Construction of tomato genomic DNA libraries in a binary-BAC (BIBAC) vector

This is the first report of large insert genomic DNA libraries constructed in a binary-BAC (BIBAC) vector. Genomic DNA libraries containing approximately 4.6 haploid nuclear genomic equivalents were constructed for Lycopersicon esculentum (cv. Mogeor) and Lycopersicon pennellii (LA716) in the BIBAC2 vector. The L. esculentum library has an average insert size of 125 kb and is comprised of 42 272 individual colonies stored as frozen cultures in a 384-well format (108 plates). The L. pennellii library has an average insert size of 90 kb and is comprised of 53 760 individual clones (140 384-well plates). In each of the libraries, it is estimated that 90% of the colonies contain genomic DNA inserts. The composition of the L. esculentum and L. pennellii libraries was determined by analyzing a series of randomly selected clones. The L. esculentum library was surveyed for clones containing chloroplast DNA (1.4%), mitochondrial DNA (0.012%) and repetitive DNA motifs. BIBAC clones that may contain a gene of interest can be identified from these libraries by colony hybridization with homologous or heterologous probes or by PCR pooling techniques. Once identified, BIBAC genomic DNA library clones are immediately suitable for Agrobacterium tumefaciens-mediated plant transformation.     

Construction of BAC and BIBAC libraries from sunflower and identification of linkage group-specific clones by overgo hybridization

Complementary BAC and BIBAC libraries were constructed from nuclear DNA of sunflower cultivar HA 89. The BAC library, constructed with BamHI in the pECBAC1 vector, contains 107,136 clones and has an average insert size of 140 kb. The BIBAC library was constructed with HindIII in the plant-transformation-competent binary vector pCLD04541 and contains 84,864 clones, with an average insert size of 137 kb. The two libraries combined contain 192,000 clones and are equivalent to approximately 8.9 haploid genomes of sunflower (3,000 Mb/1C), and provide a greater than 99% probability of obtaining a clone of interest. The frequencies of BAC and BIBAC clones carrying chloroplast or mitochondrial DNA sequences were estimated to be 2.35 and 0.04%, respectively, and insert-empty clones were less than 0.5%. To facilitate chromosome engineering and anchor the sunflower genetic map to its chromosomes, one to three single- or low-copy RFLP markers from each linkage group of sunflower were used to design pairs of overlapping oligonucleotides (overgos). Thirty-six overgos were designed and pooled as probes to screen a subset (5.1×) of the BAC and BIBAC libraries. Of the 36 overgos, 33 (92%) gave at least one positive clone and 3 (8%) failed to hit any clone. As a result, 195 BAC and BIBAC clones representing 19 linkage groups were identified, including 76 BAC clones and 119 BIBAC clones, further verifying the genome coverage and utility of the libraries. These BAC and BIBAC libraries and linkage group-specific clones provide resources essential for comprehensive research of the sunflower genome.     

BIBAC and TAC clones containing potato genomic DNA fragments larger than 100 kb are not stable in Agrobacterium

Development of efficient methods to transfer large DNA fragments into plants will greatly facilitate the map-based cloning of genes. The recently developed BIBAC and TAC vectors have shown potential to deliver large DNA fragments into plants via Agrobacterium-mediated transformation. Here we report that BIBAC and TAC clones containing potato genomic DNA fragments larger than 100 kb are not stable in Agrobacterium. We tested the possible factors that may cause instability, including the insert sizes of the BIBAC and TAC constructs, potato DNA fragments consisting of highly repetitive or largely single-copy DNA sequences, different Agrobacterium transformation methods and different Agrobacterium strains. The insert sizes of the potato BIBAC and TAC constructs were found to be critical to their stability in Agrobacterium. All constructs containing a potato DNA fragment larger than 100 kb were not stable in any of the four tested Agrobacterium strains, including two recA deficient strains. We developed a transposon-based technique that can be used to efficiently subclone a BAC insert into two to three BIBAC/TAC constructs to circumvent the instability problem.Communicated by J. Dvorak     

A large insert Thellungiella halophila BIBAC library for genomics and identification of stress tolerance genes

Salt cress (Thellungiella halophila), a salt-tolerant relative of Arabidopsis, has turned to be an important model plant for studying abiotic stress tolerance. One binary bacterial artificial chromosome (BIBAC) library was constructed which represents the first plant-transformation-competent large-insert DNA library generated for Thellungiella halophila. The BIBAC library was constructed in BamHI site of binary vector pBIBAC2 by ligation of partial digested nuclear DNA of Thellungiella halophila. This library consists of 23,040 clones with an average insert size of 75 kb, and covers 4× Thellungiella halophila haploid genomes. BIBAC clones which contain inserts over 50 kb were selected and transformed into Arabidopsis for salt tolerant plant screening. One transgenic line was found to be more salt tolerant than wild type plants from the screen of 200 lines. It was demonstrated that the library contains candidates of stress tolerance genes and the approach is suitable for the transformation of stress susceptible plants for genetic improvement.     

Construction and characterization of a bacterial artificial chromosome library of the causal agent of Black Sigatoka fungal leaf spot disease of banana and plantain, <Emphasis Type="Italic">Mycosphaerella fijiensis</Emphasis>

A bacterial artificial chromosome library of the causal agent of the Black Sigatoka leaf spot disease of banana and plantain, Mycosphaerella fijiensis, has been constructed using a non-sphaeroplasting technique and characterized using both homologous and heterologous probes. After first and a second size selection of PFGE-fractionated DNA, a ligation was obtained using a 1:4 molar ratio (insert:vector). One hundred random clones were analyzed, and the mean insert size was estimated to be 90 kb. The range of the insert sizes was between 40 and 160 kb. The highest percentage of inserts belonged to the range between 80 and 100 kb; 32% of the inserts had 2 or 3 internal NotI sites. This library consists of 1920 clones, if the genomic size is at least 35 Mb, then this represents 4.9× genome equivalents, which was supported by hybridization results with homologous and heterologous probes. Blondy Canto-Canché and Diana Karina Guillén-Maldonado contributed equally to this work and should be regarded as co-first authors.     

Development of a new transformation-competent artificial chromosome (TAC) vector and construction of tomato and rice TAC libraries

Recent research has shown that BIBAC (binary bacterial artificial chromosome) and TAC (transformation-competent artificial chromosome) vector systems are very useful tools for map-based cloning of agronomically important genes in plant species. We have developed a new TAC vector that is suitable for both dicot and monocot transformation. Using this new TAC vector, we constructed large-insert genomic libraries of tomato and rice. The tomato library contains 96,996 clones (28.3-38.5 kb insert size) and has 3.18 haploid genome equivalents. The rice TAC library has 32.7 kb average insert size and has 9.24 haploid genome equivalents. The quality of these two libraries was tested using PCR to verify genome coverage. Individual clones were characterized to confirm insert integrity by Southern analysis, end sequencing and genetic mapping. To investigate the potential application of these TAC libraries in map-based cloning, TAC constructs containing a 45 kb fragment were introduced into the rice genome via Agrobacterium-mediated transformation. Molecular analysis indicates that the 45 kb fragment was successfully transferred into the rice genome. Although rearrangements of the introduced DNA were detected, 50% of regenerated plants contained at least one intact copy of the 45 kb clone and associated vector sequences. These libraries provide us with a valuable resource to rapidly isolate important genes in tomato and rice.     

Characterization of a plant-transformation-ready large-insert BIBAC library of Arabidopsis and bombardment transformation of a large-insert BIBAC of the library into tobacco

Plant-transformation-ready, large-insert binary bacterial artificial chromosome (BIBAC) libraries are of significance for functional and network analysis of large genomic regions, gene clusters, large-spanning genes, and complex loci in the post-genome era. Here, we report the characterization of a plant-transformation-ready BIBAC library of the sequenced Arabidopsis genome for which such a library is not available to the public, the transformation of a large-insert BIBAC of the library into tobacco by biolistic bombardment, and the expression analysis of its containing genes in transgenic plants. The BIBAC library was constructed from nuclear DNA partially digested with BamHI in the BIBAC vector pCLD04541. It contains 6144 clones and has a mean insert size of 108?kb, representing 5.2× equivalents of the Arabidopsis genome or a probability of greater than 99% of obtaining at least one positive clone from the library using a single-copy sequence as a probe. The transformation of the large-insert BIBAC and analyses of the transgenic plants showed that not only did transgenic plants have intact BIBAC DNA, but also could the BIBAC be transmitted stably into progenies and its containing genes be expressed actively. These results suggest that the large-insert BIBAC library, combined with the biolistic bombardment transformation method, could provide a useful tool for large-scale functional analysis of the Arabidopsis genome sequence and applications in plant-molecular breeding.     

Construction of BAC and BIBAC libraries and their applications for generation of SSR markers for genome analysis of chickpea, Cicer arietinum L.

Large-insert bacterial artificial chromosome (BAC) libraries, plant-transformation-competent binary BAC (BIBAC) libraries, and simple sequence repeat (SSR) markers are essential for many aspects of genomics research. We constructed a BAC library and a BIBAC library from the nuclear DNA of chickpea, Cicer arietinum L., cv. Hadas, partially digested with HindIII and BamHI, respectively. The BAC library has 14,976 clones, with an average insert size of 121 kb, and the BIBAC library consists of 23,040 clones, with an average insert size of 145 kb. The combined libraries collectively cover ca. 7.0× genomes of chickpea. We screened the BAC library with eight synthetic SSR oligos, (GA)₁₀, (GAA)₇, (AT)₁₀, (TAA)₇, (TGA)₇, (CA)₁₀, (CAA)₇, and (CCA)₇. Positive BACs were selected, subcloned, and sequenced for SSR marker development. Two hundred and thirty-three new chickpea SSR markers were developed and characterized by PCR, using chickpea DNA as template. These results have demonstrated that BACs are an excellent source for SSR marker development in chickpea. We also estimated the distribution of the SSR loci in the chickpea genome. The SSR motifs (TAA)_n and (GA)_n were much more abundant than the others, and the distribution of the SSR loci appeared non-random. The BAC and BIBAC libraries and new SSR markers will provide valuable resources for chickpea genomics research and breeding (the libraries and their filters are available to the public at ).J. Lichtenzveig and C. Scheuring contributed equally to this study.     

A plant-transformation-competent BIBAC library of ginseng (Panax ginseng C. A. Meyer) for functional genomics research and characterization of genes involved in ginsenoside biosynthesis

Ginseng (Panax ginseng C. A. Mey.) is widely used as a major medicinal herb and as a feedstock for the medicine, beverage, food, cosmetic, etc. industries, in China and several other Asian countries. However, limited research has been accomplished into its genetics, genomics and breeding. To clone, characterize and utilize the genes of economic importance in the species, we have developed a large-insert plant-transformation-competent binary bacterial artificial chromosome (BIBAC) library for Jilin ginseng cv. Damaya. The library contains 141,312 clones, with an average insert size of 110 kb, each likely containing approximately 20–30 genes. The clones of the library have all been arrayed in 384-well microplates and permanently archived. We screened the library and identified BIBAC clones containing nine genes likely involved in the biosynthesis pathway of ginsenosides—the major medicinally effective compounds of ginseng—with approximately four BIBACs per gene. This result further verified the quality of the library and demonstrated its utility in cloning, characterization and utilization of economically important genes in ginseng. Furthermore, since the library is cloned in a plant-transformation-competent BIBAC vector (pCLD04541) that can be directly transformed in a variety of plants via both the Agrobacterium-mediated method and the particle bombardment method, we have also demonstrated the stability of large-insert ginseng DNA BIBACs in different Agrobacterium strains, which is crucial to large-insert BIBAC transformation in plants. Therefore, the Jilin ginseng BIBAC library provides resources and tools useful for functional genomics research, and cloning, characterization and utilization of economically important genes in the species.     

Construction of a yeast artificial chromosome library of pepper (Capsicum annuum L.) and identification of clones from the Bs2 resistance locus

A yeast artificial chromosome (YAC) library was constructed using high-molecular-weight DNA isolated from pepper (Capsicum annuum L.) leaf protoplasts. Insert DNA was prepared by partial digestion using EcoRI and subjected to electrophoretic fractionation before in-gel ligation to the pJS97/98 YAC vector. Prior to transformation of yeast spheroplasts, ligation products were subjected to a second electrophoretic size selection. The library consists of about 19 000 clones with an average insert size of 500 kb, thus representing approximately three haploid genome equivalents. Three PCR-based markers tightly linked to the pepper Bs2 resistance gene were used to assess the utility of this library for positional cloning. Three YAC clones containing pepper genomic DNA from the Bs2 resistance locus were isolated from the library. The clones ranged in size from 270 kb to 1.2 Mb and should prove useful for the cloning of the Bs2 gene. Received: 15 January 1999 / Accepted: 11 May 1999     

A plant-transformation-competent BIBAC library from the <Emphasis Type="Italic">Arabidopsis thaliana</Emphasis> Landsberg ecotype for functional and comparative genomics

The genome of the model plant Arabidopsis thaliana has been sequenced to near completion. To facilitate experimental determination of the function of every gene in the species, we constructed a large-insert library from the Landsberg ecotype using a plant-transformation-competent binary BAC vector, BIBAC2. The library contains 11,520 clones with an estimated average insert size of 162 kb. Of a sample of 102 clones, 17.6% had no inserts; further, in the library as a whole, 287 clones contained chloroplast DNA, and 25 contained mitochondrial DNA. Thus it is estimated that 9,295 clones originated from the nuclear genome, representing a 11.5 x coverage. The library was further characterized by screening with probes corresponding to 180-bp repeats, 5S rDNA, 18S-25S rDNA and 23 single-copy RFLP markers. The results showed that 92 clones contained 180-bp centromeric repeats, 78 contained 5S rDNA and 95 contained 18S-25S rDNA, approximately 1%, 0.8% and 1%, respectively, of the nuclear clones in the library. Screening the library with the 23 RFLP markers showed that each one hybridized to an average of seven clones. This library is the first large-insert DNA library for the widely studied Landsberg erecta strain. It will greatly facilitate gene identification by complementation screening, and will enhance analysis of the structure, organization and evolution of the A. thaliana genome.     

Construction and characterization of a BIBAC library of <Emphasis Type="Italic">Jatropha curcas</Emphasis> L. and identification of BIBAC clones containing genes associated with fatty acid metabolism

Jatropha curcas L. is a potentially significant bioenergy crop in the tropics and subtropics. Here we present a plant-transformation-competent binary bacterial artificial chromosome (BIBAC) library from Jatropha cultivar YN049-4. This library was constructed with BamH in the vector pCLD04541, consists of 30,720 clones and is arrayed in 80 384-well microtiter plates. Since 92.1% (28,293) of its clones were shown to contain Jatropha DNA inserts with an average size of 131.9 kb, the library is estimated to represent approximately 8.9 haploid genome equivalents of the species, thus providing a greater than 99% probability of discovering a particular single-copy sequence in the library. High-density clone filters were made from a subset of the library and hybridized with nine pairs of overgos designed from genes involved in fatty acid metabolism. Hybridization results showed that eight overgo pairs were able to identify positive clones from the subset of the library, with an average of 5.3 clones per probe, suggesting that it is suitable for Jatropha genomics and genetics research. Because this library, to our knowledge, represents the first large-insert, plant-transformation-competent BIBAC library for Jatropha, it will provide a vital resource for advanced genomics research, including isolation and characterization of genes and quantitative trait loci, integrative physical mapping and genome sequencing.     

Generation of a high-quality P1 library of Arabidopsis suitable for chromosome walking

Using improved techniques, a representative P1 library of Arabidopsis was constructed and characterized. Megabase genomic DNA was prepared from nuclei and partially digested with Sau3AI. DNA fragments of 75–100 kb were selected by size fractionation in low melting agarose, concentrated by a spot-evaporation/dialysis method, and cloned in the pAd10sacBII P1 vector. The library contains 10 080 clones individually stored in microtiter plates. With an average insert size of about 80 kb, the library represents about eight haploid genomic equivalents of this plant. This library can be screened rapidly by dot hybridization of plate and well position pools. Characterization of the library by restriction analysis, screening with RFLP probes, RFLP mapping of insert end sequences, and chromosome walking shows that the library is of high quality with respect to insert site, completeness, and absence of chimeric artifacts. With this library a contig of about 600 kb has been constructed in the cer9 locus region. This P1 library is expected to be useful for genome mapping and gene cloning in Arabidopsis research programs.     

Construction and characterization of a bacterial artificial chromosome library of banana (<Emphasis Type="Italic">Musa acuminata</Emphasis> Colla)

A bacterial artificial chromosome (BAC) library for banana was constructed from leaves of the wild diploid 'Calcutta 4' clone (Musa acuminata subsp. Burmannicoides 2n = 2 x = 22). 'Calcutta 4' is widely used in breeding programs for its resistance to the current major disease of banana and is being used to build a genetic reference map of banana. As banana leaves are particularly rich in polyphenols and polysaccharides a protocol was adapted to isolate intact nuclei and high-molecular-weight (HMW) DNA. A total of 55,152 clones with an average insert size of 100 kb were picked. The frequency of BAC clones carrying inserts derived from chloroplast and mitochondrial DNA was estimated to be 1.5%. The coverage of the library is equivalent to 9.0-times the haploid genome. The BAC library was screened with 13 RFLP probes belonging to the 8 linkage groups of the consensus molecular map of banana. A total of 135 clones were identified giving an average of 10.38 clones for each locus. This BAC library will be a valuable starting tool for many of the goals of the recently emerged International Musa Genomic Consortium. One of our initial objectives will be to develop a banana physical map by BAC-FISH (fluorescent in situ hybridization) viewing the characterization of translocation break points.     

Construction and Characterization of Two Bacterial Artificial Chromosome Libraries of Grass Carp

Bacterial artificial chromosome (BAC) library is an important tool in genomic research. We constructed two libraries from the genomic DNA of grass carp (Ctenopharyngodon idellus) as a crucial part of the grass carp genome project. The libraries were constructed in the EcoRI and HindIII sites of the vector CopyControl pCC1BAC. The EcoRI library comprised 53,000 positive clones, and approximately 99.94% of the clones contained grass carp nuclear DNA inserts (average size, 139.7 kb) covering 7.4× haploid genome equivalents and 2% empty clones. Similarly, the HindIII library comprised 52,216 clones with approximately 99.82% probability of finding any genomic fragments containing single-copy genes; the average insert size was 121.5 kb with 2.8% insert-empty clones, thus providing genome coverage of 6.3× haploid genome equivalents of grass carp. We selected gene-specific probes for screening the target gene clones in the HindIII library. In all, we obtained 31 positive clones, which were identified for every gene, with an average of 6.2 BAC clones per gene probe. Thus, we succeeded in constructing the desired BAC libraries, which should provide an important foundation for future physical mapping and whole-genome sequencing in grass carp.     

Construction and characterization of a rice YAC library for physical mapping

Genomic libraries of rice,Oryza sativa L. cv. Nipponbare, in yeast artificial chromosomes were prepared for construction of a rice physical map. High-molecular-weight genomic DNA was extracted from cultured suspension cells embedded in agarose plugs. After size fractionation of theEco RI- andNot I-digested DNA fragments, they were ligated with pYAC4 and pYAC55, respectively, and used to transformSaccharomyces cerevisiae AB1380. A total of 6932 clones were obtained containing on average ca. 350 kb DNA. The YAC library was estimated to contain six haploid genome equivalents. The YACs were examined for their chimerism by mapping both ends on an RFLP linkage map. Most YACs withEco RI fragments below 400 kb were intact colinear clones. About 40% of clones were chimeric. Genetic mapping of end clones from large size YACs revealed that the physical distance corresponding to 1 cM genetic distance varies from 120 to 1000 kb, depending on the chromosome region. To select and order YAC clones for making contig maps, high-density colony hybridization using ECL was applied. With several probes, at least one and at most ten YAC clones could be selected in this library. The library size and clone insert size indicate that this YAC library is suitable for physical map construction and map-based cloning.     

Construction of a rice bacterial artificial chromosome library and identification of clones linked to the Xa-21 disease resistance locus

A bacterial artificial chromosome (BAC) library consisting of 11 000 clones with an average DNA insert size of 125 kb was constructed from rice nuclear DNA. The BAC clones were stable in E. coli after 100 generations of serial growth. Transformation of the BAC clones by electroporation into E. coli was highly efficient and increased with decreasing size of the DNA inserts. The library was evaluated for the presence of organellar, repeated, and telomeric sequences. A very low percentage (<0.3%) of the library consisted of chloroplast and mitochondrial clones. Eighteen BACs were identified that hybridized with an Arabidopsis telomere repeat. Sixteen BACs hybridized with the AA genome-specific repetitive sequence pOs48. Twelve clones were isolated that hybridized with three DNA markers linked to the Xa-21 disease resistance locus. The results indicate that the BAC system can be used to clone and manipulate large pieces of plant DNA efficiently.     

Construction and characterisation of a large DNA insert library from the D genome of wheat

 A large DNA fragment library consisting of 144 000 clones with an average insert size of 119 kb was constructed from nuclear DNA isolated from root and leaf tissue from Triticum tauschii (syn. Aegilops tauschii), the D-genome progenitor of wheat. The library was made in a binary vector that had previously been shown to stably maintain large inserts of foreign DNA in Escherichia coli. The use of root nuclei reduced considerably the proportion of the library containing clones derived from chloroplast DNA. Several experimental parameters were investigated and optimised, leading to a high cloning efficiency. Only three ligations were needed to construct the library which was estimated to be equivalent to 3.7 haploid genomes. The accuracy of this estimation was demonstrated by screening this library with three well-defined probes. One probe containing a glutenin gene sequence identified 5 clones covering at least 230 kb of the Glu-D1 locus and contained the two tightly linked high-molecular-weight glutenin genes Glu-D1x and -D1y. Each of the other two single-copy probes derived from the Cre3 cereal cyst nematode resistance gene locus hybridised with 4 clones containing gene sequences encoding nucleotide binding sites and a leucine-rich region. This is the first representative large-insert DNA library for wheat, and the results indicated that large molecules of wheat DNA can be efficiently cloned, stably maintained and manipulated in a bacterial system. Received: 28 August 1998 / Accepted: 28 November 1998     

Construction and characterization of a soybean bacterial artificial chromosome library and use of multiple complementary libraries for genome physical mapping

Two plant-transformation-competent large-insert binary clone bacterial artificial chromosome (hereafter BIBAC) libraries were previously constructed for soybean cv. Forrest, using BamHI or HindIII. However, they are not well suited for clone-based genomic sequencing due to their larger ratio of vector to insert size (27.6 kbp:125 kbp). Therefore, we developed a larger-insert bacterial artificial chromosome (BAC) library for the genotype in a smaller vector (pECBAC1), using EcoRI. The BAC library contains 38,400 clones; about 99.1% of the clones have inserts; the average insert size is 157 kbp; and the ratio of vector to insert size is much smaller (7.5 kbp:157 kbp). Colony hybridization with probes derived from several chloroplast and mitochondrial genes showed that 0.89% and 0.45% of the clones were derived from the chloroplast and mitochondrial genomes, respectively. Considering these data, the library represents 5.4 haploid genomes of soybean. The library was hybridized with six RFLP marker probes, 5S rDNA and 18S-5.8S-25S rDNA, respectively. Each RFLP marker hybridized to about six clones, and the 5S and 18S-5.8S-25S rDNA probes collectively hybridized to 402 BACs—about 1.05% of the clones in the library. The BAC library complements the existing soybean Forrest BIBAC libraries by using different restriction enzymes and vector systems. Together, the BAC and BIBAC libraries encompass 13.2 haploid genomes, providing the most comprehensive clone resource for a single soybean genotype for public genome research. We show that the BAC library has enhanced the development of the soybean whole-genome physical map and use of three complementary BAC libraries improves genome physical mapping by fingerprint analysis of most of the clones of the library. The rDNA-containing clones were also fingerprinted to evaluate the feasibility of constructing contig maps of the rDNA regions. It was found that physical maps for the rDNA regions could not be readily constructed by fingerprint analysis, using one or two restriction enzymes. Additional data to fingerprints and/or different fingerprinting methods are needed to build contig maps for such highly tandem repetitive regions and thus, the physical map of the entire soybean genome.     

Construction of a bacterial artificial chromosome library of Medicago truncatula and identification of clones containing ethylene-response genes

 To facilitate genome analysis and map-based cloning of symbiotic genes in the model legume Medicago truncatula, a bacterial artificial chromosome (BAC) library was constructed. The library consists of 30 720 clones with an average insert size of approximately 100 kb, representing approximately five haploid-genome equivalents. The frequency of BAC clones carrying inserts of chloroplast DNA was estimated to be 1.4%. Screening of the library with single- or low-copy genes as hybridization probes resulted in the detection of 1–12 clones per gene. Hybridization of the library with repeated sequences such as rDNA genes and transposon-like elements of M. truncatula revealed the presence of 60 and 374 BAC clones containing the two sequences, respectively. The BAC library was pooled for screening by polymerase chain reaction (PCR)-amplification. To demonstrate the utility of this system, we used primers designed from a conserved region of the ein3-like loci of Arabidopsis thaliana and isolated six unique BAC clones from the library. DNA gel-blot and sequence analyses showed that these ein3-like clones could be grouped into three classes, an observation consistent with the presence of multiple ein3-like loci in M. truncatula. These results indicate that the BAC library represents a central resource for the map-based cloning and physical mapping in M. truncatula and other legumes. Received: 27 July 1998 / Accepted: 5 August 1998