• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of geneticsGeneticsCurrent IssueInformation for AuthorsEditorial BoardSubscribeSubmit a Manuscript
Genetics. Aug 2008; 179(4): 2299–2312.
PMCID: PMC2516099

Legume Anchor Markers Link Syntenic Regions Between Phaseolus vulgaris, Lotus japonicus, Medicago truncatula and Arachis

Abstract

We have previously described a bioinformatics pipeline identifying comparative anchor-tagged sequence (CATS) loci, combined with design of intron-spanning primers. The derived anchor markers defining the linkage position of homologous genes are essential for evaluating genome conservation among related species and facilitate transfer of genetic and genome information between species. Here we validate this global approach in the common bean and in the AA genome complement of the allotetraploid peanut. We present the successful conversion of ~50% of the bioinformatics-defined primers into legume anchor markers in bean and diploid Arachis species. One hundred and four new loci representing single-copy genes were added to the existing bean map. These new legume anchor-marker loci enabled the alignment of genetic linkage maps through corresponding genes and provided an estimate of the extent of synteny and collinearity. Extensive macrosynteny between Lotus and bean was uncovered on 8 of the 11 bean chromosomes and large blocks of macrosynteny were also found between bean and Medicago. This suggests that anchor markers can facilitate a better understanding of the genes and genetics of important traits in crops with largely uncharacterized genomes using genetic and genome information from related model plants.

THE legume family (Leguminosae) is the third largest family of higher plants and includes >19,000 species (Lewis et al. 2005). Plants belonging to the family are diverse, and trees, shrubby perennials, annual herbs, ornamentals as well as agriculturally important crops are represented. Legumes play a critical role in natural ecosystems, agriculture, and agroforestry, where their ability to establish symbiosis with nitrogen-fixing rhizobial bacteria makes them efficient colonizers of low-nitrogen environments and desirable protein crops. Phaseolus vulgaris L. (common bean) is a particularly important source of protein. It serves as a staple food known as the poor man's meat and contains dietary fiber, minerals, vitamins, and various health promoting compounds (Guillon and Champ 2002; Leterme 2002). Arachis hypogaea (cultivated peanut) is in addition to being consumed in many human foods, the fifth most important oil crop and also a rich source of dietary protein for the chicken and pork industries (Graham and Vance 2003).

The legume family is divided into three subfamilies: Caesalpinioideae, Mimosoideae, and Papilionoideae. Most of the economically important legumes are members of the monophyletic subfamily Papilionoideae, which can be divided into four major clades. Although the legumes included in our study, Lotus japonicus, Medicago truncatula, bean, and Arachis are all in the Papilionoideae subfamily, they belong to three different clades: halogalegina, phaseoloid/millettioid, and aeschynomenoid/dalbergioid. Lotus and Medicago are equally closely related to the bean and equally distantly related to Arachis (Doyle and Luckow 2003). Also the agronomical and genome characteristics of the legumes in our study differ. Both of the model legumes are herbaceous plants of limited agricultural use with relatively small genomes of ~470 Mb, while the common bean and cultivated peanut are major grain legumes with larger genomes of 588 Mbp and 2813 Mbp, respectively (http://data.kew.org/cvalues/homepage.html). In the bean there are >29,000 domesticated and 1300 wild accessions in germplasm banks (Broughton et al. 2003). However, the genetic base of the commercial cultivars of specific market classes is narrow. Less than 5% of the genetic diversity available has been used globally despite nearly a century of organized bean improvement (Broughton et al. 2003). Several genetic maps representing different populations have been established on a backbone of RFLP markers (Vallejos et al. 1992; Nodari et al. 1993; Freyre et al. 1998).

The cultivated peanut is an allotetraploid with an AABB genome complement. The polyploidization event in its origin left the tetraploid cultivated peanut reproductively isolated from its wild diploid relatives (Halward et al. 1993). This fact, combined with self-pollination, has entailed a narrow genetic base and a limited diversity for some traits of agricultural interest (Moretzsohn et al. 2005). No linkage map is available for the cultivated peanut. The maps that have been published in Arachis were based on interspecific crosses of diploid Arachis species or a cross of a synthetic tetraploid with the cultivated peanut (Halward et al. 1993; Burow et al. 2001; Moretzsohn et al. 2005).

Characterizing syntenic relationships between model and crop legumes seems therefore to have wide applications. If collinearity (conservation of gene order) or synteny (conservation of linkage) is found between species, information from large-scale genome sequencing of the Lotus and Medicago gene-rich regions can be used to optimize the only limited genetic and genomic information available from the genomes of bean and Arachis. Following this rationale our objective was to investigate the level and structure of synteny between the legume models (primarily Lotus) and bean and Arachis. The aim was to design robust PCR-based markers requiring relatively simple laboratory equipment and for this purpose we focused on size markers on the basis of insertions/deletions, cleaved amplified polymorphic sequence (CAPS), and derived cleaved amplified polymorphic sequence (dCAPS) markers (Konieczny and Ausubel 1993; Neff et al. 1998). The bioinformatics pipeline, which suggests comparative anchor-tagged sequence (CATS) loci for use within a set of related organisms (Fredslund et al. 2006a) and the automated primer design program PriFi (Fredslund et al. 2005) were used to identify and rank intron-spanning gene-based sets of degenerate primers. To identify conserved primer sites for use throughout the legume family, EST collections originating from Lotus, Medicago, soybean, and, to a limited extent, bean and Arachis were used as input. The pipeline then applied a number of filters optimizing the selection of single-copy orthologous genes and the likelihood of identifying polymorphisms between the mapping parents. Hence primers were designed for conserved exon regions allowing PCR amplification of an intron-spanning fragment, detection of polymorphism, and gene verification by sequencing of the intron and flanking exon regions, respectively (Fredslund et al. 2005, 2006a,b). Applying these criteria, a list of 459 universal Leg-primer sets was generated (http://cgi-www.daimi.au.dk/cgi-chili/GeneticMarkers/table).

We have tested 159 pipeline-generated primer pairs in bean and 124 in Arachis and converted ~50% into legume anchor markers. Subsequent comparison of the corresponding positions of these anchor markers on the linkage maps of Lotus, Medicago, bean, and Arachis showed different levels of genome conservation. A high level of macrosynteny was discovered between bean and Lotus where syntenic regions often span large regions, even entire linkage groups. Substantial regional synteny was also found between bean and Medicago and these regions occasionally cover segments with no apparent bean–Lotus synteny.

MATERIALS AND METHODS

Design of gene-based PCR markers:

Homologous ESTs from mainly Lotus japonicus, Medicago truncatula, and soybean were aligned together with genomic sequence from Lotus or Medicago, and primer sites in conserved exon regions were identified. The bioinformatics pipeline was set to optimize for the following criteria: (1) the targeted gene was either single copy or represented by two copies in Arabidopsis, (2) primer design was based on ESTs where the topology of the gene tree was found to match the species tree indicating orthology of targeted genes, and (3) primer pairs anneal in conserved exon regions allowing PCR amplification of an intron-spanning fragment (Fredslund et al. 2005, 2006a,b). The resulting primers are called Leg primers. At first the work was done manually, later (starting with marker Leg060) the process was automated. In addition, the ESTs were compared to the full set of inferred protein-coding sequences of the reference species Arabidopsis to estimate the number of paralogous sequences. Only sequences expected to be single copy in the legumes were used.

Starting with Leg175, a fully automated bioinformatics approach was used for both, grouping closest homologs, sequence alignment, and primer design (Fredslund et al. 2005, 2006a,b). A number of primer sets (http://bioweb.abc.hu/cgi-mt/pisprim/pisprim.pl) developed by György Kiss's group in Gödöllõ, Hungary, for the Grain Legume Integrated Project (GLIP) were also tested in bean. These primer sets were developed from essentially the same criteria. Three of them were mapped and given names of the type “mtmt_Gen_” followed by three numbers with spacing in between.

The criteria for selecting primers, from the bioinformatics pipeline, for experimental use/marker development were that they should be single copy in Arabidopsis. Homologous single-copy genes in any two genomes are likely to be orthologs, and only true orthologs are suitable for comparative studies. Moreover, the alignment underlying primer design should contain ESTs from at least three different species, or two species from different clades, to represent a broad span of sequence divergence and thereby make it likely that the PCR primers would perform well in distantly related legumes like Arachis. The predicted biological function of the gene, where the primers anneal, was never used as a criterion in the selection process.

DNA extraction and marker development:

Genomic DNA from the mapping parents and for each of the recombinant inbred lines (RILs) were isolated from frozen, young leaf tissue using a modified cetyltrimethylammonium bromide (CTAB) method. Leaf tissue from a single leaf was crushed in liquid nitrogen using mortar and pestle. The tissue was then transferred to a 13-ml Sarstedt tube and CTAB extraction was carried out as described by Sandal et al. (2006) except that the chloroform/isoamylalcohol step was repeated and 10 μl (instead of 5 μl) of RNAse (10 mg/ml stock) were added. The DNA pellet was redissolved in ~100 μl of H2O. Standard PCR conditions were used for the parental survey, where markers were amplified with 94° for 5 min, then 40 cycles of 94° denaturing for 30 sec, 48° (or 45°) annealing for 30 sec, and 72° extension for 2 min, followed by a 5-min final extension at 72°. The PCR reaction was carried out in a final volume of 20 μl containing estimated 0.1 μg of genomic DNA. Single PCR products were purified in the tube using a combination of ExonucleaseI (New England Biolabs, M0293L) and shrimp alkaline phosphatase (USB, 70092Y) in a buffer containing 20 mm Tris–HCl pH 8.0 and 10 mm MgCl2. Multiple PCR products were gel extracted using the QIAquick gel extraction kit (Qiagen, 28706), and except for eluting in 30 μl, the kit protocol was followed. Sequencing was done using ABI BigDye version 3.1. The identity of the amplification products was confirmed by Blastx searches against the Arabidopsis protein database at NCBI or comparison of the translated sequence from the amplification product to the translated sequence of the EST contig underlying primer design. Only amplification products reliably identified were considered further. If there was a significant size difference in the amplification product from the mapping parents, a size marker was made. Otherwise the web-based dCAPS Finder ver. 2.0 software (Neff et al. 2002) was used for suggestion of enzymes for CAPS markers and design of mismatched primers for development of dCAPS markers based on reliable SNPs.

Map calculation and map drawing in bean:

Linkage analysis was performed using JoinMap version 2.0 (Stam 1993). The linkage groups were numbered according to Freyre et al. (1998). MapChart 2.2 was used to draw the map (Voorrips 2002), and map modifications were done in Microsoft PowerPoint. To test the quality of the data and the mapping, the color mapping procedure was used (Kiss et al. 1998).

The map position of the bean sequences representing 18 different standard bean markers with a single best match (E-value < 10−7) to a Lotus sequence with a known map position (supplemental Table S3) were placed on the bean–Lotus/Medicago/Arachis synteny map (Figure 3) along with the Leg markers. Some of these markers (D1032, Bng12, Bng228, and Bng200) were already positioned along with the Leg primers using the BAT93 × JaloEEP558 (B × J) RILs. Others had been mapped either in the B × J F2 population or in other bean populations, and positioning of these on the B × J RIL map were made by manual evaluation according to shared markers and cM distances on the bean integrated linkage maps (http://www.ncbi.nlm.nih.gov/mapview/map_search.cgi?taxid=3885) (this applies to D1051, Bng72, Bng108, Bng148/R, Mng415, Bng57, Bng3, Hyp4-1, Bng179, Bng87, Bng95, Bng26, Bng206, and Bng1).

Figure 3. Figure 3.
Synteny between bean, Lotus, Medicago, and Arachis defined by legume anchor markers. Synteny to Lotus japonicus (Lj) and Medicago truncatula (Mt) is given by the color code. The color scheme reflects the synteny observed between Mt and Lj. For example, ...

Linkage analysis in Lotus:

For the anchor markers that could not be mapped in the Gifu × MG20 population, primers were tested for polymorphism between L. japonicus Gifu and the related species L. filicaulis (Sandal et al. 2006). Lotus BAC/TAC sequences and the web-based computer program Tandem Repeat Occurrence Locator (TROLL) (http://wsmartins.net/webtroll/troll.html) were used for identifying microsatellites. Due to suppression of recombination and resulting poor resolution in parts of the Lotus genetic map (Sandal et al. 2006), some of the Leg markers were not mapped to a specific location on the Lotus map, but rather to a region. If this region was <10 cM the marker was artificially put on the map on a location equivalent to the midpoint of the interval where it was mapped (this applies to 12M-Gm, 33M-Gm, Leg036, Leg055, Leg081, Leg183, Leg187, Leg191, Leg203, Leg213, Leg228, and Leg231). Leg markers mapping to a region >10 cM are not shown on the Lotus map (this applies to Leg111, Leg177, Leg202, Leg206, and Leg762).

RESULTS

Gene-based PCR markers:

To validate the application of the Leg primers for comparative mapping, a subset corresponding to single-copy genes in Arabidopsis was selected for generating legume anchor markers in bean and Arachis. The chosen Leg primers were used to amplify PCR fragments from the BAT93 and JaloEEP558 parents of the bean RILs (Freyre et al. 1998), as well as the A. duranensis K7988 and A. stenosperma V10309 parents (Moretzsohn et al. 2005) of the Arachis population, subsequently used for mapping. Following the flow scheme outlined in Figure 1 an overlapping set of 159 and 124 primer pairs was tested in bean and Arachis, respectively. To verify the identity of amplified fragments, the terminal exon sequences of the PCR fragments were compared to a translated version of aligned ESTs used for primer design or alternatively a Blastx search was performed (Figure 1). For bean, ~70% of the PriFi primer pairs on the basis of ESTs from four, three, or two different legume species amplified a product that was verified (supplemental Table S1). In Arachis ~55% of the primer pairs on the basis of four or three ESTs produced a PCR fragment that could be verified (supplemental Table S1). Following a positive confirmation of gene identity, sequence polymorphism between the mapping parents was used to develop the CAPS, dCAPS, or size markers applied for mapping the Leg-marker loci. In bean, 50% of the original primer sets tested were developed into markers and in Arachis the yield was 45% (Table 1). Prior to the development of the bioinformatics pipeline a number of markers were developed manually using the same principles, and adding these markers a total of 111 bean markers were made.

Figure 1.
Flow scheme of marker production. The Leg primers were tested on DNA from mapping parents. Single products were enzymatically purified before sequencing, verification of gene identity, and detection of polymorphism. Multiple products were gel extracted ...
TABLE 1
Summary of primer efficiency for exhaustively tested primer pairs

Gene-based genetic map of the bean:

To develop the bean genetic linkage map, 77 lines from the recombinant inbred (RI) population B × J (Freyre et al. 1998) were used. The segregation information collected from the 111 Leg markers in this study was analyzed, along with data from 150 framework markers from the core linkage map (Freyre et al. 1998). Using JoinMap version 2.0 (Stam 1993) and a 3.5-LOD threshold, we were able to assign 104 Leg markers to 99 loci in 11 linkage groups, corresponding to the haploid chromosome number of bean (Figure 2). The linkage groups range from 40.8 cM (PvLG10) to 99.8 cM (PvLG2), and the total map length is 758 cM (Table 2). The number of Leg markers per chromosome range from 3 to 19 with an average of 9.5 and the average interval between Leg markers is 7.3 cM. The generated Leg markers include 15 polymorphic size markers, 48 CAPS, and 48 dCAPS markers. Marker information can be seen in supplemental Table S2.

Figure 2.
Genetic linkage map of the bean. Each vertical bar represents individual bean linkage groups (LG). Marker names are shown on the right side of each linkage group, connecting lines indicate the position of each locus on the linkage map, and centimorgan ...
TABLE 2
Synteny summary of the 11 Phaseolus vulgaris linkage groups

Besides the described linkage groups, one Leg marker (7-Gm) showed linkage to one of the framework markers (DROD3a) and four Leg markers (Leg191, Leg194, Leg223, and Leg232) came out as unlinked loci.

Mapping in both the B × J and the DOR364 × G19833 (D × G) population, had previously proved useful for positioning microsatellites that were monomorphic in one or the other of the populations (Blair et al. 2003). In an attempt to improve the marker yield 17 different Leg-primer pairs, which amplified a correct but monomorphic product in B × J, were tested in the D × G mapping parents. However, a SNP polymorphism was detected in only two of the PCR fragments (data not shown) implying that a mapping population originating from more distantly related parents is optimal for this approach.

Allele frequencies in the B × J RI population:

The allele frequency was calculated for each marker locus to evaluate the degree of deviation from the expected 0.5 transmission frequency for each parental allele at each locus of the RIL lines. All markers with segregation data were subjected to this analysis, including the framework markers and interestingly there were large regional differences in allele frequencies. At the bottom part of PvLG7 there is a major segregation distortion. The maternal BAT93 alleles are overrepresented compared to the paternal JaloEEP558 alleles (supplemental Figure S1). The segregation distortion has a maximum at the bottom of PvLG7 with 85% of the alleles originating from BAT93, but is normalized to the expected value of 50% at the middle of PvLG7 after ~35 cM. This linkage group was previously reported to have a preference for the BAT93 allele (Nodari et al. 1993; Freyre et al. 1998). On PvLG10, a modest but consistent segregation distortion was observed. PvLG10 carries only 11 markers, but at all loci, the transmission frequency was higher for the BAT93 alleles. Moreover, on PvLG1, PvLG4, and PvLG6 large regions of segregation distortion were found. On PvLG1 and PvLG6 the JaloEEP558 alleles were favored in regions of ~30 cM, whereas on PvLG4 the BAT93 alleles were more prominent, again in a region of ~30 cM. The preference for the Andean JaloEEP558 alleles on linkage group PvLG1 is consistent with the observations of Freyre et al. (1998) and Nodari et al. (1993). Preferential transmission of parental alleles could be caused by an allele specific advantage in viability or fertility and Leg markers may represent or be linked to alleles selected during the eight generations used for development of the RILs (Freyre et al. 1998).

Bean–Lotus synteny:

To evaluate the extent of macrosynteny between Lotus and bean the individual positions of Leg markers on the two genetic linkage maps were compared. Most of the Leg-marker positions in Lotus were known from their presence on BAC or TAC clones, which were mapped in the intraspecific cross between L. japonicus ecotypes Gifu and L. japonicus Miyakojima MG-20 (Hayashi et al. 2001; Kato et al. 2003) and the collinear L. filicaulis × L. japonicus Gifu map (Sandal et al. 2006). Where map positions were absent, we took advantage of the Lotus genome-sequencing program. A corresponding BAC or TAC clone was included in the genome sequencing and genetically mapped using the standard microsatellite marker procedure. A total of 81 of the legume anchor markers were positioned on the genetic map of both bean and Lotus (Figure 3). To add more loci, 99 of the previously developed standard bean markers from the Build 1.1 linkage map (http://www.ncbi.nlm.nih.gov/mapview/map_search.cgi?taxid=3885) that had sequence information attached was searched (Blastn) against the Lotus genome sequence. This approach found bean sequences representing 18 different bean markers with a single best match (E-value < 10−7) to a Lotus sequence with a known map position on the Lotus genetic map (supplemental Table S3).

A total of 99 unique positions were shared between bean and Lotus, ranging from 2 (PvLG5) to 20 (PvLG2) shared markers per linkage group and an average of 9 common markers per linkage group. Extensive synteny between Lotus and bean was found for about half of the bean linkage groups. In PvLG4, PvLG7, PvLG9, and PvLG11 the syntenic regions span entire linkage groups (with a maximum of one nonsyntenic marker) and in PvLG2 and PvLG8, half a linkage group. PvLG5 and PvLG10 have only 2 and 3 markers, respectively, shared with Lotus. Only 4 of the 11 bean linkage groups have markers that map to >2 different Lotus linkage groups, and even among these, 3 clearly show synteny toward one particular Lotus chromosome. For example, 6 of 11 markers on PvLG3 belong to LjLG4, 7 of 12 markers on PvLG6 belong to LjLG6, and 5 of 8 markers on PvLG8 belong to LjLG2. PvLG1 contains 9 shared markers mapping to 3 different Lotus linkage groups.

The syntenic marker positions on Lotus and bean chromosomes shown in Figure 3 indicate that only a limited number of large-scale rearrangements occurred during the evolution of legume chromosomes. To evaluate this suggestion further, marker collinearity was investigated. Three examples of this analysis are shown in Figure 4. It is clear that the blocks of syntenic loci are only partly collinear and that smaller internal chromosome rearrangements have occurred in addition to the large scale rearrangements. Figure 4 (bottom) shows an example of one bean linkage group having blocks of anchor loci shared with two Lotus linkage groups and one Lotus linkage group having syntenic blocks shared with two bean linkage groups.

Figure 4.
Detailed examples of macrosynteny between bean (Pv) and Lotus japonicus (Lj). Each vertical bar represents individual bean and Lotus linkage groups (LG). Marker names are shown on one side of each linkage group and centimorgan values are given on the ...

Bean–Medicago synteny:

Positions of Leg markers in Medicago were found by Blastn searches against the Medicago pseudomolecules (http://www.medicago.org/genome/cvit_blast.php), using a combination of bean sequences amplified with the Leg primers and the Medicago EST sequence included in the CATS alignment for Leg-primer design. Search information is presented in supplemental Table S4. For the previously developed bean standard markers with sequence information known, the bean sequence was used (supplemental Table S3). Among the Leg markers, 65 had a match with E-value < 10−7 in Medicago, and among the 18 marker sequences 10 had a match with E-value < 10−7. All in all this amounts to 75 shared positions between bean and Medicago, ranging from 3 (PvLG5 and PvLG7) to 15 (PvLG2) markers per bean linkage group and an average of 6.8 markers per linkage group (Figure 3). Substantial synteny is seen. Two of the bean linkage groups, having 3 (PvLG5) and 5 (PvLG9) shared positions to Medicago, respectively, have only markers mapping to a single Medicago chromosome and at the bottom part of PvLG2 a large syntenic segment spanning 52 cM is seen. Six of 11 bean linkage groups have markers that map to >2 different Medicago linkage groups, but 5 of these have 1 (PvLG3 and PvLG11) or 2 (PvLG2, PvLG6, and PvLG8) uninterrupted syntenic regions of at least 3 markers. A total of 11 uninterrupted regions with ≥3 shared markers are distributed over the bean linkage groups.

Figure 5 shows two detailed alignments between bean and Medicago. In both examples, blocks of loci show synteny, but only partial collinearity indicating small internal chromosome rearrangements.

Figure 5.
Detailed examples of macrosynteny between bean (Pv) and Medicago truncatula (Mt). Vertical bars represent individual bean linkage groups (LG) or segments of Medicago pseudochromosomes (hatched). Marker positions are given in centimorgans for bean and ...

Bean–Arachis synteny:

Positions of Leg markers in Arachis were found by mapping in a diploid F2 mapping population, originating from a cross between A. duranensis K7988 and A. stenosperma V10309 (Moretzsohn et al. 2005). Eighty-two Leg markers were mapped in Arachis and 38 of them have a linkage map position in bean. The number of shared loci between bean and Arachis range from zero (PvLG1) to eight (PvLG3), with an average of 3.5 shared loci per bean linkage group. In three of the bean linkage groups (PvLG2, PvLG4, and PvLG8) nondisrupted syntenic segments of ≥3 loci mapped to the same Arachis linkage group. Of these, PvLG4 and PvLG8 only shared markers to one Arachis linkage group. However, several of the bean linkage groups have a patchy pattern of loci shared with more Arachis linkage groups (Figure 3).

Figure 6 shows two detailed examples of macrosynteny between Medicago, bean, Lotus, and Arachis. The syntenic blocks between Arachis and Lotus consist of three and four markers, respectively.

Figure 6.
Detailed examples of macrosynteny between bean (Pv), Lotus japonicus (Lj), Arachis (Ara), and Medicago truncatula (Mt). Vertical bars represent individual bean, Lotus, and Arachis linkage groups (LG) or segments of Medicago pseudochromosomes (hatched). ...

Bean–Lotus–Medicago–Arachis synteny:

Besides giving a quick overview of the pairwise synteny observed between bean, Lotus, Medicago, and Arachis, respectively, the color scheme given in Figure 3 reflects the previously described synteny blocks between Lotus and Medicago (Cannon et al. 2006). Accordingly, it is possible to relate the synteny detected by Leg markers to the chromosome correspondences between the two model legumes. In the large-scale sequence-based comparison reported by Cannon et al. (2006) 10 large-scale synteny blocks were identified. Notably most of the syntenic regions we have identified fall into these blocks. For example, three large synthenic segments between bean, Lotus, and Medicago were found on PvLG2 and PvLG9 (Figure 3 and Table 2). Our synteny analysis therefore supports and is supported by the large-scale sequence-based comparison.

Defining syntenic segments as noninterrupted regions with at least three markers, 48% of the bean linkage map is syntenic to Lotus (367 of 758 cM) and 35% is syntenic to Medicago (263 of 758 cM). However, bean has more markers shared with Lotus than with Medicago and therefore the numbers are not directly comparable.

Bean–Arabidopsis synteny:

In our setup for primer design, we have used the GenBank accession number of the Arabidopsis ortholog as a reference for the alignment of legume EST sequences, and therefore we have shared positions for some 100 loci in bean and Arabidopsis. Five of the bean linkage groups (PvLG3, PvLG5, PvLG6, PvLG8, and PvLG11) have uninterrupted regions of ≥3 loci extending up to 33 cM (Leg133–Leg220 on PvLG11), which share linkage on an Arabidopsis chromosome (supplemental Figure S2). Given the ~112 million years thought to separate Arabidopsis from the legumes (Bowers et al. 2003), the gene linkage is remarkable although macrosynteny has been observed before (Grant et al. 2000). Moreover, our experimental setup focuses on single-copy and highly conserved genes, and this group of genes is optimal for identifying ancestral synteny.

DISCUSSION

Here we present the application of an automated and systematic approach for developing anchor markers in two distantly related legumes, bean and Arachis, chosen for testing the feasibility of the approach. Among the 159 Leg-primer pairs created by the bioinformatics pipeline, which were exhaustively tested in bean and an overlapping set of 124 Leg-primer pairs tested in Arachis, 113 (71%) amplified the correct product in bean and 67 (54%) amplified the correct product in Arachis. PriFi suggests and ranks 4 different primer pairs per alignment. For each of these we have only used the highest ranking and tested rankings from 187 to 22 without finding a difference in efficiency between high-ranking and low-ranking pairs. Use of ≥1 of the 3 alternative PriFi primer sets would therefore most likely increase the marker yield. Similarly, there were no major differences in efficiency between primer pairs based on ESTs from four, three, or two different legume species. The PriFi settings used to generate the 459 primer sets appear therefore to be robust. Since the Arachis-mapping parents were more polymorphic than the bean-mapping parents the marker yield was nearly the same, 50% vs. 45%. Here we present a bean linkage map with 104 new legume anchor-marker positions. The linkage map adds additional genetic resolution to the existing bean maps and expands the map with 1.9 cM at the outermost top of PvLG6 and 4.1 cM at the top of PvLG10. Leg markers distribute over all of the 11 bean linkage groups, and in only five cases do 2 Leg markers share the same locus. This absence of clustering may indicate that suppression of recombination is rare in the B × J RILs. The distribution of Leg markers and thus the genetic resolution varies. For example, linkage group PvLG2 has 19 Leg markers, whereas PvLG10 has only 3 Leg markers. This difference might be a consequence of the bioinformatics setup designing Leg primers. The bioinformatics pipeline selects primers in regions of high sequence conservation and gives preference to highly expressed genes represented in EST collections. Neither conserved genes nor highly expressed genes are necessarily randomly distributed in genomes, and an inverse relationship between the frequency of single-copy genes and clustered retrotransposons has been observed (D. Bertioli, M. Moretzsohn, L. H. Madsen, N. Sandal, S. Leal-Bertioli, P. Guimarães, B. K. Hougaard, J. Fredslund, L. Schauser, A. M. Nielsen, S. Sato, S. Tabata, S. Cannon and J. Stougaard, unpublished results). Most importantly our results provide proof of concept for the bioinformatics-driven identification of anchor-tagged loci that, together with positional information from homologous loci in Lotus, Medicago, and Arachis, provide a basis for comparative genomics and an overview of the synteny among these four legumes.

Synteny:

One of the ideas behind the model-legume concept was to exploit synteny between model and crops to accelerate isolation and comparative characterization of genes from the less characterized crop legumes. This approach requires that the target genome regions contain the same genes in approximately the same order in models and crops. A substantial effort has been invested in investigating this question. Regional synteny between Medicago and/or Lotus models and pea (Aubert et al. 2006), alfalfa (Choi et al. 2004a), red clover (Sato et al. 2005), lens (Phan et al. 2006, 2007), and lupin (Nelson et al. 2006; Phan et al. 2006) lend support to this approach. In the present analysis, we have anchor-tagged loci covering 758 cM of the bean genetic map and a set of 99 shared loci made it possible to compare this map with the genetic map of Lotus. All of the 11 bean linkage groups had noninterrupted regions of at least two markers also showing linkage in the Lotus genome, and in several of the bean linkage groups synteny spanned entire linkage groups. All in all, the legume anchor markers revealed a broad conservation of gene linkage (macrosynteny) between bean and Lotus.

Seventy-five loci were shared between bean and Medicago and substantial conservation of gene linkage between bean and Medicago was found (Figure 3). However, even though Medicago and Lotus phylogenetically are equally distant from bean, the conserved bean–Medicago segments are generally shorter than observed for bean–Lotus. Looking at uninterrupted syntenic regions >20 cM, five can be found between bean and Medicago [PvLG2 (two regions) PvLG6, PvLG9, and PvLG10] whereas seven are seen between bean and Lotus [PvLG2 (two regions), PvLG4, PvLG7, PvLG8, PvLG9, and PvLG11]. In certain regions linkage is more conserved between bean and Medicago than between bean and Lotus, for example, the region in PvLG3 flanked by Leg187 and Leg164N. Here five Leg markers all map to MtChr8, whereas the same region maps to four different Lotus linkage groups. Likewise, in PvLG6 the region flanked by Leg217 and Leg081 have four Leg markers all mapping to MtChr7, whereas in Lotus, they map to three different linkage groups. However, in PvLG2, PvLG4, PvLG7, and PvLG11 macrosynteny between bean and Lotus is more pronounced than between bean and Medicago.

Despite the fact that bean and Arachis have similar numbers of chromosomes (11 and 10, respectively), the wild Arachis species have genome sizes that are considerably larger than the genomes of bean and Lotus. A. duranensis has a gametic DNA content (1C) of 1243 Mbp (http://data.kew.org/cvalues/homepage.html). Moreover, Arachis is a rather distant relative of bean and Lotus, so the evidence of macrosynteny documented between Arachis, bean, Lotus, and Medicago is remarkable. In both of the detailed examples shown in Figure 6, the loci shared between Lotus and Arachis span a larger region in Lotus than in Arachis. Given the difference in DNA content between these two species, this seems surprising. However, whereas the average physical equivalent of 1 cM corresponds to 1.97 Mbp in Arachis (Moretzsohn et al. 2005), Lotus chromosome 1 has a mean ratio of 0.77 Mbp/cM (Pedrosa et al. 2002).

A comparative map spanning cold- and warm-season legumes was presented by Choi et al. (2004b). In their study, marker segregation was analyzed in M. truncatula, M. sativa, P. sativum, V. radiata, and bean. Using a limited number of markers, 8 of 11 bean linkage groups were included from their already-described relation to V. radiata (mung bean) (Boutin et al. 1995). No markers were directly shared between bean and Medicago and bean–Medicago synteny was inferred using several bridging legume species. The bean–Medicago relations presented here are documented by directly corresponding shared loci and include all of the 11 bean linkage groups. In addition, 9 of the bean linkage groups show synteny to >1 Medicago linkage group, and in this study we have precise indication of which regions on the individual bean linkage groups share homology to which Medicago chromosomes.

Macrosynteny between bean and Lotus was described by Zhu et al. (2005). A circle diagram included 8 of 11 bean linkage groups and the relations to Lotus were inferred through positions shared between Lotus and Medicago. In contrast to Zhu et al. (2005), we present markers directly bridging the genomes of bean and Lotus and thereby we define regions of shared macrosynteny directly on each of the 11 bean linkage groups.

Previous studies have demonstrated substantial macrosynteny between Lotus and Medicago (Choi et al. 2004b; Young et al. 2005; Cannon et al. 2006), and in a bioinformatics approach comparing 121 Mbp of nonredundant map-anchored genomic sequence from the Lotus build with 149 Mbp from the Medicago build, 10 large-scale synteny blocks were identified (Cannon et al. 2006). On the bean–Lotus/Medicago/Arachis synteny map (Figure 3), the color code reflects the described synteny between Lotus and Medicago. As seen, in large parts of the bean genome, the synteny to Lotus and Medicago converges on the 10 large-scale synteny blocks defined by direct sequence comparison. We infer that large syntenic segments are conserved in the genomes of legumes and that anchor markers can link these segments between different species and directly to the genome sequence of the model legumes.

Given the extensive linkage observed between bean and the two model legumes in this study, we therefore conclude that Lotus and Medicago will have a significant potential for future identification of tightly linked markers for marker-assisted selection and map-based isolation of candidate genes.

Acknowledgments

Paul Gepts and Valérie Geffroy kindly provided seeds for the BAT93 x Jalo EEP558 population. Joe Tohme and Matthew Blair provided seeds for the DOR364 x G19833 mapping parents and DNA from the corresponding mapping population. The work of B.K.H. was funded by the Council for Development Research (RUF) under the Danish Ministry of Foreign Affairs (project no. 91210). In addition, there has been funding by The European Union INCO-DEV Programme (ARAMAP reference: ICA4-2001-10072); the Generation Challenge Program (project no. 31); the EU FP6 Grain Legume Integrated Project (GLIP reference: FOOD-CT-2004-506223); and the Danish National Research Foundation (Centre for Carbohydrate Recognition and Signalling).

References

  • Aubert, G., J. Morin, F. Jacquin, K. Loridon, M. C. Quillet et al., 2006. Functional mapping in pea, as an aid to the candidate gene selection and for investigating synteny with the model legume Medicago truncatula. Theor. Appl. Genet. 112 1024–1041. [PubMed]
  • Blair, M. W., F. Pedraza, H. F. Buendia, E. Gaitan-Solis, S. E. Beebe et al., 2003. Development of a genome-wide anchored microsatellite map for common bean (Phaseolus vulgaris L.). Theor. Appl. Genet. 107 1362–1374. [PubMed]
  • Boutin, S. R., N. D. Young, T. C. Olson, Z.-H. Yu, R. C. Shoemaker et al., 1995. Genome conservation among three legume genera detected with DNA markers. Genome 38 928–937. [PubMed]
  • Bowers, J. E., B. A. Chapman, J. Rong and A. H. Paterson, 2003. Unravelling angiosperm genome evolution by phylogenetic analysis of chromosomal duplication events. Nature 422 433–438. [PubMed]
  • Broughton, W. J., G. Hernández, M. Blair, S. Beebe, P. Gepts et al., 2003. Beans (Phaseolus spp.): model food legumes. Plant Soil 252 55–128.
  • Burow, M. D., C. E. Simpson, J. L. Starr and A. H. Paterson, 2001. Transmission genetics of chromatin from a synthetic amphidiploid to cultivated peanut (Arachis hypogaea L.): broadening the gene pool of a monophyletic polyploid species. Genetics 159 823–837. [PMC free article] [PubMed]
  • Cannon, S. B., L. Sterck, S. Rombauts, S. Sato, F. Cheung et al., 2006. Legume genome evolution viewed through the Medicago truncatula and Lotus japonicus genomes. Proc. Natl. Acad. Sci. USA 103 14959–14964. [PMC free article] [PubMed]
  • Choi, H. K., D. Kim, T. Uhm, E. Limpens, H. Lim et al., 2004. a A sequence-based genetic map of Medicago truncatula and comparison of marker colinearity with M. sativa. Genetics 166 1463–1502. [PMC free article] [PubMed]
  • Choi, H. K., J. H. Mun, D. J. Kim, H. Zhu, J. M. Baek et al., 2004. b Estimating genome conservation between crop and model legume species. Proc. Natl. Acad. Sci. USA 101 15289–15294. [PMC free article] [PubMed]
  • Doyle, J. J., and M. A. Luckow, 2003. The rest of the iceberg. Legume diversity and evolution in a phylogenetic context. Plant Physiol. 131 900–910. [PMC free article] [PubMed]
  • Fredslund, J., L. Schauser, L. H. Madsen, N. Sandal and J. Stougaard, 2005. PriFi: using a multiple alignment of related sequences to find primers for amplification of homologs. Nucleic Acids Res. 33 W516–W520. [PMC free article] [PubMed]
  • Fredslund, J., L. H. Madsen, B. K. Hougaard, A. M. Nielsen, D. Bertioli et al., 2006. a A general pipeline for the development of anchor markers for comparative genomics in plants. BMC Genomics 7 207. [PMC free article] [PubMed]
  • Fredslund, J., L. H. Madsen, B. K. Hougaard, N. Sandal, J. Stougaard et al., 2006. b GeMprospector–online design of cross-species genetic marker candidates in legumes and grasses. Nucleic Acids Res. 34 W670–W675. [PMC free article] [PubMed]
  • Freyre, R., P. W. Skroch, V. Geffroy, A.-F. Adam-Blondon, A. Shirmohamadali et al., 1998. Towards an integrated linkage map of common bean. 4. Development of a core linkage map and alignment of RFLP maps. Theor. Appl. Genet. 97 847–856.
  • Graham, P. H., and C. P. Vance, 2003. Legumes: importance and constraints to greater use. Plant Physiol. 131 872–877. [PMC free article] [PubMed]
  • Grant, D., P. Cregan and R. C. Shoemaker, 2000. Genome organization in dicots: genome duplication in Arabidopsis and synteny between soybean and Arabidopsis. Proc. Natl. Acad. Sci. USA 97 4168–4173. [PMC free article] [PubMed]
  • Guillon, F., and M. M. Champ, 2002. Carbohydrate fractions of legumes: uses in human nutrition and potential for health. Br. J. Nutr. 88 S293–S306. [PubMed]
  • Halward, T., H. T. Stalker and G. Kochert, 1993. Development of an RFLP linkage map in diploid peanut species. Theor. Appl. Genet. 87 379–384. [PubMed]
  • Hayashi, M., A. Miyahara, S. Sato, T. Kato, M. Yoshikawa et al., 2001. Construction of a genetic linkage map of the model legume Lotus japonicus using an intraspecific F2 population. DNA Res. 8 301–310. [PubMed]
  • Kato, T., S. Sato, Y. Nakamura, T. Kaneko, E. Asamizu et al., 2003. Structural analysis of a Lotus japonicus genome. V. Sequence features and mapping of sixty-four TAC clones which cover the 6.4 mb regions of the genome. DNA Res. 10 277–285. [PubMed]
  • Kiss, G. B., A. Kereszt, P. Kiss and G. Endre, 1998. Colormapping: a non-mathematical procedure for genetic mapping. Acta Biol. Hung. 49 125–142.
  • Konieczny, A., and F. M. Ausubel, 1993. A procedure for mapping Arabidopsis mutations using co-dominant ecotype-specific PCR-based markers. Plant J. 4 403–410. [PubMed]
  • Leterme, P., 2002. Recommendations by health organizations for pulse consumption. Br. J. Nutr. 88 S239–S242. [PubMed]
  • Lewis, G., B. Schrire, B. Mackinder and M. Lock, 2005. Legumes of the World. The Royal Botanic Gardens, Kew, Richmond, UK.
  • Moretzsohn, M. C., L. Leoi, K. Proite, P. M. Guimaraes, S. C. Leal-Bertioli et al., 2005. A microsatellite-based, gene-rich linkage map for the AA genome of Arachis (Fabaceae). Theor. Appl. Genet. 111 1060–1071. [PubMed]
  • Neff, M. M., J. D. Neff, J. Chory and A. E. Pepper, 1998. dCAPS, a simple technique for the genetic analysis of single nucleotide polymorphisms: experimental applications in Arabidopsis thaliana genetics. Plant J. 14 387–392. [PubMed]
  • Neff, M. M., E. Turk and M. Kalishman, 2002. Web-based primer design for single nucleotide polymorphism analysis. Trends Genet. 18 613–615. [PubMed]
  • Nelson, M. N., H. T. Phan, S. R. Ellwood, P. M. Moolhuijzen, J. Hane et al., 2006. The first gene-based map of Lupinus angustifolius L.: location of domestication genes and conserved synteny with Medicago truncatula. Theor. Appl. Genet. 113 225–238. [PubMed]
  • Nodari, R. O., S. M. Tsai, R. L. Gilbertson and P. Gepts, 1993. Towards an integrated linkage map of common bean 2. Development of an RFLP-based linkage map. Theor. Appl. Genet. 85 513–520. [PubMed]
  • Pedrosa, A., N. Sandal, J. Stougaard, D. Schweizer and A. Bachmair, 2002. Chromosomal map of the model legume Lotus japonicus. Genetics 161 1661–1672. [PMC free article] [PubMed]
  • Phan, H. T. T., S. R. Ellwood, R. Ford, S. Thomas and R. Oliver, 2006. Differences in syntenic complexity between Medicago Truncatula with Lens culinaris and Lupinus albus. Funct. Plant Biol. 33 775–782.
  • Phan, H. T., S. R. Ellwood, J. K. Hane, R. Ford, M. Materne et al., 2007. Extensive macrosynteny between Medicago truncatula and Lens culinaris ssp. culinaris. Theor. Appl. Genet. 114 549–558. [PubMed]
  • Sandal, N., T. R. Petersen, J. Murray, Y. Umehara, B. Karas et al., 2006. Genetics of symbiosis in Lotus Japonicus: recombinant inbred lines, comparative genetic maps, and map position of 35 symbiotic loci. Mol. Plant-Microbe Interact. 19 90–91. [PubMed]
  • Sato, S., S. Isobe, E. Asamizu, N. Ohmido, R. Kataoka et al., 2005. Comprehensive structural analysis of the genome of red clover (Trifolium pratense L.). DNA Res. 12 301–364. [PubMed]
  • Stam, P., 1993. Construction of integrated genetic linkage maps by means of a new computer package: Joinmap. Plant J. 3 739–744.
  • Vallejos, C. E., N. S. Sakiyama and C. D. Chase, 1992. A molecular marker-based linkage map of Phaseolus vulgaris L. Genetics 131 733–740. [PMC free article] [PubMed]
  • Voorrips, R. E., 2002. MapChart: software for the graphical presentation of linkage maps and QTLs. J. Hered. 93 77–78. [PubMed]
  • Young, N. D., S. B. Cannon, S. Sato, D. Kim, D. R. Cook et al., 2005. Sequencing the genespaces of Medicago truncatula and Lotus japonicus. Plant Physiol. 137 1174–1181. [PMC free article] [PubMed]
  • Zhu, H., H.-K. Choi, D. R. Cook and R. C. Shoemaker, 2005. Bridging model and crop legumes through comparative genomics. Plant Physiol. 137 1189–1196. [PMC free article] [PubMed]

Articles from Genetics are provided here courtesy of Genetics Society of America
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...