (Hodges et al. 2006) and VirE2 (Simone et al. 2001; Vergunst et al. 2003, 2005) both contain C-terminal signals for export from the bacterial cell through the type IV secretion system. Also, both GALLS (Hodges et al. 2006, 2009) and VirE2 (Citovsky et al. 1992, 1994; Tzfira and Citovsky 2001; Ziemienowicz et al. 2001) accumulate in the plant cell nucleus. Unlike VirE2, however, GALLS contains ATP-binding and helicase motifs (Hodges et al. 2006). The sequences of GALLS and VirE2 also do not share any homology.

Intranuclear Movement of the T-Complex and Its Uncoating

Chromatin Targeting of the T-Complex

Little is known about movement of the T-complex within the host nucleus toward the chromatin. Similarly to its transport in the cytoplasm, interactions of proteins coating the T-DNA with the host factors are likely to be involved (Figure 1.2, step 1).

Figure 1.2 The fate of the T-complex in the host nucleus. (1) The T-complex is targeted to the host chromatin by a mechanism that might rely on the VIP1–nucleosome interaction and/or by interactions with proteins that target the DNA repair machinery to DSBs in the host genome. (2) Proteins associated with the T-DNA are removed by proteasomal degradation via the SCFVirF pathway; at this stage, the T-strand is likely converted into a double-stranded form. (3) The T-DNA is integrated in the host genome by the host DNA repair machinery. (For a color version of this figure, see Plate 2.)

When discussing chromatin targeting, it is important first to understand whether this targeting aims at specific sites within the genome or it is random. Several analyses of the T-DNA integration sites have shown that T-DNA integrates randomly into the host genome (Tinland 1996; Alonso and Stepanova 2003). Other studies suggested bias toward transcriptionally active chromatin and toward the regulatory regions of genes (Barakat et al. 2000; Chen et al. 2003; Schneeberger et al. 2005). However, this apparent bias might be an artifact of selection of high-expression transgenic plants. Recovery of transgenic plants in these studies relied on the expression of reporter or selectable marker genes that favored detection of integration events in transcriptionally active chromatin regions and caused underrepresentation of integration events in regions of low transcriptional activity. Indeed, two recent studies have shown that there is no integration bias when transgenic plants are recovered without selection that is dependent on expression of the transgene (Dominguez et al. 2002; Kim et al. 2007). Consequently, the Agrobacterium T-DNA most likely integrates randomly and, thus, has access to all areas of the host chromatin.

Several plant factors could assist the targeting of the T-complex to the host chromatin. CAK2M, a conserved plant ortholog of cyclin-dependent kinase-activating kinases, was identified as an interactor of VirD2 (Bako et al. 2003). CAK2M is a nuclear protein that also interacts with the largest subunit of RNA polymerase II, which recruits TATA box-binding proteins (TBPs). VirD2 was also found tightly associated with the TBP in vivo (Bako et al. 2003). Thus, VirD2 could play a role in the T-complex chromatin targeting by associating with CAK2M and/or TBP, which, in turn, naturally associate with the chromatin.

VIP1 is another candidate for a host factor involved in chromatin targeting of the T-complex. VIP1 is a transcription factor (Djamei et al. 2007), and as such it is expected to associate with the chromatin. Indeed, VIP1 was shown to bind to all four types of purified Xenopus core histones in vitro (Loyter et al. 2005), and to at least one plant core histone, H2A, in vivo (Li et al. 2005a). It was thus suggested that VIP1 acts as a molecular link between the VirE2 component of the T-complex and the core histone component of the host chromatin. This hypothesis is consistent with the known requirement of several core histones, and particularly H2A, for the T-DNA integration (Mysore et al. 2000b; Yi et al. 2002). Recent data indicate that VIP1 can have a strong interaction with purified plant nucleosomes in vitro that can be competitively inhibited by free histone H2A. In the same experiment, VIP1 also mediated binding of free VirE2 as well as a synthetic T-complex composed of VirE2 and single-stranded DNA to nucleosomes (Lacroix et al. 2008).

Because T-DNA integration occurs preferentially into double-stranded breaks (DSBs) in the host genome (Salomon and Puchta 1998; Chilton and Que 2003; Tzfira et al. 2003) (see also below), the proteins that recognize and target the DNA repair machinery to DSBs might also assist the targeting of the T-complex. Furthermore, another plant protein, VIP2, interacting with VirE2, was found necessary for stable plant transformation, but not for transient T-DNA gene expression (Anand et al. 2007). Thus, VIP2 presumably is involved in the T-complex chromatin targeting and/or T-DNA integration, although the mechanism of this involvement is not yet understood. VIP2 is a transcriptional regulator that modifies the expression levels of many genes, including the core histones (Anand et al. 2007), and its effect on the Agrobacterium T-DNA integration may also be indirect, via altering the expression of histones.

Proteasomal Uncoating of the T-Complex

The removal (“uncoating”) of the proteins protecting the T-strand is necessary to allow the second strand synthesis, which likely occurs before integration (Chilton and Que 2003; Tzfira et al. 2003) (Figure 1.2, step 2) as well as to expose the T-DNA to the host cell DNA repair machinery that mediates the integration event (Tzfira et al. 2004a).

The first indication that proteasomal degradation may be involved in the uncoating process was provided by the presence of an F-box domain in VirF (Regensburg-Tuink and Hooykaas 1993), an Agrobacterium virulence protein exported to the host plant (Vergunst et al. 2000). In eukaryotic cells, F-box proteins represent a component of the Skp1/Cullin/F-box protein (SCF) E3 ligase complex, and they function to recognize and direct specific substrates to degradation by the 26S proteasome (reviewed in Deshaies 1999; Kipreos and Pagano 2000; Cardozo and Pagano 2004). VirF interacts with Arabidopsis Skp1-like protein 1 (ASK1), a plant homolog of the yeast Skp1 protein (Schrammeijer et al. 2001; Tzfira et al. 2004b), and both VirF and ASK1 localize in the plant cell nucleus (Tzfira et al. 2004b), where the T-complex uncoating is expected to occur. One of the cellular substrates recognized by VirF is VIP1; VirF binds VIP1 and destabilizes it in plants and in yeast cells (Tzfira et al. 2004b), which are known to be genetically transformed by Agrobacterium (Bundock et al. 1995; Piers et al. 1996). In addition, VirF, which does not bind VirE2, promotes VirE2 destabilization in the presence of VIP1 (Tzfira et al. 2004b), suggesting that VirF can destabilize the entire VIP1–VirE2 complex. In yeast, VIP1 and VirE2 destabilization by VirF is Skp1-dependent as it does not occur in an skp1-4 mutant (Connelly and Hieter 1996), indicating that this destabilization occurs via the SCFVirF pathway (Tzfira et al. 2004b). That VirF might help to uncoat the T-complex, docked at the host chromatin, is supported by the ability of VirF to associate simultaneously with purified VIP1, VirE2, single-stranded DNA, and nucleosomes in vitro (Lacroix et al. 2008). The involvement of the 26S proteasome in Agrobacterium infection is consistent with the inhibitory effect of the proteasomal inhibitor MG132 on the transformation process (Tzfira et al. 2004b).

Historically, VirF has been considered to be a bacterial host range factor (Melchers et al. 1990; Regensburg-Tuink and Hooykaas 1993). For example, VirF enhances Agrobacterium infectivity in tomato and Nicotiana glauca (Regensburg-Tuink and Hooykaas 1993), but it is not required for infection of tobacco or Arabidopsis. Thus, in plant species for which infection does not require VirF, the plant might produce proteins that have F-box protein functions that can substitute for VirF during transformation. Among several Arabidopsis F-box proteins induced by Agrobacterium infection (Ditt et al. 2006), we have identified one, designated VIP1-binding F-box protein (VBF), that binds VIP1 and promotes proteasomal destabilization of VIP1 and VIP1–VirE2 complexes in yeast and plants. Moreover, suppression of VBF expression in Arabidopsis reduced their susceptibility to Agrobacterium-induced tumor formation (Adi Zaltsman and Vitaly Citovsky, unpublished data).

T-DNA Integration into the Host Genome

Recent advances have substantially enhanced our understanding of the T-DNA integration pathways and uncovered many host factors that participate in these events (Tzfira et al. 2004a). The likely model for Agrobacterium T-DNA integration includes two major steps: first, the T-strand is converted to a double-stranded form; second, the host cell DNA repair machinery mediates the double-stranded T-DNA integration into DSBs in the host genome (Tzfira et al. 2004a).

Early studies of the T-DNA integration mechanisms focused on the role of virulence proteins accompanying the T-DNA. In particular, VirD2 was suggested to act as an integrase or a ligase because it contains an H-R-Y motif typical of the phage λ integrase (Tinland et al. 1995), and it can cleave and ligate single-stranded DNA in vitro (Pansegrau et al. 1993). However, mutations in the H-R-Y motif reduce precision of T-DNA integration, but not its efficiency (Tinland et al. 1995), and the cleavage/ligation activity was strictly sequence-specific (Pansegrau et al. 1993), which is not consistent with the direct function of VirD2 in integration. A later study revealed that, in fact, VirD2 itself does not act as a DNA ligase in vitro, suggesting that T-DNA integration is more likely to be mediated by host enzymes (Ziemienowicz et al. 2000). It cannot be ruled out, however, that VirD2 is involved in T-DNA integration by recruiting host plant factors that mediate integration.

The first proposed T-DNA integration model (Tinland 1996), named SSGR (single-stranded gap repair), was based on the sequence of a few T-DNA integration sites. In this model, T-DNA integration started by annealing of the T-DNA right border to microhomologies in the host genomic DNA, followed by synthesis of the second strand and ligation of the left border (Tinland 1996). This model was challenged by subsequent data (Tzfira et al. 2004a). On the one hand, several T-DNA integration patterns, incompatible with the SSGR model, have been discovered. Analysis of a large number of T-DNA integration sites in plant genomes revealed that microhomologies are not consistently observed at these sites (Alonso et al. 2003). Moreover, some complex integration patterns involving multiple T-DNAs, which can be integrated at the same site in direct or reverse orientation and with or without filler DNA, cannot be explained by the SSGR model (De Neve et al. 1997; De Buck et al. 1999). Specifically, the occurrence of two T-DNA molecules integrated in a head-to-head orientation is not compatible with the SSGR model because head-to-head recombination is not possible for single-stranded DNA. In addition, the presence of filler DNA cannot be explained by the SSGR model. On the other hand, increasing evidence points to a role for DSBs and the DSB repair machinery in T-DNA integration, suggesting that T-strands are converted to double-strand molecules before integration. Generation of DSBs in the plant genome by a rare-cutting DNA endonuclease resulted in higher frequencies of foreign DNA integration, in direct transformation (Salomon and Puchta 1998) as well as in