INTRODUCTION

TOP INTRODUCTION L. JAPONICUS M. TRUNCATULA G. MAX PHASEOLUS VULGARIS LITERATURE CITED

Genomic research has and will continue to revolutionize plant biology. It is clear that the adoption of Arabidopsis as a model species has done much to speed the development of plant genomics and to hasten our increased understanding of basic plant biology. However, Arabidopsis is not an "omniscient" model because this plant does not encompass all of the diverse physiological, developmental, and environmental processes seen throughout the plant kingdom. Thus, to study these other processes and to bring the genomic revolution to crop species, additional genomic resources must be developed in other plants.

Over the past several years, this realization has led to the adoption of the model species concept to the study of legumes. Unlike Arabidopsis, legumes develop important and interesting symbioses with nitrogen (N)-fixing rhizobia and with mycorrhizal fungi. They also exhibit interesting differences in secondary metabolism, pod development, and other processes that cannot be adequately modeled with Arabidopsis. The impetus for the development of legume models has come primarily from researchers interested in the rhizobium-legume symbiosis. Because of this, two models, not one, have been developed: Lotus japonicus and Medicago truncatula. In reality, these two models have evolved due to the energy of their proponents but, scientifically, they can also be justified because they exhibit two developmental systems for nodulation as well as other differences. L. japonicus forms determinate nodules, in which the root subepidermal cortical cells initiate nodule formation and a persistent, terminal nodule meristem does not develop. In contrast, M. truncatula nodules initiate from the division of inner cortical cells and continue to grow from a terminal, persistent meristem. As can be seen by the summaries below, both legume model species are now well established with a large number of laboratories involved. Therefore, in the long run, legume biology can only benefit by a comparison of the results between these models and legume crop plants.

In contrast to the effort focused on the legume models, with the possible exception of soybean (Glycine max), significantly smaller efforts exist to study the genomics of legume crop species. This is unfortunate because it remains to be seen just how much of the information developed from legume models can be directly applied to the improvement of legume crops.

It is clear that legume biology is rapidly undergoing a revolutionary transformation due to the application of genomic methods. The future is exceedingly bright, and one would expect rapid progress in our understanding of basic plant processes and the unique aspects of legume physiology and development.

The edited summaries below represent the currently funded genomic activities focused on legume models and crops. For convenience, these are listed by relevant species, although similar trends and interests are apparent throughout.

	L. JAPONICUS

TOP INTRODUCTION L. JAPONICUS M. TRUNCATULA G. MAX PHASEOLUS VULGARIS LITERATURE CITED

A Domestic Weed Goes Worldwide. Recent Progress on Lotus Research in Japan

Makoto Hayashi, Osaka University, Japan, hayashi{at}bio.eng.osaka-u.ac.jp; Toshio Aoki, Nihon University, Japan, taoki{at}brs.nihon-u.ac.jp; Sachiko Isobe, National Agricultural Research Center for Hokkaido Region, Japan, sisobe{at}affrc.go.jp; Kyuya Harada, Chiba University, Japan, haradaq{at}faculty.chiba-u.jp; Hiroshi Kouchi, National Institute of Agrobiological Sciences, Japan, kouchih{at}nias.affrc.go.jp; Kiwamu Minamisawa, Tohoku University, Japan, kiwamu{at}ige.tohoku.ac.jp; Kazuhiko Saeki, Osaka University, Japan, ksaeki{at}bio.sci.osaka-u.ac.jp; Shusei Sato, Kazusa DNA Research Institute, Japan, ssato{at}kazusa.or.jp; Satoshi Tabata, Kazusa DNA Research Institute, Japan, tabata{at}kazusa.or.jp; and Masayoshi Kawaguchi, Niigata University, Japan, masayosi{at}env.sc.niigata-u.ac.jp

This work was supported by grants from the Japan Science and Technology Corporation; the Japanese Ministry of Education, Culture, Sports, Science and Technology; the Kazusa DNA Research Institute Foundation; and the Bio-oriented Technology Research Advancement Institution.

L. japonicus was first recognized at the ancient capital of Japan, Kyoto, centuries ago. Its Japanese name, "Miyakogusa," means "capital weed," but the reason for this is not clear. It might be due to the fact that this weed was common in Kyoto, or the bright and showy color of its flowers might evoke the luxury ofthe capital city. In Japan, people used the weed as a remedy. In the 1950s, Professor Isao Hirayoshi (Kyoto University) collected L. japonicus plants growing on a riverbank in Gifu. Professor William F. Grant (McGill University, Montreal) collected its progeny as the accession B-129. In 1992, Kurt Handberg and Jens Stougaard of University of Aarhus (Denmark) obtained B-129 and established the weed as a valuable tool for modern legume research. Now, Lotus is regarded as one of the most useful plants for legume study. Four genes that condition nodulation phenotypes have already been cloned. Researchers who have interests in nodulation and other aspects of legume biology use it worldwide.

The research activity of Lotus in Japan is mostly carried out by a nonprofit organization, the Miyakogusa Consortium. Started at the end of 1999, it develops and maintains public resources essential for Lotus research: linkage maps, expression arrays (of both plant and endosymbiont genes), transformation techniques, as well as making accessions available and fostering communication among researchers. Approximately 30 laboratories all over Japan are involved in Lotus.

An advantage to researchers in Japan is that many ecotypes of Lotus can be found growing in the wild. Genotypic variety is necessary to investigate actual and potential traits of agronomic importance, such as seed yield, plant height, cold tolerance, and disease resistance, which can be identified by quantitative trait locus (QTL) mapping. Dozens of accessions have been collected, from the northernmost Hokkaido to the southernmost island Miyakojima. The L. japonicus Seed Center was recently established at the National Agricultural Research Center for Hokkaido Region (http://cryo.naro.affrc.go.jp/sakumotu/mameka/lotus-e.htm). As of October 2002, more than 60 accessions are in distribution, and almost 80 accessions are under production. Also available in the near future will be recombinant inbred lines (RILs) between "Gifu" B-129 and "Miyakojima" MG-20 (Kawaguchi, 2000), the most divergent accession of L. japonicus from "Gifu" known to date (Kawaguchi et al., 2001). The lines, established by Kazusa DNA Research Institute, consist of 221 lines at the F₈ stage. Almost one-half of the RILs will be distributed by March 2003. Using RILs, stable assessment of mapping loci and sharing of the mapping population will be possible.

Genetic transformation is a prerequisite to modern molecular biology and molecular genetics. Leguminous plants are relatively recalcitrant to transformation, although this is highly dependent on species. Since the first report of Lotus transformation, several articles have dealt with the improvement of transformation, and the technique is now readily at hand. We established a new Agrobacterium tumefaciens-mediated transformation technique for "Gifu" (Aoki et al., 2002) and "Miyakojima" with minimal contamination by non-transformants, thus largely eliminating the need for confirmation of the transgene by PCR or other time-consuming tests. This reliable transformation method is efficient enough for large-scale experiments using insertional mutagenesis or gene tagging. More than 1,300 T₁ plants of "Gifu" have been generated for gene tagging. The tagged lines were constructed using A. tumefaciens EHA101 and a tagging vector pEKB35SEXTra, which can be used for both activation tagging and exon trapping (I. Nakamura, unpublished data). Genomic DNA gel-blot analysis of arbitrarily selected T₁ plants showed that more than 60% had single copy of T-DNA, implying a relatively low frequency of multicopy insertion and genomic rearrangement. Some putative mutants show modified morphology and/or nodulation (T. Aoki, unpublished data).

Although molecular genetic studies mainly deal with the traits of whole plants, cell suspension cultures can serve as an alternative for investigating cell biology and physiology of metabolism. Cultured cell lines were established from "Gifu" and "Miyakojima" (K. Syono, unpublished data) and will be used for comprehensive profiling of metabolites in future studies. Cultured cells under various conditions will also provide sources for new cDNA libraries, which could be mined for unusual and invaluable gene transcripts.

With the aim of understanding the whole genome of Lotus, both cDNA and genome sequencing are in progress at the Kazusa DNA Research Institute. As of October 2002, 93,000 5' and 3' expressed sequence tags (ESTs) have been obtained from normalized and size-selected cDNA libraries constructed from seven different organs, such as nodules, pods, and flower buds (Asamizu et al., 2000). A total of 70,137 3' ESTs have been clustered into 20,127 nonredundant groups. The sequence data from these ESTs are available at the Web site: http://www.kazusa.or.jp/en/plant/lotus/EST/. For initiating genome sequencing, genomic clones corresponding to multiple seed points were selected using sequence information from ESTs and cDNA markers of Lotus and other legumes. As of October 2002, a total of 975 seed clones have been selected. Ninety-two of them are in the library phase, seven are being sequenced, 411 are in the finishing phase, 88 are being annotated, and 199 have been annotated. The annotated sequences are being made available on the public databases and on our Web database at http://www.kazusa.or.jp/lotus.

A large-scale cDNA macroarray was constructed using Lotus ESTs. This contains 18,144 nonredundant ESTs on a set of nylon membranes, which were selected from the EST resources (about 69,000 clones) established in the Kazusa DNA Institute. By way of example, we have analyzed comprehensive gene expression during early stages of Lotus nodule formation by means of the cDNA array. These studies detected more than 1,000 genes that are significantly up-regulated during nodulation.

The isolation of plant mutants will be required to fully elucidate the molecular mechanisms underlying plant-microbe symbioses. In our case, 33 stable mutant lines affecting nodule number and organogenesis were isolated by ethyl methanesulfonate (EMS) or ion beam mutagenesis. They include Nod, Hist, Fix, Myc, Nod²⁺, and Myc²⁺ phenotypes. We found three loci conferring an increased nodule number, i.e. astray (Ljsym77), har1 (Ljsym78), and beading nodule (bel). Among them, two genes have been cloned. The Astray gene was demonstrated to encode a bZIP protein similar to Arabidopsis HY5 that is known as a key regulator of photomorphogenesis (Nishimura et al., 2002b). On the other hand, the Har1 gene encodes a Leu-rich repeat (LRR) receptor-like kinase having the highest identity with Arabidopsis CLAVATA1 (Nishimura et al., 2002a). Sequence analysis of the Har1 ortholog in soybean revealed that the hypernodulating mutant En6500 that is allelic to nts1 has a stop codon near the transmembrane domain. These cloned genes would serve as starting points to understand light and systemic regulation of nodule development at a molecular level.

Some symbiotic mutants were isolated by means of possible somaclonal variation during tissue culture (Y. Umehara and H. Kouchi, unpublished data). Because active retrotransposons were found in the process of positional cloning of LjSYMRK, their use would facilitate the cloning of symbiotic genes. Besides nodulation, mutants affecting nyctinastic movement (sleepless), root hair (slippery root; Kawaguchi et al., 2002), and anthocyanin accumulation (viridicaulis; T. Aoki, unpublished data) were also isolated. For the positional cloning of EMS and other mutants lacking tagged genes, linkage maps were constructed between "Gifu" and "Miyakojima" (Hayashi et al., 2001). Using amplification length polymorphisms (AFLP; recombinations of EcoRI-MseI and HindIII-TaqI), simple sequence repeat (SSR), and derived cleaved amplified polymorphic sequences, almost 480 markers covering roughly 500-cM distance are mapped in both linkages. The Kazusa DNA Research Institute is now generating more SSR markers (Sato et al., 2001; Nakamura et al., 2002). This will make it possible to map and isolate mutated genes by simple PCR.

In summary, a wide spectrum of work has been initiated to develop resources for Lotus research. As a result, research on the molecular genetic, functional genomics, molecular breeding, and metabolomics of Lotus will be facilitated by the availability of linkage maps, genome sequences, ESTs, expression arrays, mutant lines, and accessions. It is now up to the choice of individual investigators to find a new edge in legume research using Lotus as a model plant.

Research Training Using Lotus japonicus. A Model Legume for Functional Genomics

Principal Investigator (PI): Michael Udvardi, Max Planck Institute of Molecular Plant Physiology, Germany, udvardi{at}mpimp-golm.mpg.de; Co-PI: Maurizio Chiurazzi, Institute of Genetics and Biophysics, Naples, Italy, chiurazz{at}iigbna.iigb.na.cnr.it; Co-PI: Panagiotis Katinakis, Agricultural University of Athens, Greece, bmbi2kap{at}auadec.aua.gr; Co-PI: Antonio Marquez, University of Seville, Spain, cabeza{at}cica.es; Co-PI: Martin Parniske, John Innes Centre, UK, martin.parniske{at}bbsrc.ac.uk; Co-PI: Gerhard Saalbach, Risoe National Laboratory, Denmark, g.saalbach{at}risoe.dk; Co-PI: Herman Spaink, Leiden University, The Netherlands, spaink{at}rulbim.leidenuniv.nl; Co-PI: Jens Stougaard, University of Aarhus, Denmark, stougaard{at}mbio.aau.dk; and Co-PI: Judith Webb, Institute of Grassland and Environmental Research (IGER), UK, judith.webb{at}bbsrc.ac.uk)

This work was supported by the European Union (FP5 Project No. HPRN-CT-2000-00086; http://improving.cordis.lu/rtn/).

Beneficial plant-microbe interactions are extremely important to agriculture and the ecology of our planet. Root symbioses between plants (specifically legumes) and bacteria of the family Rhizobiaceae (called simply rhizobia), and between plants and arbuscular mycorrhizal (AM) fungi are perhaps the most important of all such interactions. N-fixing symbioses between legumes and rhizobia enable the plants to grow in the absence of fertilizer N. AM symbioses, on the other hand, often play a crucial role in the phosphorous (P) nutrition of plants.

N-fixing and AM symbioses have played an important role in sustainable agricultural systems for hundreds, if not thousands, of years and have the potential to play an even greater role in the future. Realization of this potential requires further fundamental research on these symbioses. The legume L. japonicus is a valuable model species for symbiosis research because it has a relatively small diploid genome, it is self-fertile, and it can be transformed easily. Thus, it is amenable to forward and reverse genetics, genomics, and functional genomics.

The Lotus project is an international, multidisciplinary effort funded by the European Union to promote research training at the cutting edge of plant science. The two principal research objectives of the project are to develop resources for functional genomics of L. japonicus, and to use these resources to understand better how N-fixing and AM symbioses develop and how they function to provide plants with N, P, and other nutrients. With respect to the first objective, a number of essential public resources are being developed. These include: large populations of genetically tagged and untagged mutants; a high-resolution genetic map of the L. japonicus genome; large libraries of root and nodule ESTs; facilities for high-throughput analysis of mRNA, protein, and metabolites; standardized protocols for growth and physiological analysis of plants; advanced microscopy protocols for cell biology; and, finally, capabilities to collect, store, analyze, and distribute large amounts of data. With respect to the second objective, work will focus on the identification of genes and signals involved in development of N-fixing nodules and AMs, as well as on metabolism and transport in nodules.

There are several highlights of progress after 2 years. To facilitate transcriptome and classical molecular/cell biological studies in Lotus, approximately 10,000 ESTs have been obtained from rhizobium-infected roots and mature nodules of Lotus. Using bioinformatics, we have ascribed putative functions to the proteins encoded by many of these genes, and classified them into various functional categories including putative signaling proteins, putative transcription factors, enzymes involved in primary and secondary metabolism, and many different types of transporters.

Full-length cDNAs encoding enzymes of carbon metabolism (phosphoenolpyruvate carboxylase, Suc synthase, Suc transporters, trehalose phosphatase, and carbonic anhydrase-, -, and - type) and N metabolism (Asn synthetase, Gln synthetase, Asp aminotransferase, Orn decarboxylase, Arg decarboxylase, and Glu decarboxylase) were identified in our EST database, and RNA in situ hybridization analysis was used to localize expression of these genes in nodules. Immunolocalization of phosphoenolpyruvate carboxylase and carbonic anhydrase (-type) protein confirmed the gene expression data. This information will contribute to a better understanding of how nodule metabolism is organized and regulated.

Genes that are up-regulated during nodule or AM development in Lotus, and which may be essential for these processes, were identified by two complementary approaches. The first approach utilized DNA arrays, produced by spotting 2,000 EST clones onto nylon membranes, to identify genes that were differentially expressed in nodules compared with roots. In this way, 83 genes were identified that may play important roles in nodule development or function of mature nodules (Colebatch et al., 2002). Among these were genes for primary C and N metabolism, metabolite transport, hormone metabolism, cell wall biosynthesis, signal transduction, and regulation of transcription. The size of arrays was recently increased to 5,000 clones, and this will be increased to over 10,000 in the near future, using additional clones obtained from Japanese colleagues (see Hayashi et al., above). The second approach employed cDNA-AFLP analysis to identify genes that were induced or repressed in roots shortly after inoculation with Mesorhizobium loti or Glomus intraradices. In total, 1,200 differentially regulated cDNA fragments were isolated. Research has been focused on those genes that are induced by both microsymbionts ("symbiosins"). RNA interference (RNAi) construct design for silencing of selected symbiosins has been initiated. Some of these genes may play regulatory roles during early stages of symbiosis. Differentially expressed genes will be used to produce a temporal map of the molecular events that occur during symbiosis development. This map will be useful in establishing a hierarchy of mutants affected in nodulation and/or AM development.

To accelerate reverse genetics in Lotus, we have improved methods for Lotus transformation. An optimized, in vitro transformation regeneration protocol using root explants has been developed that increases transformation and regeneration efficiencies and decreases the plant regeneration time. Other transformation techniques are being tested to find the most time- and labor-efficient method for generation of transgenic plants/roots.

To facilitate forward genetics in Lotus, populations of transposon and T-DNA insertion mutants are being developed. To create Ds insertion mutants in L. japonicus, two gene trap constructs and two activation-tagging constructs were transferred into Lotus by Agrobacterium transformation. At present, 200 lines are being raised for seed production. From this material, selection of double resistant lines was initiated for isolation of Ds lines for nodulation mutant screening. An additional 800 transformed calli are going through the regeneration procedure and will be added to the collection of Ds launching lines. A large collection of independent transformants obtained with two different promoterless reporter gene T-DNA constructs is also in preparation. Most recently, an active retrotransposon, LORE1, has been found in Lotus, and simple conditions for activation of this element are being investigated. Activation of this endogenous element by a controllable environmental condition would facilitate efficient tagging procedures.

Using polymorphic markers (AFLP, RFLP, and sequence/gene-specific PCR), the genetic map of Lotus has been developed into a very effective tool for map-based cloning of symbiotic genes and other genes of interest. We have developed a series of codominant sequence-known and gene-based markers that resulted in a consolidated genetic map of L. japonicus (Sandal et al., 2002). This collaboration has already facilitated the map-based cloning of a gene indispensable for root symbioses (Stracke et al., 2002) and another that controls nodule number (Krusell et al., 2002). Cloning of three other symbiotic loci involved in Nod-factor signal perception or immediate downstream signal transduction is in a very advanced stage.

A proteomics approach has been taken to identify proteins at the symbiotic interface in N-fixing nodules. A method to isolate the peribacteroid membrane from Lotus nodules was developed, which utilizes aqueous two-phase partitioning of membrane fractions. Using this method together with mass spectrometry (MS), several novel putative PBM proteins have been identified, including a sulfate transporter that matches an EST, which showed nodule-induced expression pattern on DNA arrays (Weinkoop and Saalbach, 2003).

Metabolite profiling, using gas chromatography (GC)-MS, has also commenced in two of our groups and has revealed quantitative changes in flavonoids after mycorrhizal infection. Comparisons were made of wild-type leaf, stem, and root tissues of L. corniculatus and L. japonicus, using HPLC-photodiode array, HPLC-photodiode array/MS and GC/MS. This comparison showed that although shoot profiles are similar, the roots of these two species are different. GC-MS analysis has also been used to profile changes in metabolites in nodules of mutant, non-N-fixing plants and wild-type nodules containing mutant rhizobia.

In summary, the Lotus consortium has made significant progress in legume functional genomics. The tools for rapid map-based cloning of genes have been developed to the point that Lotus is now a premier model legume for forward genetics. As a result, members of the Lotus consortium have been among the first to identify several genes that are essential for beneficial symbiosis in plants. The state-of-the-art of legume reverse genetics has also been advanced by the Lotus project, which has developed insertion mutant populations, protocols for RNAi suppression, or overexpression of genes in Lotus, and a large population of EMS mutants and facilities for TILLING. Advances have also been made in transcriptome, proteome, and metabolome analysis for legume research.

A TILLING Reverse Genetics Tool for L. japonicus

PI: Martin Parniske, The Sainsbury Laboratory, UK, martin.parniske{at}bbsrc.ac.uk; Co-PI: Trevor Wang, John Innes Centre, UK, trevor.wang{at}bbsrc.ac.uk; Co-Investigators: Jillian Perry, Sarah Gardener, and Jodie Pike, The Sainsbury Laboratory, UK; and Tracey Welham, John Innes Centre, UK

This work was supported by the Biotechnology and Biological Science Research Council (grant no. 83/D15167), by the John Innes Centre (two rapid response interdepartmental research grants), and by the Gatsby Charitable foundation (http://www.lotusjaponicus.org; to The Sainsbury Laboratory).

A strategy for reverse genetics that is based on EMS-mutagenesis was first described by McCallum et al.(2000) using the acronym TILLING (targeted induced local lesions in genomes). A specific advantage of EMS mutagenesis is that the series of allelic mutations can serve as the basis of detailed structure-function studies. In addition, this has the potential to recover weak alleles with subtle changes in functionality of genes that would be lethal when more strongly affected. TILLING identifies individuals carrying point mutations in any gene of interest within a large population of EMS-mutagenized plants.

We have established a TILLING reverse genetics tool for the legume L. japonicus with the objective of establishing a resource for the scientific community. The methods and early results of this endeavor are described in detail in this issue (Perry et al., 2003). Taking into account that root symbiosis is a large and legume-specific area of interest, we generated two populations directly accessible for TILLING: a general TILLING population, biased against the occurrence of severe developmental phenotypes, and a series of smaller populations of siblings exhibiting defects in the root nodule symbiosis. For the general TILLING population, our aim was to include fertile individuals only, so that progeny of a plant carrying a mutant allele can be directly recovered from progeny of the TILLed plant.

Within a population of preselected symbiotic mutants, a series of functionally impaired alleles of the SYMRK gene could be identified. This gene is required for the formation of root symbioses (Stracke et al., 2002). Only a fraction of these alleles were represented in the corresponding siblings of the general TILLING population. We concluded that in the case where a gene of interest has already been implicated in a particular biological process, the inclusion of a forward screen for that particular trait can increase the frequency at which functionally affected alleles can be recovered.

To cover research interest in other aspects of legume biology, mutant siblings were isolated that exhibited abnormal root branching patterns, abnormal growth habit (dwarf and stature mutants), abnormal leaf or flower development, or were affected either in starch synthesis or breakdown. The mutant phenotypes including photographs were entered into a Web-accessible database (www.lotusjaponicus.org/finder.htm). We have collected seed from these developmental mutants, and trait-specific TILLING populations could be set up on demand.

Plant-Insect Interactions as a Response to Metabolic Engineering of Natural Product Synthesis Studied by Functional Genomics in L. japonicus

PI: Søren Bak, Royal Veterinary and Agricultural University, Denmark, bak{at}kvl.dk; Co-PI: Karin Forslund, Royal Veterinary and Agricultural University, Denmark, kaf{at}kvl.dk; Co-PI: Birger Lindberg Møller, Royal Veterinary and Agricultural University, Denmark, blm{at}kvl.dk; Co-PI: Bodil Jørgensen, Danish Institute of Agricultural Sciences, Denmark, b.jorgensen{at}dias.kvl.dk; International Collaborators: David Galbraith, University of Arizona, Tucson, galbraith{at}arizona.edu, Kazusa DNA Research Institute, Japan, lotus{at}kazusa.or.jp; Michel Udvardi, Max-Planck-Institut, Golm, Germany, udvardi{at}mpimpgolm.mpg.de; and Clas M. Naumann, Leibniz Institute for Research in Terrestrial Biology, Germany, c.naumann.zfmk{at}uni-bonn.de

This work was supported by the Danish National Research Foundation (grant to the Center for Molecular Plant Physiology, [PlaCe]) and by the Danish Agricultural and Veterinary Research Council (http://www.place.kvl.dk; grant no. 23-02-0095).

We are introducing L. japonicus as a genetic model system to study cyanogenic glucosides. Cyanogenic glucosides are -glucosides of -hydroxynitriles and are classified as phytoanticipins. Upon disruption of plant tissue containing cyanogenic glucosides, these are degraded by -glucosidases and -hydroxynitrilases, resulting in the release of toxic hydrogen cyanide as well as of Glc and an aldehyde or ketone. This binary systemtwo sets of components that separately are chemically inertprovides plants with an immediate chemical defense response to herbivores and pathogens that cause tissue damage (Møller and Seigler, 1999). To study the specific effect of cyanogenic glucosides on plant-insect interactions, we have previously transferred the entire pathway for synthesis of the Tyr-derived cyanogenic glucoside dhurrin into the model plant Arabidopsis using the sorghum (Sorghum bicolor) genes CYP79A1, CYP71E1, and UGT85B. In free-choice tests, the crucifer specialist flea beetle (Phyllotreta nemorum) was found to avoid the dhurrin-containing Arabidopsis plants (Tattersall et al., 2001). Arabidopsis belongs to the Brassicaceae, which do not contain cyanogenic glucosides but produce a related group of amino acid-derived natural products classified as glucosinolates. The two pathways have aldoximes as common intermediates and, in accordance, the introduction of the Tyr-derived cyanogenic glucoside dhurrin also results in simultaneous accumulation of a new Tyr-derived glucosinolate, p-hydroxybenzyl glucosinolate (Bak et al., 1999, 2000). Furthermore, the level of endogenous -glucoside activity available to hydrolyze dhurrin is low in Arabidopsis, rendering cyanide release slow (Tattersall et al., 2001). To overcome these experimental problems, we are introducing L. japonicus as a new experimental model system, and taking a functional genomics approach.

L. japonicus contains the two cyanogenic glucosides, linamarin and lotaustralin, derived from Val and Ile, respectively. Lotaustralin constitutes the major glucoside. The -glucosidase activity is high, causing rapid cyanide release upon tissue damage. In collaboration with the Kazusa DNA Research Institute (http:www.kazusa.or.jp/lotus/), the L. japonicus genome has been found to contain two CYP79 orthologs assigned as CYP79D3 and CYP79D4, which show different expression patterns at the tissue level. Bioinformatic approaches are currently being used to identify the L. japonicus genes orthologous to S. bicolor CYP71E1 and UGT85B1 (Paquette et al., 2000, 2003). Promoter fusion constructs of key regulatory enzymes in biosynthesis and degradation of cyanogenic glucosides will be generated in collaboration with Dr. David Galbraith to delineate the tissue-specific expression pattern in planta and facilitate in vivo expression studies after abiotic and biotic stress.

Metabolic engineering of cyanogenic glucoside synthesis in L. japonicus proceeds following two parallel approaches. Transgenic plants overexpressing pathway enzymes from cassava (Manihot esculenta Crantz.; Andersen et al., 2000) or sorghum are being analyzed for altered patterns of glucoside accumulation. Second, L. japonicus genes involved in cyanogenic glucoside synthesis or degradation will be silenced by either RNAi (posttranslational gene silencing), or mutants will be isolated by screening the collection of TILLING mutants available at The Sainsbury Laboratory (see Parniske et al., above; Perry et al., 2003). Metabolite profiling based on HPLC/MS/MS has been established. Transcriptome analyses are being carried out using DNA macroarrays provided by Dr. Michael Udvardi, according to previously established protocols (Xu et al., 2001).

L. japonicus has co-evolved with Zygaenae moths. Members of the Zygaenae sequester cyanogenic glucosides in special glands and utilize them in defense against its predators. Together with Clas Naumann, who is able to rear Zygaena trifolii, we will investigate interplay between Z. trifolii and the transgenic L. japonicus plants engineered to have no or different cyanogenic glucoside profiles or to be unable to degrade such glucosides. The L. japonicus plants expressing promoter fusions of the key regulatory enzymes in biosynthesis, and degradation of cyanogenic glucosides will be included in these studies. The plant/insect studies will be carried out with additional insects with a focus on insects for which DNA macroarray chips are available. In this way, the chemical warfare between plants and insects can be followed at the transcriptional level by transcriptome analyses and by promoter reporter gene fusions, as well as through metabolite profiling, thereby providing a detailed understanding of the relative importance of complete metabolism, detoxification, and sequestering.

Cytochromes P450, UDPG-glycosyltransferases, and -glucosidases are key biocatalysts in the formation and degradation of natural products in plants (Vogt and Jones, 2000) and enable recruitment of new functions to maintain competitiveness in the chemical warfare toward herbivores and pathogens (Paquette et al., 2003). We will annotate L. japonicus genes from these three multigene families and include them on our bioinformatics Web site (The Arabidopsis P450, cytochrome b5, P450 reductase, and Glycosyltransferase Site at Center for Molecular Plant Physiology (PlaCe) at http://www.biobase.dk/P450/) to ensure correct annotation and naming according to the nomenclature rules. We will carry out comparative genomic phylogenetic analyses of cytochrome P450s and UDPG-glycosyltransferase from green algae, Lotus, and Arabidopsis to obtain a better understanding of the expansion of these two multigene families during plant evolution.

Identifying Symbiotic Genes in Model and Crop Legumes

Co-PI: K. Judith Webb, IGER, Aberystwyth, UK, judith.webb{at}bbsrc.ac.uk. Co-PI: Leif Skøt, IGER, leif.skot{at}bbsrc.ac.uk; and Co-PI: William Eason, IGER, william.eason{at}bbsrc.ac.uk

This work was supported by the Biotechnology and Biological Sciences Research Council, UK (Competitive Strategic Grant).

We aim to increase understanding of the genetic basis of interactions between legumes and microorganisms that affect plant health and performance. We are exploiting a model legume, L. japonicus, and a crop legume, white clover (Trifolium repens), to study both rhizobium bacteria (free-living symbionts) and AM fungi (AMF; obligate symbionts). This project has two main aims: (a) to identify and analyze plant genes involved in early recognition events in interactions of legumes with their rhizobium and AMF symbionts, and (b) to identify and analyze plant genes involved in efficient functioning in rhizobium and AMF symbioses.

Gene expression analysis has provided evidence of similarities between nodulation and mycorrhizal colonization. We have created and are exploiting populations of mutants (including Nod and Myc; Bonfante et al., 2000; Novero et al., 2002) and genetically tagged transformants (promoterless-GUS) of the model L. japonicus to dissect common early stages in development of both symbioses. We have identified putative promoter regions and genes from these promoter-trapped plants (Webb et al., 2000). In one of these lines, -glucuronidase (GUS) is expressed in mature and senescing nodules, whereas in another, GUS expression is apparent soon after challenge with rhizobium. This line is providing a tool for further analysis of symbiotic signaling. We have also identified plants expressing GUS in other tissues, such as embryos, from the population of 284 independently tagged lines.

We are also exploiting unique material generated at IGER: near-isogenic lines (NILs) of white clover that show phenotypic differences in plant response to AM infection (Eason et al., 2001). We have isolated functional phenotypes of NILs of white clover, which vary in AMF colonization and effectiveness. We have identified potential differences in gene expression in both leaves and roots, using differential display. Some of these sequences have homology to EST sequences reported by other researchers working with AMF. We plan to generate ESTs from these white clover lines, targeting NILs with contrasting phenotypes and specific cells involved in the symbiotic interaction, to yield more precise genetic information on establishment and maintenance of successful symbioses.

	M. TRUNCATULA

TOP INTRODUCTION L. JAPONICUS M. TRUNCATULA G. MAX PHASEOLUS VULGARIS LITERATURE CITED

Toward the Complete Gene Inventory and Function of the M. truncatula Genome

PI: Douglas Cook, University of California, Davis, drcook{at}ucdavis.edu; Co-PI: Steve Gantt, University of Minnesota, steve{at}cbs.umn.edu; Co-PI: Michael G. Hahn, University of Georgia, hahn{at}ccrc.uga.edu; Co-PI: Maria Harrison, The Samuel Roberts Noble Foundation (SRNF), harrison{at}noble.org; Co-PI: Dongjin Kim, University of California, Davis, djim{at}ucdavis.edu; Co-PI: Ernest Retzel, University of Minnesota, ernest{at}mail.ahc.umn.edu; Co-PI: Deborah Samac, U.S. Department of Agriculture (USDA) and University of Minnesota, dasamac{at}tc.umn.edu; Co-PI: Christopher Town, The Institute for Genomics Research (TIGR), cdtown{at}tigr.org; Co-PI: Kathryn A. VandenBosch, University of Minnesota, kvandenb{at}cbs.umn.edu; Co-PI: Carroll Vance, USDA and University of Minnesota, vance004{at}maroon.tc.umn.edu; Co-PI: Nevin Young, University of Minnesota, neviny{at}tc.umn.edu; Collaborators: David Bird, North Carolina State University, david_bird{at}ncsu.edu; Julia Frugoli, Clemson University, jfrogol{at}clemson.edu; Michael Grusak, USDA-Agricultural Research Service (ARS) Children's Nutrition Research Center, mgrusak{at}bcm.tmc.edu; and Gary Stacey, University of Missouri, staceyg{at}missouri.edu

This work was supported by the National Science Foundation (NSF; Plant Genome Project no. DBI-0110206, a continuation of DBI-0196179; http://www.medicago.org/).

This project involves the large-scale genomic analysis of M. truncatula, a model legume with a small genome for efficient molecular, genetic, and reverse genetic analyses (Barker et al., 1990; Cook, 1999). This multiinstitutional U.S. effort is part of an international collaboration to develop a complete inventory and functional analysis of the Medicago genome. The emphases of this project include: (a) construction of genetic and physical maps (involving participants Douglas Cook, Dongji Kim, and Nevin Young), (b) analysis of gene function in legume biology (Steve Gantt, Michael G. Hahn, Maria Harrison, Deborah Samac, Christopher Town, Carroll Vance, Kathryn A. VandenBosch, Nevin Young, and collaborators), and (c) analysis and public distribution of data (Ernest Retzel and Christopher Town). The results are expected to accelerate the discovery of agronomically important genes, in both Medicago and crop legumes, and to enhance understanding of gene and genome evolution within the Leguminosae.

A Sequence-Based M. truncatula Genetic Map Facilitates Comparisons between Species

M. truncatula Resistance Gene Evolution and Genomic Organization

Development of a Medicago Physical Map to Aid Positional Cloning and Full Genome Sequencing

Analysis of Medicago Gene Function during Interactions with Microbes

Sifting for Novel Expressed Sequences among Legume ESTs

The Institut National de la Recherche Agronomique (INRA) Project. Genetics and Genomics of the Model Legume M. truncatula

Coordinator: Jean Dénarié, Centre National de la Recherche Scientifique (CNRS)-INRA, Castanet-Tolosan, France, denarie{at}toulouse.inra.fr; PI: J.M. Prosperi, INRA, Montpellier, France, prosperi{at}ensam.inra.fr; PI: T. Huguet, CNRS-INRA, Castanet-Tolosan, France, thuguet{at}toulouse.inra.fr; PI: C. Rameau, INRA, Versailles, France, rameau{at}versailles.inra.fr; PI: P. Gamas, CNRS-INRA, Castanet-Tolosan, France, gamas{at}toulouse.inra.fr; PI: J. Gouzy, CNRS-INRA, Castanet-Tolosan, France, gouzy{at}toulouse.inra.fr; PI: B. Tivoli, INRA, Rennes, France, tivoli{at}rennes.inra.fr; and PI: R. Thompson, INRA, Dijon, France, thompson{at}epoisses.inra.fr

This work was supported by INRA (AIP 242; http://medicago.toulouse.inra.fr/ATS/).

This project has two main objectives: (a) to contribute to the development of M. truncatula genetic and genomic resources, and (b) to initiate studies to prepare for transfer of genetic and genomic information collected on this model species to cultivated legumes important in France such as pea and alfalfa. The project involves 17 laboratories, will last 3 years (2001-2003), and is organized in the following areas.

Genetic Resources and Development of Genetic Maps

Functional Genomics

The Integrated Structural, Functional, and Comparative Genomics of the Model Legume M. truncatula

Coordinator: Jean Dénarié, INRA-CNRS, Toulouse, France, denarie{at}toulouse.inra.fr; Vivienne Gianinazzi-Pearson, INRA-CMSE, Dijon, France, gianina{at}epoisses.inra.fr; Alfred Puehler, Universität Bielefeld, Germany, puehler{at}genetic.uni-bielefeld.de; Helge Kuester, Universität Bielefeld, Germany, helge.kuester{at}genetik.uni-bielefeld.de; Philipp Franken, Max Planck Institute for Terrestrial Microbiology, Germany, frankenp{at}mailer.uni-marburg.de; Jörn Kalinowski, Universität Bielefeld, Germany, joern.kalinowski{at}genetic.uni-bielefeld.de; Adam Kondorosi, CNRS, Gif-sur-Yvette, France, kondorosi{at}isv.cnrs-gif.fr; Michael Schultze, University of York, UK, ms47{at}york.ac.uk; Noel Ellis, John Innes Centre, UK, noel.ellis{at}bbsrc.ac.uk; György Kiss, Biological Research Center of the Hungarian Academy of Sciences, Hungary, kgb{at}nucleus.szbk.u-szeged.hu; Ton Bisseling, Wageningen University, The Netherlands, ton.bisseling{at}wur.nl; and Anne Schneider, European Association for Grain Legume Research, France, a.schneider-aep{at}prolea.com

This is project no. QLG2-CT-2000-00676 of the European Commission Fifth Framework Program (http://medicago.toulouse.inra.fr/EU/).

This large consortium includes 11 participating sites in five European countries (France, Germany, Hungary, The Netherlands, and UK). The principal investigators at the project's 11 sites are listed above. The diverse expertise of the participants allows a comprehensive approach to genomics of M. truncatula.

Comparative Analysis of Legume Genome Structure

Functional Genomics

Center for Medicago Genomics Research

Coordinator: Gregory D. May, Samuel Roberts Noble Foundation (SRNF), gdmay{at}noble.org; Principal Investigators: Lloyd W. Sumner, SRNF, lwsumner{at}noble.org; Richard A. Dixon, SRNF, radixon{at}noble.org; Richard S. Nelson, SRNF, rsnelson{at}noble.org; Maria J. Harrison, SRNF, mjharrison{at}noble.org; Robert A Gonzales, SRNF, ragonzales{at}noble.org; Liangjiang Wang, SRNF, lwang{at}noble.org; Xiaoqiang Wang, SRNF, xwang{at}noble.org; Nancy L. Paiva, SRNF, nlpaiva{at}noble.org; Kiran Mysore, SRNF, ksmysore{at}noble.org; Rujin Chen, SNRF, rchen{at}noble.org; and Elison Blancaflor, SRNF, eblancaflor{at}noble.org

This work was supported by the SRNF (http://www.noble.org/medicago/index.htm).

A Center for Medicago Genomics Research was established at the SRNF in the fall of 1999. We have taken a global approach in the study of the genetic and biochemical events associated with the growth, development, and environmental interactions of M. truncatula. Our approach includes large-scale EST sequencing, gene expression profiling, and high-throughput metabolite and protein profiling. We are interfacing these multidisciplinary data types to provide an integrated set of tools to address fundamental questions pertaining to legume biology. These questions include the analysis and understanding of: (a) the biosynthesis of natural products that affect forage quality and human health, (b) the cellular and molecular basis for the directional growth response of roots to gravitropism and the role of the cytoskeleton in this process, (c) legume root development and elucidating molecular mechanisms of polar auxin transport, (d) non-host pathogen resistance, (e) the RNA silencing pathway, and (f) the use of M. truncatula in combination with an AM fungus G. versiforme for analyses of the AM symbiosis.

The Medicago Genome Initiative, established at the National Center for Genome Resources, is a database of EST sequences of the model legume M. truncatula (Bell et al., 2001). From January 2000 to December 2002, almost 100,000 M. truncatula ESTs have been characterized at SRNF, and a total of approximately 175,000 have been characterized worldwide. Unidirectional cDNA libraries representing different stages of M. truncatula development and exposure to biotic and abiotic stresses have been generated. The international Medicago research community has characterized ESTs from more than 24 different cDNA libraries. The goal of the SRNF's EST project is to identify and characterize 20,000 to 40,000 unique Medicago cDNA isolates.

Changes in gene expression underlie many biological phenomena. The use of DNA microarrays and serial analysis of gene expression (SAGE) will provide insights into tissue- and developmental-specific expression of genes and the response of gene expression to environmental stimuli. Qiagen Operon, in collaboration with SRNF, Chris Town (The Institute for Genomic Research), and Kathryn VandenBosch (University of Minnesota), are developing a commercially available Array Ready Genome Oligonucleotide Set for M. truncatula. This set of 16,000 bioinformatically optimized oligonucleotides will be used as probes in our microarray analysis, and will provide a uniform platform for gene expression analysis around the globe.

The protein complement of the genome, the proteome, serves as a biological counterpart to the Medicago EST and gene expression analyses. Given that many biological phenomena lack the requirement for de novo gene transcription, proteomics studies provide a mechanism to study proteins and their modifications under developmental changes and in response to environmental stimuli. An automated system has been established for the electrophoretic separation of complex protein mixtures and differential analysis to discover changes in proteome content (Asirvatham et al., 2002; Watson et al., 2003).

A state-of-the-art biological MS laboratory has been established as part of the Medicago genomics activities (Sumner et al., 2002). Instrumentation within the laboratory included liquid chromatography/MS, GC/MS, and matrix-assisted laser-desorption ionization time of flight MS. M. truncatula ecotypes and elicited cell cultures will be screened for changes in the levels of a wide range of primary and secondary metabolites.

Our aim is to develop a program that will integrate gene expression and protein and metabolite profiling in conjunction with M. truncatula genetics to provide a global view of Medicago biology. Medicago resources are publicly available through the SRNF.

BAC-Based Sequencing of the Gene-Rich Euchromatic Regions of M. truncatula

PI: Bruce A. Roe, The University of Oklahoma, broe{at}ou.edu

This work was supported by the SRNF (http://www.genome.ou.edu).

To better understand biological processes associated with legumes, as a pilot project we initially cloned and end sequenced 25,000 random 2-to 4-kb whole-genome shotgun fragments. Although this data represented only 2% of the approximately 500 million-bp M. truncatula genome, approximately 1,000 individual sequence reads match the M. truncatula EST database, including an approximately 20-kb contig matching 26S rRNA genes and several classes of extensive repeated sequences representing between 60% and 80% of the genome. In addition, we now have completed the 124,034-bp sequence of the M, truncatula chloroplast genome, which has been deposited into GenBank (accession no. AC093544).

Funding from the SRNF is allowing us to obtain the working draft (4-6-fold shotgun coverage) of mapped BACs that are provided to us by Douglas Cook and Dongjin Kim (University of California). To date, we have approximately 480 BACs in our sequencing pipeline and anticipate that by September 2003, we will have obtained the working draft sequence of an additional 480 BACs representing approximately 10% of the M. truncatula genome and over one-half of the gene coding regions in the genome. By September 2003, we also propose to have sequenced approximately one-half of the approximately 960 BACs to phase 2, i.e. ordered and oriented contigs.

Our present sequencing strategy entails random shotgun cloning of isolated and mapped BAC clones into pUC vectors, followed by colony picking on a Q-pick system and cell growth in a 384-well format in HiGrow shaker-incubators. The template isolations now are fully automated and performed on a Zymark 384 well pipetting SiClone robot with Twister II arm designed with four 384-well plate shakers, a newly designed, key component in our work flow. The cycle sequencing reactions with non-fluorescent universal forward and reverse primers, and fluorescent-labeled terminators are pipetted on a Velocity 11 V-Prep and incubated in 384-well format ABI Viper thermocyclers. After electrophoresis and data collection on the ABI 3700 and data transfer, the individual contigs assembled by Phrap are analyzed by several automated scripts to predict efficient closure strategies. An Oracle database has been developed to facilitate automated data analysis and preliminary sequence annotation. Implementation of these robotic procedures and automated data analysis protocols has reduced the cost of finished DNA sequence to less than 10 cents/base while improving the overall efficiency from cells to sequence to over 95%.

In accord with the Bermuda Rules for data access, all our sequence data is being deposited into GenBank within 24 h after it is assembled into contigs greater than 2 kb. A genome browser showing the results of our automated annotation pipeline is available from links on our Web site (http://www.genome.ou.edu).

An Integrated Approach to Functional Genomics and Bioinformatics in a Model Legume

PI: Pedro Mendes, Virginia Bioinformatics Institute, mendes{at}vt.edu; Co-PI: Richard A. Dixon, SRNF, radixon{at}noble.org; Co-PI: Lloyd W. Sumner, SRNF, lwsumner{at}noble.org; Co-PI: Gregory D. May, SRNF, gdmay{at}noble.org; and Co-PI: J. Tim Smith, Southeastern Oklahoma State University, tsmith{at}sosu.edu

This work was supported by the NSF (Plant Genome Project No. 0109732; http://medicago.vbi.vt.edu).

This project studies the responses of M. truncatula root cell cultures to three elicitations: exposure to high-UV light, methyl-jasmonic acid, and yeast elicitor. During the course of this project, methodologies are being developed to produce, store, and analyze integrated functional genomics data sets including the transcriptome, the proteome, and the metabolome. Analysis of these data sets is expected to facilitate the functional identification of many new genes and proteins in M. truncatula and other legumes, many of which will be associated with natural product biosynthesis. We aim to explain in a quantitative way, through computer models and simulations, the working of the cellular machinery as plant cells reprogram themselves in response to the three stresses. This information will facilitate future manipulations of legumes to improve beneficial traits such as disease resistance, radiation protection, and nutritional content by decreasing the levels of toxic natural products and increasing beneficial ones.

There is currently little knowledge available concerning the biosynthesis of the large majority of plant natural products, many of which are lead compounds for pharmacological drug development. This project is aimed at identification of new natural products and, importantly, also the genes involved in their biosynthesis. The data generated consist of hundreds of microarrays, two-dimensional gels, chromatography and electrophoresis runs, and various types of mass spectra. These data, together with processed data, consisting of ratios of protein, mRNA, and metabolite levels, will be available through a public database system that will provide and integrated view of the transcriptome, proteome, and metabolome.

MolMyk. Molecular Basics of Mycorrhizal Symbioses

PI: Alfred Pühler, Universität Bielefeld, Germany, puehler{at}genetik.uni-bielefeld.de; Co-PI: Helge Küster, Universität Bielefeld, helge.kuester{at}genetik.uni-bielefeld.de; Co-PI: Karsten Niehaus, Universität Bielefeld, Karsten.Niehaus{at}Genetik.Uni-Bielefeld.DE; Co-PI: Andreas Perlick, Universität Bielefeld, Andreas.Perlick{at}Genetik.Uni-Bielefeld.DE; Co-PI: Philipp Franken, Institute for Vegetable and Ornamental Plants, Grossbeeren, franken{at}igzev.de; Co-PI: Thomas Fester, Leibniz-Institut für Pflanzenbiochemie, Halle, tfester{at}ipb.uni-halle.de; Co-PI: Willibald Schliemann, Leibniz-Institut für Pflanzenbiochemie, Halle, wschliemann{at}ipb-halle.de; Co-PI: Dieter Strack and Bettina Hause, Leibniz-Institut für Pflanzenbiochemie, Halle, dstrack{at}Ipb.uni-Halle.DE, bhause{at}ipb.uni-halle.de; Co-PI: Michael H. Walter, Leibniz-Institut für Pflanzenbiochemie, Halle, mhwalter{at}ipb-halle.de; Co-PI: Franziska Krajinski, Universität Hannover, krajinski{at}lgm.uni-hannover.de; Co-PI: Hermann Bothe, Universität zu Köln, hbothe{at}novell.biolan.uni-koeln.de; Co-PI: Natalia Requena-Sanchez, Universität Tübingen, Natalia.Requena{at}Uni-Tuebingen.de; Co-PI: Rainer Hedrich and Peter Ache, Universität Würzburg, hedrich{at}botanik.uni-wuerzburg.de, ache{at}botanik.uni-wuerzburg.de; and Co-PI: Rainer Kaldenhoff, Universität Würzburg, kaldenhoff{at}botanik.uni-wuerzburg.de

This is Deutsche Forschungsgemeinschaft Focus Program no. SPP1084 (MolMyk; http://MolMyk.Genetik.Uni-Bielefeld.DE).

More than 80% of flowering plants are able to form either endomycorrhiza or ectomycorrhiza symbioses with certain groups of fungi. In return for carbon compounds, the symbiotic fungi supply phosphate, N, potassium, and trace elements to the plant, resulting in an increased plant growth and an improved resistance against biotic and abiotic stress.

The targeted molecular analysis of plant endo- and ectomycorrhiza is difficult due to the complex development of the interaction. The Deutsche Forschungsgemeinschaft Focus Program MolMyk was established in the year 2000 to apply untargeted techniques of plant genomics to identify plant genes essential for the establishment and functioning of mycorrhizal symbioses. We have chosen the interaction of M. truncatula with Glomus spp. as a model for endomycorrhiza and the interaction of poplar (Populus tremula) with Amanita muscaria as a model for ectomycorrhiza.

Within MolMyk, the collaboration of groups working on M. truncatula endomycorrhiza is supported primarily by establishing and applying modern methods and technologies of plant genomics and bioinformatics. As a basis for this collaboration, we have carried out and continue to perform an EST sequencing project using the interaction between M. truncatula and G. intraradices as a model. During this project, two cDNA libraries of M. truncatula endomycorrhiza were constructed and sequenced: a general cDNA library of mycorrhizal roots harvested at different stages of development and a suppression-subtractive hybridization cDNA library that was enriched for mycorrhiza-specific or mycorrhiza-amplified transcript sequences. To date, about 6,500 ESTs from these two libraries were generated and deposited in the EMBL and GenBank databases. Using the software tool BioMake that was developed within the MolMyk project, all EST sequences obtained were clustered and automatically annotated. Using electronic northern approaches, in silico patterns for the genes represented by the MolMyk EST collection were obtained based on comparisons with the TIGR M. truncatula Gene Index (http://www.tigr.org/tdb/mtgi/).

In collaboration with the French EST sequencing project "Functional Genomics in M. truncatula EST Analysis as a Tool to Explore M. truncatula Root Symbiotic Program" (http://medicago.toulouse.inra.fr/Mt/EST/DOC/MtB.html), we constructed macro- and microarrays covering the root interaction transcriptome of M. truncatula. These arrays, designated 6k-RIT, comprise about 6,000 EST clusters that are derived from cDNA libraries from root nodules, root endomycorrhiza, and uninfected M. truncatula roots. To widen the scope of our DNA arrays, we are switching to the construction and hybridization of 70-mer oligonucleotide microarrays covering about 16,000 M. truncatula EST clusters that are based on the TIGR M. truncatula Gene Index (see May et al., above). Both the 6k-RIT cDNA and the 16k oligo arrays are and will be used within MolMyk to obtain expression profiles of genes relevant for root endomycorrhiza under different symbiotic or physiological conditions. To evaluate expression profiles generated from DNA arrays, the EMMA ("ESTs meet microarrays") software was developed as an integral part of the bioinformatics section of MolMyk. The EMMA software is part of a relational database that allows the linkage of EST annotations with expression profiles obtained on the DNA arrays. The information gained by this integrated genomics approach will be complemented by functional analyses using gene silencing techniques in transgenic plants and transgenic roots of M. truncatula.

EST sequence and expression data obtained within the MolMyk genome project are mined by the project participants to address biological questions related to different aspects of legume endomycorrhiza. These questions can be grouped into three main sections. The section "development of mycorrhizae" covers the identification of signals involved in the recognition of symbiotic partners, the analysis of signal transduction pathways playing a role in the establishment of the symbiosis, and the identification of legume symbiosis-specific genes (symbiosins) that are common to the root nodule and the root endomycorrhiza symbiosis. The second section, "fluxes," addresses the transport of monosaccharides, amino acids, N and sulfur compounds, and the adaptation of the plant's and the fungal metabolism to the symbiotic status. Finally, applied aspects, such as mycorrhiza-induced plant resistance, are considered.

We envisage that the MolMyk project will deliver an integrated view of sequence properties, expression patterns, and functions for M. truncatula genes that are relevant for different symbiotic and physiological conditions of an arbuscular mycorrhiza.

Progress Report on TILLING. High-Throughput Genotyping in M. truncatula

PI: Douglas Cook, University of California, Davis, drcook{at}ucdavis.edu; and Co-PI: R. Varma Penmetsa, University of California, Davis, rvpenmetsa{at}ucdavis.edu

This work was supported by a U.S. DA-ARS cooperative agreement with the University of California (Davis).

For decades, biologists have used genetic screens to link the function of individual genes to specific biological processes. Although powerful, such "phenotype-first" screens are limited in the types of individual genes and range of alleles that can be identified. In contrast, "genotype-first" screens identify allelic variation in genes independent of phenotype. Genotype-first screens allow a systematic search for variation in any gene whose sequence is known and identify a greater diversity of alleles than do phenotype-first screens. Moreover, genotype-first screens are a logical follow-on from the massive EST sequencing projects that have been conducted for numerous plant species.

For reverse genetic analysis in M. truncatula, we have adapted the TILLING technology originally developed in Arabidopsis (McCallum et al., 2000). We have established an efficient pipeline for germplasm curation and DNA extraction from EMS mutagenized populations. To date, approximately 2,000 individual germlines have been prepared for genotyping, and an additional set of several thousand mutagenized, single-seed decent germlines are being prepared for incorporation into the screening pipeline. We have developed a high-throughput genotyping procedure based on the Cel I mismatch enzyme detection system to enable screening for single-nucleotide polymorphisms (SNPs) in genes of interest. DNA mismatches induced by EMS mutagenesis are cleaved by the Cel I endonuclease to identify plant lines containing SNPs (Oleykowski et al., 1998; Colbert et al., 2001). We have instituted a DNA multiplex strategy that involves pooling DNA from four unique germlines. The pooling increases throughput of screening and reduces cost, without compromising sensitivity of SNP detection.

To date, we have interrogated approximately 2,000 germlines for SNPs in 1-kb regions in four genes of interest. This screening has identified an average of approximately one SNP lesion per 1 kb per 500 germlines. After the incorporation of the remaining germlines to the existing group, it will be possible to screen an anticipated 4,000 unique germlines. We estimate that we can identify approximately eight SNPs in genes of interest, and this number would be augmented by screening of additional germplasm as this material is developed.

Proteome Analysis of Meristematic Development of M. truncatula and the Investigation of the Molecular Interactions between Rhizobium, Phytohormones, and Plant Nodulation Mutants

PI: Barry Rolfe, The Australian National University (ANU), Canberra, Australia, rolfe{at}rsbs.anu.edu.au; Collaborators and Co-Pis: Ulrike Mathesius, ANU, ulrike.mathesius{at}anu.edu.au; Michael Djordjevic, ANU, michael{at}rsbs.anu.edu.au; Jeremy Weinman, ANU, weinman{at}rsbs.anu.edu.au; Charles Hocart, ANU, hocart{at}rsbs.anu.edu.au; Georg Weiller, ANU, weiller{at}rsbs.anu.edu.au; Peter Gresshoff, University of Queensland, Australia; Ray Rose, University of Newcastle, Australia, birjr{at}alinga.newcastle.edu.au; Mohan Singh, University of Melbourne, Australia, mohan{at}unimelb.edu.au; and W. Dietz Bauer, Ohio State University, bauer.7{at}osu.edu

This work was supported by the ANU (block grant), by the U.S. Department of Agriculture (project no. 741734), by the University of Newcastle (project grant), and by the ARC Centre of Excellence (5-year support for CILR; http://biology.anu.edu.au/research-groups/gig/gig.html; http://semele.anu.edu.au/2d/2d.html).

Meristems Are the Key to Plant Growth and Adaptation

Integrated Legume Research Program

Research Tools

Meristem Development of M. truncatula Cells Grown in Culture

Legume Roots, Nodules, and Induction of Meristems

Communication between Meristems

Resistance of M. truncatula to Pathogens and Insect Pests. R Gene Isolation, Defense Gene Expression Profiling, Proteomics, and Comparative Mapping within the Papillonidae

PI: Karam B. Singh, Commonwealth Scientific and Industrial Research Organization (CSIRO) Centre for Environment and Life Sciences, Australia, karam.singh{at}csiro.au; Co-PI: Owain Edwards, CSIRO Centre for Environment and Life Sciences, Australia, owain.edwards{at}csiro.au; Co-PI: Richard P. Oliver, Murdoch University, Australia, roliver{at}central.murdoch.edu.au; Co-PI: John Klingler, Centre for Environment and Life Sciences, Australia, john.klingler{at}csiro.au; and Co-PI: Simon Ellwood, Murdoch University, Australia, sellwood{at}central.murdoch.edu.au

Disease and pests are major problems for legume crops worldwide. The development of M. truncatula as a model system for plant research opens up new opportunities to study biotic stress responses in these important crops. Our research program is using genomic approaches in M. truncatula to study some of the key biotic stresses impacting on legume crops in Australia. Our program is linked with legume breeders through the Centre for Legumes In Mediterranean Agriculture and the South Australian Research and Development Institute to facilitate the transfer of results obtained with M. truncatula into other agronomically important species. The accumulating evidence of extensive synteny among legumes will aid rapid transfer between species. Of particular relevance to our research program is the fact that M. truncatula is grown in Australia as an important pasture crop and the Australian Medicago Genetic Resource Centre in Adelaide, South Australia has one of the world's largest collections of annual and perennial Medicago germplasm (http://www.sardi.sa.gov.au/pages/pastures/pas_genetres.htm).

Molecular Genetic and Genomic Approaches to Study
Aphid-Plant Interactions

Genetic Dissection of Necrotrophic Fungal Disease Resistance in Legumes Using M. truncatula

	G. MAX

TOP INTRODUCTION L. JAPONICUS M. TRUNCATULA G. MAX PHASEOLUS VULGARIS LITERATURE CITED

The Public Soybean EST Project

PI: Randy Shoemaker, USDA-ARS, Iowa State University, rcsshoe{at}iastate.edu; Co-PI: Ernest Retzel, University of Minnesota, ernest{at}ahc.umn.edu; Co-PI: Lila Vodkin, University of Illinois, l-vodkin{at}uiuc.edu; Co-PI: Paul Keim, northern Arizona University, paul.keim{at}nau.edu; and Co-PI: Sandra Clifton, Washington University School of Medicine, sclifton{at}watson.wustl.edu

This work was jointly supported by the United Soybean Board and the North Central Soybean Research Program (http://soybean.ccgb.umn.edu/).

The size and complexity of the soybean genome currently make whole-genome sequencing a nontrivial problem to be solved. Because it is primarily the coding regions of genes that provide the information many geneticists desire, transcript sequencing provides an efficient and cost-effective method of sampling the coding portion of the genome.

The Public Soybean EST Project began in 1998 and was funded by check-off dollars through the North Central Soybean Research Program and the United Soybean Board, and by the USDA-ARS. The project was developed as a truly "public" project with input from many members of the soybean community, and data placed immediately into the public sector.

More than 80 cDNA libraries were generated through the project. These libraries represent transcripts from a very broad range of genotypes, developmental and reproductive stages, organs, tissues, and abiotic and biotic stresses. Libraries are sampled deeply when the frequency of discovery of "new" ESTs remains high, and less deeply when redundant sampling significantly decreases the efficiency of discovery. Most cDNA libraries were generated by members of the PI's or CoPIs' laboratories. Other libraries were created and donated by scientists with unique specialization with particular soybean diseases or stresses. Still other scientists contributed tissues representing unique mutants or unique soybean genotypes.

The success of the project has been gratifying. More than 285,000 EST sequences have been deposited in dbEST, and more are being processed. The soybean ESTs represent the largest collection of ESTs for any plant. The collection currently coalesces into approximately 50,000 unigenes and singletons. The average length of each sequence is greater than 400 bases. Because of the size of the EST collection, contigs are often comprised of more than six ESTs. On average, each contig spans more than 750 bases.

The project is providing more to the legume community than simply "gene discovery." When the EST sample is large and random, the frequency of recurrence of any EST provides information on the level of expression of the corresponding gene in the tissue or organ sampled. As a consequence, we are able to draw inferences about gene relationships and gene ontology based upon similarities in expression patterns. In addition, we are learning much about the evolution of duplicated genes in this ancient polyploid and can estimate a coalescence time for the duplicates.

This project has already resulted in publications based directly upon analyses of the EST collections (Granger et al., 2002; Shoemaker et al., 2002), and led to a project funded by the NSF Plant Genome Program (A Functional Genomics Program for Soybean; L. Vodkin, PI; see below).

A Functional Genomics Program for Soybean

PI: Lila Vodkin, University of Illinois, l-vodkin{at}uiuc.edu; Co-PI: Paul Keim, Northern Arizona University, paul.keim{at}nau.edu; Co-PI: Joseph Polacco, University of Missouri, bcjoecpo{at}muccmail.missouri.edu; Co-PI: Ernest Retzel, University of Minnesota, ernest{at}mail.ahc.umn.edu; Co-PI: Randy Shoemaker, USDA/ARS and Iowa State University, rcsshoe{at}iastate.edu; and Co-PI: Nevin Young, University of Minnesota, neviny{at}umn.edu

This work was supported by the NSF (Plant Genome Project No. 9872565; http://soybeangenomics.cropsci.uiuc.edu/ and http://soybean.ccgb.umn.edu/).

This project is a collaborative effort to stimulate basic and applied research in functional genomics of soybean through development of tools for global gene expression analysis and physical mapping. The main objectives are to build a soybean "unigene" set defined by 5' and 3' sequence data, to construct and use microarrays for global expression, to generate and sequence SAGE-tagged libraries, and to produce BAC end sequences that are anchored on the soybean genetic map. The objectives of this NSF-sponsored project and progress toward those goals are described briefly below:

Development of a Soybean "Unigene Set" and Validation by 3' Sequencing

Development and Use of Global Expression Methods for Soybean

Building an Infrastructure of BAC Contigs Anchored to the Genetic Map

Integrative Physical Mapping of the Soybean Genome

PI: David A. Lightfoot, Southern Illinois University, ga4082{at}siu.edu; Co-PI: Khalid Melssem, Southern Illinois University, kmeksem{at}siu.edu; and Co-PI: Hongbin Zhang, Texas A&M University, hbz7049{at}tamu.edu

This work was supported by the NSF (Plant Genome Award no. 9872635; http://www.siu.edu, http://hbz.tamu.edu/info).

The development of a physical map of the soybean genome has the potential to accelerate the rate of discovery and cloning of economically important genes by severalfold. A detailed physical map would also provide for the future sequencing of the whole soybean genome. The objectives of this research are to develop a physical map of the soybean genome and to integrate it with the genetic map. The ultimate goal is to facilitate discovery, cloning, manipulation, and utilization of soybean genes for genetic improvement and agricultural production. The research also provides soybean researchers with electronic access to BAC clones encompassing regions likely to contain genes of economic importance. This report overviews the public resources created by the project and gives a progress report on results to date.

Community Access to Project Resources

Summary of the Results of the Completed Work

Conclusions

Dissecting Phytophthora Resistance in Soybean Using Expression Profiling and Analysis of QTLs

PI: Brett M. Tyler, Virginia Polytechnic and State University, bmtyler{at}vt.edu; Co-PI: Glenn Buss, Virginia Polytechnic and State University, gbuss{at}vt.edu; Co-PI: Anne E. Dorrance, Ohio State University, dorrance.1{at}osu.edu; Co-PI: Ina Hoeschele, Virginia Polytechnic and State University, inah{at}vt.edu; Co-PI: M. A. Saghai Maroof, Virginia Polytechnic and State University, smaroof{at}vt.edu; Co-PI: Steven St. Martin, Ohio State University, stmartin+{at}osu.edu; and Co-PI: Keying Ye, Virginia Polytechnic and State University, keying{at}vt.edu

This work was supported by the NSF (Plant Genome Project No. 0211863; initiated October 1, 2002; http://www.vbi.vt.edu/staff/perspage_fcty/fcty_tyler_bret.htm).

Much has been learned over the last decade regarding the mechanism by which plants are protected from disease by major resistance genes. Much less is known about mechanisms of quantitative resistance, which protect a plant species against all genetic forms of a pathogen, albeit not completely. However, this form of durable resistance often retains its effectiveness over time against pathogen populations that can change rapidly to overcome major resistance genes in the host.

This project will use functional genomics to dissect mechanisms of quantitative resistance in soybean against one of its most destructive pathogens, the oomycete Phytophthora sojae. cDNA and oligonucleotide microarrays will be used to examine gene expression profiles of both the host and the pathogen during P. sojae infection of soybean lines containing varied levels of quantitative resistance and in progeny segregating for quantitative resistance. QTL analysis of the expression profiles in a segregating population will enable us to identify soybean loci that contribute to quantitative resistance.

Thirteen major resistance alleles (Rps genes) have been identified in soybeans at seven loci that provide protection against P. sojae (Diers et al., 1992; Lohnes and Schmitthenner, 1997), and matching avirulence genes have been identified in P. sojae for 11 of these (May et al., 2002). Many populations of P. sojae have adapted to overcome the Rps genes that have been deployed in commercially grown soybean cultivars. Quantitative resistance to P. sojae is also present in soybeans and appears to be controlled by multiple genes (Walker and Schmitthenner, 1984; Glover and Scott, 1998). It is an important means for controlling the disease, especially in locations where Rps genes are no longer effective. Preliminary data from an analysis of three soybean populations indicate that three or more loci may be involved with this trait (A.E. Dorrance and S. St. Martin, unpublished data). Heritability estimates range from 0.57 to 0.87 for root lesion length in these F₄:F₆ populations.

The soybean-P. sojae system is one of very few pathosystems in which extensive genomic resources are available to examine both plant and pathogen simultaneously (Tyler, 2001; see also summaries by Lightfoot et al., Shoemaker et al., and Vodkin et al.). The overall design of this project is to screen a large population of soybean genetic progeny segregating for quantitative resistance against P. sojae, using soybean and P. sojae microarrays. QTL analysis of the plant and pathogen profiles will identify soybean loci that contribute to quantitative resistance, and the mechanisms of resistance conferred by genes at those loci. By assaying pathogen expression profiles in addition to plant expression profiles, we should more readily identify minor resistance genes, especially if there are several such genes segregating. For example, a small decrease in lesion size caused by one minor gene may be very difficult to assay and map using conventional QTL analysis, but if that gene triggers large-scale changes in the levels of certain pathogen mRNAs, then the resistance gene would be easily detected. Further, just as expression profiles can reveal how antimicrobial drugs debilitate human pathogens, they should reveal mechanisms by which plant defenses debilitate plant pathogens. Therefore, analyzing plant and pathogen expression profiles simultaneously will provide unique insights into the interplay between plant and pathogen genes that regulate the outcome of an infection.

	PHASEOLUS VULGARIS

TOP INTRODUCTION L. JAPONICUS M. TRUNCATULA G. MAX PHASEOLUS VULGARIS LITERATURE CITED

The Phaseomics International Consortium

Coordinators: William J. Broughton, University of Geneva, Switzerland, william.broughton{at}bioveg.unige.ch; Gina Hernández, C. de Investigación sobre Fijación de Nitrógeno-Universidad Nacional Autónoma de México (UNAM), Mexico, gina{at}cifn.unam.mx; Matthew Blair, International Center for Tropical Agriculture (CIAT), Colombia, m.blair{at}cgiar.org; Paul Gepts, University of California, Davis, plgepts{at}ucdavis.edu; Jos Vanderleyden, Katholieke University Leuven, Belgium, jozef.vanderleyden{at}agr.kuleuven.ac.be; and Nancy Terryn, Ghent University, Belgium, nater{at}gengenp.rug.ac.be

One-half the dry edible legumes consumed worldwide as human food are common beans. Beans are one of the world's most ancient crops, originating in Central America and the Andean region of South America. Beans and their relatives (Phaseolus spp.) are extremely diverse crops in terms of cultivation methods, uses, the range of environments to which they have been adapted, and morphological variability. Their genetic resources exist as a complex array of major and minor gene pools, races, and intermediate types, with occasional introgression between wild ancestors and domesticated types. Thus, beans are a crop that is adapted to many niches, both in agronomic and consumer preference terms. Although many diverse associations contribute to symbiotic N fixation, in most agricultural settings the primary source (80%) of biological fixed N is through the Rhizobium-legume symbiosis. Important agricultural goals include enhancing the use of and improving the management of biologically fixed N, as well as increasing the adaptation to abiotic stresses. The original microsymbiont of common bean is the gram-negative bacterium Rhizobium etli that also originated in the Americas and has co-evolved with beans for millennia. However, other rhizobia, such as Rhizobium tropici, which is resistant to acid soils, M. loti, or Rhizobium leguminosarum bv phaseoli, can also nodulate the species.

Genomics, transcriptomics, and proteomics permit the study of many (and sometimes all) genes of a particular organism. Significant discoveries concerning interrelationships between some of the basic metabolic functions of the organisms have been made this way for model plant species. As a consequence, an integrated, almost holistic view of the organism is evolving. An international consortium called Phaseomics (Phaseolus genomics) has been formed to establish the necessary framework of knowledge and materials for the advancement of genomic studies of bean. Phaseomics includes around 80 scientists from 20 different countries. The principal goal of the consortium is to increase the genetic resources and tools available for the crop, especially large insert and cDNA libraries, genomic sequences, ESTs, and genetic markers. The ultimate objective of the research carried out within Phaseomics is to assist in the generation of new common bean varieties that are not only suitable for but also desired by local farmer and consumer communities. An additional long-term goal is to more rationally use the large germplasm collections held for the crop at the CIAT (36,000 accessions), and at national germplasm repositories in the U.S. (USDA), Brazil (Brazilian Agricultural Research Corporation), and Mexico (National Institute for Forest and Agricultural Research). A detailed description of the global project proposed by the Phaseomics partners will be published this year (Broughton et al., 2003). Phaseomics partners have met in two occasions: in 2001 in Cuernavaca, Mexico, and in 2002 in Geneva. During the second Phaseomics meeting, different working groups that concentrate on certain of the Phaseomics goals were formed, with the aim of providing efficient exchange of information among the scientists performing related individual projects and to avoid duplication.

One of the working groups is the EST sequencing group that is coordinated by G. Hernández. EST projects provide an inexpensive and efficient method to get information on the genes that are expressed in a certain tissue or organ of the plant and when converted to genetic markers, breeders can use them to position genes on the genetic map of the species. Judicious selection of the type of tissue from which to isolate the mRNA (and hence prepare a cDNA library) provides valuable information not only on the type of genes found in a particular plant, but also on the conditions in which they are expressed. Thus, EST projects permit "skimming" gene expression information and, by extension, the genome itself. The following Phaseomic partners are performing projects on common bean ESTs that are funded by different sources depending on the groups and/or country. M. Blair, S. Beebe, and J. Tohme from (CIAT, Colombia) have sequenced approximately 4,000 ESTs from leaves of an Andean genotype G19833 and from young lateral and basal roots treated with limited or unlimited phosphorus of the Mesoamerican genotype DOR364. M. Melotto et al. (University of São Paulo, Brazil) are preparing two cDNA libraries from seedling shoots of the anthracnose-resistant genotype SEL1308 with and without inoculation with Colletotrichum lindemuthianum, and they will sequence a total of 5,000 ESTs from each library. J. Vanderleyden et al. are sequencing ESTs from a cDNA library from nodules (10 d postinoculation) of the BAT477 genotype inoculated with R. etli CNAPF512. G. Hernández, M. Lara, and M. Ramírez (Nitrogen Fixation Research Center-UNAM, Mexico), in collaboration with C.P. Vance (University of Minnesota, USDA) have sequenced 3,000 ESTs from a cDNA library of nodules (18 d postinoculation) from the variety Negro Jamapa 81 inoculated with R. tropici 899. Libraries were also prepared from mRNA isolated from developing pods and phosphorus-stressed roots for which 3,000 cDNA clones will be sequenced each. W.J. Broughton et al. (University of Geneva) are currently isolating mRNA from root hairs (BAT93) treated with Rhizobium sp. NGR234 and its Nod factors. Each group has or will have their internal database and will share the EST data with the public through GenBank. The research from the several groups is moving toward functional genomics analyses, through macro- and microarray technologies.

An efficient and reliable genetic transformation system is crucial to any genomic project. The working group coordinated by N. Terryn focuses on Phaseolus transformation mediated by Agrobacterium. It is now clear that two Phaseomics groups can routinely transform Phaseolus acutifolius: the group in Ghent (N. Terryn) and the group at CIAT (A. Mejía-Jiménez and J. Tohme). Most work is done with NI576, a wild P. acutifolius line, whereas the Ghent group is now also exploring the use of a cultivated variety. The problem is that the protocol is rather laborious and time consuming. As for P. acutifolius crosses to common bean, embryo rescue is needed, and crosses are not always successful. Therefore, CIAT is working on hybrid lines between P. acutifolius and common bean that will have greater transformability. This could make it easier to cross transgenics obtained in these lines with common bean varieties of interest. Although it has been shown that A. tumefaciens can transfer DNA to common bean, an efficient transformation system for common bean has not yet been established. The groups of M. Lara and G. Hernandez (Mexico) are working in trying to develop a transformation system for common bean based in A. tumefaciens and in vitro regeneration protocols that they have developed for several cultivars.

Other Phaseomics working groups and their coordinators are: resources and libraries, P. Gepts (U.S.); biological N fixation, G. Hardarson (Austria); germplasm, M. Blair (Colombia); genome sequencing, E. Triplett (U.S.); bioinformatics, J.P. Nap (The Netherlands); and coordination, W.J. Broughton (Switzerland).

Development of Genetic and Genomic Tools to Study Nutritional Quality and Aluminum (Al) Tolerance in Common Bean

PI: M.W. Blair, CIAT, Colombia, m.blair{at}cgiar.org; Co-PI: S. Beebe, CIAT, Colombia, s.beebe{at}cgiar.org; Co-PI: P. Beyer, University of Freiburg, Germany, beyer{at}uni-freiburg.de; Co-PI: D. Dellapena, Michigan State University, dellapen{at}msu.edu; Co-PI: M. Grusak, USDA-ARS Children's Nutrition Research Center, Baylor College of Medicine, mgrusak{at}bcm.tmc.edu; Co-PI: W. Horst University of Hannover, Germany, horst{at}mbox.pflern.uni-hannover.de; Co-PI: J. Tohme, CIAT, Colombia, j.tohme{at}cgiar.org; and Co-PI: I. Rao, CIAT, Colombia, i.rao{at}cgiar.org

This work was supported by the U.S. Agency for International Development (grant no. DAN-G-IN-89-00048-00) and by the German Federal Ministry for Economic Cooperation and Development (Project no. 2000.7860.0-001.11, contract no. 81043147; http://www.ciat.cgiar.org/biotechnology/index.htm).

Micronutrient deficiencies in human populations, especially for iron and zinc, are a serious health concern in the developing world, where more than 1 billion people live in a state of abject poverty and have limited access to sufficient quantities of these minerals in their daily diet. Common beans are by far the world's most important food grain legume and, therefore, have a role in addressing this problem (Graham et al., 1999). Genetic variability exists for mineral content in common bean and, surprisingly, seed iron and zinc concentrations tend to be correlated, leading to the promise of incorporating these traits into a bean improvement program (Beebe et al., 2000). However, the physiology and genetic mechanisms controlling mineral transport pathways are still being worked out and remain a priority if nutritional breeding is to be effective.

Breeding for tolerance to Al phytotoxicity is another important objective of bean breeding programs in the tropics, because this is a serious edaphic constraint affecting over 40% of the world's arable land, especially acid soils of tropical and subtropical regions (Von Uexkull and Mutert, 1995). New bean varieties with Al tolerance would reduce the dependence of farmers on lime and fertilizer (inorganic/organic) inputs that are used to solve these soil problems. Genetic variation exists for Al resistance among common bean genotypes (Thung and Rao, 1999), and screening procedures have been or are being established. However, plant breeders are only in the early stages of developing Al-resistant cultivars for farmers in areas that need them, and much work is needed to understand the genetic control of Al toxicity tolerance and how best to select for this trait.

At the CIAT, two projects are investigating the genes that control micronutrient content and Al stress tolerance in common bean. The issue of seed micronutrient accumulation and abiotic stress tolerance are related given the link between soil and plant nutrition and their affect on grain quality and human nutrition. Therefore, in the research for these two projects, we are asking similar questions, such as how plant roots adapt to acid soils or how root uptake of minerals translates into increased accumulation in the grain, and for both projects, we are developing similar genetic tools that are described below.

Marker Development

Genetic Mapping

Future Plans