pmc logo image
Logo of geneticsJournal URL: http://www.genetics.org/
Journal List > Genetics > v.181(2); Feb 2009

Formats:
Genetics. 2009 February; 181(2): 405–419.
doi: 10.1534/genetics.108.093583.
PMCID: PMC2644936
Copyright © 2009 by the Genetics Society of America
Molecular Analysis of a Large Subtelomeric Nucleotide-Binding-Site–Leucine-Rich-Repeat Family in Two Representative Genotypes of the Major Gene Pools of Phaseolus vulgaris
Valérie Geffroy,*1 Catherine Macadré,* Perrine David,* Andrea Pedrosa-Harand,†2 Mireille Sévignac,* Catherine Dauga,‡ and Thierry Langin*
*Institut de Biotechnologie des Plantes, UMR-CNRS 8618, INRA, Université Paris-Sud, 91 405 Orsay, France, †Department of Chromosome Biology, University of Vienna, 1030 Vienna, Austria and ‡Plate-Forme 4-Intégration et Analyse Génomiques, Génopole, Institut Pasteur, 75724 Paris, Cedex 15, France
1Corresponding author: Institut de Biotechnologie des Plantes, Bât. 630 UMR-CNRS 8618, INRA Université Paris-Sud, 91 405 Orsay, France. E-mail: valerie.geffroy/at/u-psud.fr
2Present address: Laboratório de Citogenética Vegetal, Departamento de Botânica–CCB, Universidade Federal de Pernambuco, Recife–PE, 50670-420, Brazil.
Communicating editor: T. Brutnell
Received July 4, 2008; Accepted December 11, 2008.
Abstract
In common bean, the B4 disease resistance (R) gene cluster is a complex cluster localized at the end of linkage group (LG) B4, containing at least three R specificities to the fungus Colletotrichum lindemuthianum. To investigate the evolution of this R cluster since the divergence of Andean and Mesoamerican gene pools, DNA sequences were characterized from two representative genotypes of the two major gene pools of common bean (BAT93: Mesoamerican; JaloEEP558: Andean). Sequences encoding 29 B4-CC nucleotide-binding-site–leucine-rich-repeat (B4-CNL) genes were determined—12 from JaloEEP558 and 17 from BAT93. Although sequence exchange events were identified, phylogenetic analyses revealed that they were not frequent enough to lead to homogenization of B4-CNL sequences within a haplotype. Genetic mapping based on pulsed-field gel electrophoresis separation confirmed that the B4-CNL family is a large family specific to one end of LG B4 and is present at two distinct blocks separated by 26 cM. Fluorescent in situ hybridization on meiotic pachytene chromosomes revealed that two B4-CNL blocks are located in the subtelomeric region of the short arm of chromosome 4 on both sides of a heterochromatic block (knob), suggesting that this peculiar genomic environment may favor the proliferation of a large R gene cluster.
 
PLANT disease is one of the major limitations in crop production throughout the world and is responsible for huge economic losses (Madden and Wheelis 2003). Use of resistant genotypes is the most economic and ecologically safe means for controlling plant diseases (Hulbert et al. 2001; Hammond-Kosack and Parker 2003; Michelmore 2003). In the past 15 years, >60 resistance (R) genes following the classic gene-for-gene model (Flor 1955) have been cloned from various plant species (Martin et al. 2003; Meyers et al. 2005). The most prevalent plant R genes encode proteins containing a nucleotide-binding site (NBS) and a C-terminal leucine-rich-repeat (LRR) domain (Martin et al. 2003; Belkhadir et al. 2004; Rairdan and Moffett 2007). R genes belonging to this class have been identified in various plant species, in monocots as well as in dicots, and correspond to R genes effective against all types of pathogens and pests, including fungi, bacteria, viruses, nematodes, oomycetes, and insects (Dangl and Jones 2001; Hammond-Kosack and Parker 2003; Mchale et al. 2006). This NBS–LRR protein class can be divided into two subclasses on the basis of their amino-terminal sequence, corresponding to two ancient lineages (Bai et al. 2002; Meyers et al. 2003; Ameline-Torregrosa et al. 2008). One subclass is composed of toll interleukin 1 receptor (TIR)-NBS-LRR encoding genes characterized by the TIR domain homologous to the Drosophila toll and mammalian interleukin-1 receptor (Hammond-Kosack and Jones 1997). The second subclass is composed of NBS–LRR encoding genes without the TIR motif, which often includes a coiled-coil (CC) domain (Pan et al. 2000). Although these two subclasses are present in gymnosperm and dicot genomes, TIR–NBS–LRR are completely absent from monocot genomes (Zhou et al. 2004; Meyers et al. 2005).
Annotation of the Arabidopsis thaliana, rice, poplar (Populus trichocarpa), Medicago truncatula, grape, and papaya genomes identified 149, 480, 317, 333, 233, and 55 genes encoding NBS–LRR proteins, respectively (Bai et al. 2002; Meyers et al. 2003; Zhou et al. 2004; Tuskan et al. 2006; Velasco et al. 2007; Ameline-Torregrosa et al. 2008; Kohler et al. 2008; Ming et al. 2008). Furthermore, the genes encoding NBS–LRR proteins have been shown to be the most polymorphic of all Arabidopsis genes (Clark et al. 2007). This richness of NBS–LRR sequences represents an important resource for combating pathogen attack, since the function of these proteins is to serve as surveillance molecules that detect infection by specific pathogens (Chisholm et al. 2006). NBS–LRR sequences are often tightly linked at complex loci (Hulbert et al. 2001; Leister 2004; McDowell and Simon 2006). For example, in the Arabidopsis genome, 73.2% (109 of 149) of the NBS–LRR sequences are located in clusters of genes and 40 sequences are singletons (Meyers et al. 2003). In eukaryotes, genetic systems for generating new functions are often based on complex loci comprising tandemly organized genes of related function (Borst and Greaves 1987). Complex resistance gene clusters identified in plants fit this situation. Indeed, the clustered organization is supposed to favor sequence exchanges, such as unequal crossing over and/or gene conversion events, which can give rise, in some cases, to new (nonparental) R specificities (Sudupak et al. 1993; Richter et al. 1995; Chin et al. 2001). The relative importance of unequal crossing over and gene conversion compared to point mutations in the evolution of R gene clusters has been discussed (Michelmore and Meyers 1998). If these two processes were prevalent, homogenization of the sequences within a haplotype would be observed. However, comparisons of intra- and interspecies resistance haplotypes, such as at the rice Pi2/9 cluster (Zhou et al. 2007), revealed that orthologs are often more similar than paralogs, suggesting a low rate of sequence homogenization from unequal crossing over and gene conversion (Michelmore and Meyers 1998). Consequently, evolution of R gene clusters results from a balance between mechanisms that, in addition to creating new genes and reassorting them, can also homogenize them and mechanisms leading to sequence diversification (Meyers et al. 2005). Strong positive selection on specific residues (x) of the LRR domain has been proposed as a mechanism leading to sequence diversification. Indeed, in the LRR domain, the “xxLxLxx” motif is predicted to form a β-strand/β-turn structure in which the x residues are solvent exposed and available to detect potential pathogen ligands (Bent and Mackey 2007). Recently, illegitimate recombination within the LRR domain was also proposed to be a frequent source of variability in the evolution of the R gene cluster (Wickler et al. 2007). If clustering of resistance genes is a widely discussed explanation for the evolution of specific resistance genes (Hulbert et al. 2001), studies on the evolution of R gene clusters are scarce, compared with studies on R genes organized as “singletons” (Stahl et al. 1999; Tian et al. 2002; Mauricio et al. 2003). One such example is the analysis of the complex RGC2 cluster of lettuce where two types of R genes (I and II) were identified. Type I genes present a high frequency of sequence exchange, whereas type II genes show few sequences exchanges and “orthologs” can be identified (Kuang et al. 2004).
Common bean, Phaseolus vulgaris, is a plant species for which diversity of wild and cultivated forms has been extremely well documented (Kami et al. 1995; Broughton et al. 2003; Chacon et al. 2005). On the basis of morphological traits, phaseolin electrophoretic types, isozymes, and molecular markers, the P. vulgaris natural populations can be divided into three geographical regions of diversity: the Mesoamerican, the South Andean, and the North Andean centers. Genetic analyses based on neutral markers provide evidence for two independent domestication processes within the South Andean and Mesoamerican centers of diversity, leading to the development of two distinct groups of cultivated beans, called the Andean and Mesoamerican gene pools (Kami et al. 1995; Broughton et al. 2003; Chacon et al. 2005), respectively. These two gene pools are supposed to have diverged from a common ancestor ≤0.6 MYA (Chacon et al. 2007). In P. vulgaris, a complex R gene cluster, referred to as the B4 R gene cluster, is localized at the end of linkage group (LG) B4. This cluster comprises at least three R specificities of either Andean or Mesoamerican origin (Co-9, Co-y, and Co-z) and two R QTL effective against the fungal pathogen Colletotrichum lindemuthianum, the causal agent of anthracnose (Geffroy et al. 1999, 2000). Four expressed CC–NBS–LRR (CNL) encoding genes, mapped at the B4 R gene cluster, were previously characterized from two bean genotypes chosen to represent the Andean (JaloEEP558) and Mesoamerican (BAT93) cultivated gene pools (Ferrier Cana et al. 2003).
To further characterize the organization of the B4 R gene cluster and to understand its evolution since the divergence of Andean and Mesoamerican gene pools, selected genomic sequences of the B4 R gene cluster were determined and compared in two common bean genotypes, BAT93 and JaloEEP558. In addition to the four previously sequenced CNL, 25 B4-CNL encoding genes were determined—10 from JaloEEP558 (Andean) and 15 from BAT93 (Mesoamerican). No features associating the R-like genes to either gene pool have been identified. A combination of fluorescent in situ hybridization (FISH) and genetic mapping based on pulsed-field gel electrophoresis (PFGE) separation confirmed that the B4-CNL family is a large CNL family specific to one end of LG B4, corresponding to chromosome 4. B4-CNL members were mapped at two distinct blocks separated by 26 cM. FISH on meiotic pachytene chromosomes revealed that the B4 R gene cluster is localized in a peculiar genomic environment, since B4-CNL sequences are located in the subtelomeric region of the short arm of chromosome 4 adjacent to a major heterochromatic block and on both sides of a minor heterochromatic block (knobs).
MATERIALS AND METHODS
Plant material and genetic mapping:
Seventy-seven F9 recombinant inbred lines (RILs) derived from a cross between the Mesoamerican BAT93 genotype and the Andean JaloEEP558 genotype were used to map the B4-CNL sequences (phage inserts) on the integrated linkage map of common bean (Freyre et al. 1998). A PCR approach, using specific oligonucleotide primers, was utilized to map phage inserts. The list of primers used to map each phage insert is presented in supplemental Table 1. PCR reactions were completed in a volume of 25 μl containing 50 ng of template DNA, 1× PCR reaction buffer, 3 pmol of each primer, 50 μm of each dNTP, and 0.5 units of Red Taq Goldstar polymerase (Eurogentec, Seraing, Belgium). Amplifications were performed in a GeneAmp PCR system 2720 (PerkinElmer, Norwalk, CT). Presence/absence or size polymorphisms were scored on 77 F9 RILs. The MAPMAKER software, version 3.0, was used to map the genomic clone on the integrated linkage map (Lander et al. 1987). Linkage groups were established with a LOD threshold of 3.0 and a recombination fraction of 0.3. Marker order was estimated with a LOD threshold of 2.0, based on multipoint compare, order, and ripple analyses.
Pulsed-field gel electrophoresis and hybridization:
Conditions for high-molecular-weight DNA preparation in agarose blocks and restriction digestion with the rare cutting restriction endonuclease SalI have been described in Creusot et al. (1992) except that a different extraction buffer (10 mm Tris, pH 7.5; 15 mm KCl; 15 mm NaCl; 0.15 mm spermine; 0.5 mm spermidine; 2 mm EDTA; 0.5% Triton X-100 and 4.5% glucose) was used in this study for the DNA preparation. PFGE was carried out using a CHEF-DRIII apparatus commercialized by BIO-RAD. One-percent PFGE-grade agarose in 0.5× TBE was used as the matrix. Resolution of DNA fragments in the size range 50–600 kb was obtained by electrophoresis for 22 hr at 6 V/cm with a 12.5/40-sec pulse ramp at 10° in 0.5× TBE. The molecular weight standards were λ concatemers designed for use as size markers for PFGE (New England Biolabs). DNA transfer, hybridization conditions, and probe preparation were done as described in Geffroy et al. (1998). PRLJ1 is a NBS probe encompassing the P-loop–hydrophobic domain (HD) region of the NBS–LRR family specific to the B4 R gene cluster (Geffroy et al. 1999).
Library screening and sequence determination:
BAT93 and JaloEEP558 genomic λ libraries were screened with the PRLJ1 probe (Geffroy et al. 1999) as described in Ferrier Cana et al. (2003, 2005). Phage DNA was isolated using Nucleobond AX columns (Macherey-Nagel). Restriction analysis, Southern blot hybridization with the PRLJ1 probe, partial sequencing of the phage insert ends, and genetic mapping allowed the identification of contigs of overlapping phages. Three of these phage contigs, referred to as ContB1 (72,929 bp; nine phages), ContJ1 (29,543 bp; six phages), and ContJ2 (33,523 bp; four phages), and one phage, “λB54” (16,690 bp, which for convenience will be referred to as “contig” ContB2 in this article), were completely sequenced, except the 3′-end of ContJ1 that consists of tandem repeats preventing proper sequencing. For the remaining phages, sequencing was focused on B4-CNL sequences, using a primer walking strategy directly on phage DNA and/or on phage subcloned DNA, when two CNL were present on the same phage insert. Sequencing was performed using an automated ABI PRISM 3100 sequencer and the ABI prism BigDye Terminator Cycle sequencing kit (Applied Biosystems, Roissy, France) with custom primers or standard Sp6, T7, M13 reverse, or M13 (-21) primers. Raw DNA sequence data were visually inspected and assembled using the computer program Sequencher (Gene Codes, Ann Arbor, MI). The genomic sequence was annotated by using the gene prediction program Fgenesh (Softberry website) and was manually edited by a homology search against available databases. The genomic sequence of CNL-BA8 was used as a reference for the annotation of the NBS–LRR encoding genes (Ferrier Cana et al. 2003). The nucleotide sequence instead of the protein sequence was used for comparison because some genes, referred to as “pseudogenes,” do not contain the entire coding sequence. These pseudogenes contain a start methionine but present frameshift(s) or premature stop codon(s) leading to truncated predicted proteins compared to “full-length” encoding NBS–LRR (for details of their annotation, see supplemental Figure 2). The GenBank accession numbers are EU856768–EU856792.
Sequence analysis:
Multiple sequence alignments were performed using the Clustal X program and edited in GENEDOC (http://www.psc.edu/biomed/genedoc). Nucleotide sequence identities were established using the Gap program of the Genetics Computer Group (Madison, WI). COILs analyses (Lupas et al. 1991) were performed with the Macstripe, version 2.0b1, software, using a window of 14 and the MTK matrix. The average of distances between all pairs of CNL (the observed number of mutations with correction for probable invisible substitutions using the transversional model for the NBS domains and the general time reversible model for all other comparisons) was calculated with the PAUP software.
FISH mapping:
Phage genomic clones λB35, λB10, λB62, and λB61, mapped, respectively, at genetic position (GP) 1, GP 2, GP 3, and GP 4, were used as probes. Probes were labeled by nick translation (Roche Diagnostics) with Cy3-dUTP (Amersham Pharmacia Biotech) or Spectrum Green-dUTP (Vysis). Mitotic chromosome preparation is described in Pedrosa-Harand et al. (2006). Meiotic chromosomes were prepared from young flower buds fixed in ethanol/acetic acid 3:1 (v/v). Buds were macerated in 0.4% cellulase/0.4% pectolyase/0.4% citohelicase in 0.01 m citric acid–sodium citrate buffer, pH 4.8, for 3 hr at 37°, incubated in 60% acetic acid for 30 min, and squashed after removal of petals and sepals. Slide selection and pretreatment are described in Pedrosa et al. (2001). Chromosome and probe denaturation and post-hybridization washes were performed according to Heslop-Harrison et al. (1991), with modifications described in Pedrosa-Harand et al. (2006), except that meiotic preparations were denatured at 73° for 3 min. Reprobing of slides was performed according to Heslop-Harrison et al. (1992). Photographs were taken on a Zeiss Axioplan (Carl Zeiss) equipped with a mono cool view CCD camera (Photometrics, Tucson, AZ). Images from the camera were combined and pseudocolored using the IPLab spectrum software (IPLab, Fairfax, VA). Distance of FISH signals to the closest telomere and total chromosome length were measured in 10 metaphases using the “analyze–measure length” function of the same software. Digital images were imported into Adobe Photoshop version 8 for final processing.
Positive selection assessment:
The program CODEML from the PAML package (Yang 1997) was used to calculate the ω-ratio (of nonsynonymous-to-synonymous changes; dN/dS for each site). B4-CNL members corresponding to pseudogenes were excluded from this analysis. Consequently, this analysis was conducted on the 18 nucleotide sequences corresponding to “full-length” B4-CNL (12 from BAT93 and 6 from JaloEEP558), whose multiple alignment was carefully manually optimized in GENEDOC (http://www.nrbsc.org/gfsc/genedoc/index.html). Evolutionary models M7 and M8 were tested: model M7 assumed that amino acid site substitutions are conservative (ω ~ 0) and model M8 allows the occurrence of positively selected sites (ω > 1). M7 and M8 assume a β-distribution for the ω-value between 0 and 1. Diversifying selection was confirmed using a likelihood-ratio test by comparing the likelihood of models M8 and M7.
Phylogeny:
Phylogenetic trees were built for complete CNL sequences and separated CNL domains (A, B, C, D, and E) defined in Figure 3Figure 3. using three methods provided by PAUP 4.0b10 software: distance (BIONJ), maximum likelihood (ML), and maximum parsimony (MP). For all studied sequence sets, models of nucleotide substitution were evaluated with MODELTEST 3.0 (Posada and Crandall 2001) to identify the model that best fit the data. Distance- and ML-based phylogenetic trees were constructed with different nucleotide substitution models: the Tamura and Nei model (Tamura and Nei 1993) for domain E, the transversional model (proposed only by MODELTEST) for domains A and B, and the general time reversible model (Rodriguez et al. 1990) for domains C and D and for complete NBS–LRR sequences. The heterogeneity of nucleotide substitution rates among sites was approximated by a gamma distribution (G) and an assumption of invariable sites (I). Distance analyses were performed with the BIONJ algorithm (Gascuel 1997). ML analyses were carried out with a heuristic search strategy to find the best trees. The MP trees were built with the full heuristic search and the tree bisection reconnection branch-swapping option. A strict consensus tree was drawn when multiple best trees were obtained. Statistical confidence levels for MP, ML, and BIONJ trees were evaluated by the nonparametric bootstrap method on the basis of 100 resamplings. Bootstrap for ML analysis was performed without branch swapping to reduce computational time. Nodes represented in >70% of bootstrap replicates were considered well supported.
Figure 3.
Figure 3.
Figure 3.—
Amino acid sequence for the predicted full-length transcript of CNL-B7. The amino acid sequence is shown in single-letter code and is divided into six regions. (A) The N terminus. (B) The NBS. (C) The “spacer” region. (D) The LRR region. (more ...)
To test whether congruence between tree topologies of domains exists, four different methods were employed: (1) the Kishino–Hasegawa (KH) test (Kishino and Hasegawa 1989), (2) the Shimodaira–Hasegawa (SH) test (Shimodaira 2002), (3) the Swofford–Olsen–Waddell–Hillis (SOWH) test (Goldman et al. 2000), and (4) the expected likelihood weight (ELW) test (Strimmer and Rambaut 2002). The KH and SH tests were performed with PAUP. The SOWH and ELW tests were performed using PhyML (Guindon and Gascuel 2003). The best trees agreeing with the null hypotheses and the unconstrained ML tree, inferred in PhyML, employed the general time reversible (GTR), I, G nucleotide substitution model.
RESULTS
Distribution of PRLJ1-like sequences in the common bean genome and estimation of the size of the B4 R gene cluster through PFGE genetic mapping:
The bean NBS–PRLJ1 probe, encompassing the P-loop–HD region of plant R genes, was previously shown to be specific to the B4 R gene cluster, since all the polymorphic bands mapped at three different genetic positions (GP 1, GP 2, GP 3) defining a 2.7-cM interval at the end of linkage group B4 (Geffroy et al. 1999) (Figure 1Figure 1. ). In an attempt to estimate the physical size of the genomic region carrying the PRLJ1 family, we have used the PFGE technique. Hybridization experiments using PRLJ1 as a probe on a Southern blot of PFGE revealed a very complex hybridization pattern in both BAT93 and JaloEEP558 in the size range of 50–500 kb (Figure 1AFigure 1. ). To determine the genomic location of all these P(FGE)–PRLJ1 bands, high-molecular-weight DNA was extracted for each individual of the BAT93 × JaloEEP558 RILs progeny, digested with SalI and separated by PFGE. All the polymorphic bands, referred to as P-PRLJ1.a–k, mapped at the end of linkage group B4 at two blocks separated by 24.6 cM (Figure 1BFigure 1. ). First, P-PRLJ1.b, -d, -j, -g, and -k (from BAT93) and P-PRLJ1c, -h, and -i (from JaloEEP558) mapped in a 2.7-cM interval spanning GP 1, GP 2, and GP 3 (Figure 1BFigure 1. ). GP 1 and GP 2 are separated by 2.0 cM while GP 2 and GP 3 are separated by 0.7 cM. Second, P-PRLJ1.a (from JaloEEP558) and P-PRLJ.f (from BAT93) mapped 26.4 cM apart from the first block at GP 4 (Figure 1BFigure 1. ). Consequently, B4-CNL members present at the end of linkage group B4 are grouped at two distinct blocks separated by 26.4 cM: block 1 (spanning GP 1, GP 2, and GP 3) and block 2 (corresponding to GP 4) (Figure 1BFigure 1. ). From these experiments, by summing the size of the bands mapped at GP 1, GP 2, and GP 3, the size of block 1 was estimated to be at least 1000 kb in BAT93 and 700 kb in JaloEEP558. This is a minimal estimation because the intervening regions between GP 1, GP 2, and GP 3 might contain SalI fragments without PRLJ1 members. The intensity of the P-PRLJ1 bands is greatly variable, suggesting that several SalI fragments are extremely rich in PRLJ1 family members. For example, P-PRLJ1.g and -k from BAT93 (mapped at GP 2) and P-PRLJ1.h from JaloEEP558 (mapped at GP 1) present a very strong hybridization signal (Figure 1AFigure 1. ).
Figure 1.
Figure 1.
Figure 1.—
(A) High-stringency hybridization of an NBS–PRLJ1 probe to SalI-digested high-molecular-weight DNA separated by PFGE. Molecular weights in kilobases are indicated to the right. The region corresponding to the PRLJ1 probe is boxed in the amino (more ...)
General features of 29 B4-CNL members issued from two representative genotypes of the two major gene pools of common bean: BAT93 (Mesoamerican) and JaloEEP558 (Andean):
To estimate the number and the nucleotide polymorphism of the NBS–LRR homologs present at the B4 R gene cluster, genomic λ-phage libraries of both BAT93 and JaloEEP558 genotypes were screened with NBS–PRLJ1 probe, shown to be strictly specific to the B4 R gene cluster (this article and Geffroy et al. 1999). Complete or partial sequencing of the positive phage inserts allowed the identification of a total of 29 NBS–LRR homologs, 17 from BAT93 and 12 from JaloEEP558 (Figure 2Figure 2. ; supplemental Figure 1). A COILS analysis (Lupas et al. 1991) revealed that most of these NBS–LRR homologs present a significant CC structure (supplemental Figure 2) and were consequently referred to as CNL. Given that the B4 R gene cluster in both BAT93 and JaloEEP558 contains multiple members, the CNL encoding genes were named in numeric order according to their genomic and genetic position and were discriminated by the suffixes“B” (for BAT93) or “J” (for JaloEEP558). Among the 29 B4-CNL, 4 CNL correspond to previously identified sequences: CNL-BA8, CNL-BA11, CNL-JA78, and CNL-JA71 (Ferrier Cana et al. 2003). Coexistence of a full-length open reading frame (ORF) and pseudogenes (see materials and methods for their definition) were observed for both genotypes: in BAT93, of the 17 CNL-B, 5 correspond to pseudogenes; in JaloEEP558, of the 12 CNL-J, 6 correspond to pseudogenes. These 29 B4-CNL (Figure 2Figure 2. ; supplemental Figure 1), were mapped at the B4 R gene cluster at the four previously defined genetic positions (GP 1, GP 2, GP 3, and GP 4) (Figure 1BFigure 1. ). For each GP, CNL from both BAT93 and JaloEEP558 were identified, indicating that, as already revealed by the PFGE results, the occurrence of CNL sequences at these four GP predates the divergence between Andean and Mesoamerican gene pools (Figure 1BFigure 1. ).
Figure 2.
Figure 2.
Figure 2.—
Comparative analysis of sequenced phage contigs from BAT93 (ContB1 and ContB2) and JaloEEP558 (ContJ1 and ContJ2) mapped at the B4 R gene cluster. (A) ContB1 (72,919 bp) sequenced from BAT93 and its putative corresponding orthologous region from JaloEEP558 (more ...)
Sequence analysis revealed that the 29 B4-CNL present a very high percentage of nucleotide identity, ranging from 80 to 95% (supplemental Table 2). The 18 full-length B4-CNL potentially encode proteins ranging from 1066 aa (CNL-JA71) to 1186 aa (CNL-J1) (supplemental Figure 2). For these 18 sequences, no introns were identified between the start codon and the terminal stop codon, as predicted by previous results obtained for the B4-CNL encoding genes (Ferrier Cana et al. 2003) and in agreement with what has been previously found for plant CNL-encoding genes in general (Meyers et al. 2003). As shown in Figure 3Figure 3. , for the longest B4-CNL, CNL-B7, all the potential proteins can be divided into five regions (A–E) defined in Ferrier Cana et al. (2003). Region A, the N-terminal region, lacks a TIR domain and shares significant similarities only with the members of the non-TIR–NBS–LRR class of R proteins. A significant CC structure was identified for most B4-CNL as described above (supplemental Figure 2). Furthermore, an EDVID motif, shown to be more consistently present in CNL proteins than a predicted CC (Rairdan et al. 2008), was also identified. Region B includes the NBS region with typical P-loop (Walker A), kinase 2 (Walker B), and kinase 3a motifs, as well as other motifs such as GLPL and MHDL motifs found in all plant NBS–LRR proteins (Takken et al. 2006). This region also carries extra features specific to the non-TIR–NBS–LRR: a tryptophan residue (W) at the end of the kinase 2 motif, the RNBS-A-non-TIR motif (consensus: FDLxAWVCVSQxF), and the RNBS-D-non-TIR motif (consensus: CFLYCALFPED) (Meyers et al. 1999; Takken et al. 2006). Region C, referred to as the “spacer region,” has no obvious similarity with other R proteins except with previously published bean CNL (Creusot et al. 1999; Ferrier Cana et al. 2003), and soybean RPG1b CNL (Ashfield et al. 2004). A single hit was also identified in both A. thaliana (At3g14460) and M. truncatula (one CNL on BAC clone AC153125) genomes (Meyers et al. 2003; Ameline-Torregrosa et al. 2008). Region D can be organized in imperfect LRRs of variable length (15–41 amino acids) on the basis of the amino acid consensus pattern of the β-strand/β-turn motif (xxLxLxx) (Jones and Jones 1997). The deduced proteins harbor between 19 (CNL-JA71) and 22 LRR repeats (CNL-B7 and CNL-J1) (supplemental Figure 2). Compared to the 20 LRRs defined in Ferrier Cana et al. (2003) with CNL-BA8 as a reference, we have added LRR10′, which was omitted on CNL-BA8 in Ferrier Cana et al. (2003). Compared to these 20 LRRs, we have also identified 2 additional LRRs, referred to as LRR3′ (present in CNL-B7 and CNL-B23) and LRR12′ (present in CNL-J1) but absent in CNL-BA8 (supplemental Figure 2). Finally, region E corresponds to the C-terminal region and includes the conserved motif KIAHI, identified in our B4-CNL sequences. For the 11 B4-CNL pseudogenes, complete ORFs could be restored after a few base-pair changes, showing that they are highly similar in regions A, B, C, and E and present from 3 (CNL-J8*) to 22 LRRs (CNL-J11*) (supplemental Figure 2B). In addition to the previously identified Co-2 bean CNL (Creusot et al. 1999), the most closely related R proteins to our B4-CNL members are the soybean RPG1b CNL members exhibiting ~70% of amino acid similarity with B4-CNL (Ashfield et al. 2004; Hayes et al. 2004).
Identification of highly conserved intergenic regions:
Concerning the completely sequenced phage contigs, annotation of ContB1 (72,919 bp), ContJ1 (29,543 bp), ContB2 (16,690 bp), and ContJ2 (33,523 bp) revealed that all the putative ORFs present strong homology to CNL, but two ORFs, one at each 3′-end of ContB2 and ContJ2, show homology to a A. thaliana P-type transporting ATPase-like protein (At3g27870; E-value = 0; 86% of similarity) (Figure 2BFigure 2. ). Mapping experiments revealed that both ContB1 and ContJ1 are located at GP 3, while ContB2 is located at GP 2. Mapping of ContJ2 revealed that its 3′-end mapped at GP 2 while its 5′-end mapped at GP 3 because of the occurrence of a recombination event in the RIL progeny in the 22-kb region between these two PCR-based markers (Figure 2BFigure 2. ). These contigs are extremely rich in CNL sequences: ContB1, ContJ1, and ContJ2 present 8, 3, and 3 CNL sequences, in 72,919, 29,543, and 33,523 bp, respectively, giving an average CNL density of one CNL every 4 kb. On each contig, all the CNL (full length and pseudogenes) were present in the same transcriptional orientation, suggesting that they might have resulted from local tandem duplication (Figure 2Figure 2. ).
Nucleotide pairwise comparisons revealed that, although the intergenic regions display a degree of structural diversity, blocks of closely related sequences occur within and between the phage contigs. Three CNL belonging to ContB1 and ContJ1 appear to be more closely related than average. These are CNL-B1*, CNL-B6*, and CNL-J1, which share ≥95% of nucleotide identity (Figure 2AFigure 2. ; supplemental Table 2). Further sequence analysis revealed that this high level of sequence identity also extends into noncoding regions surrounding CNL-B1*, CNL-B6*, and CNL-J1 (Figure 2AFigure 2. ). Indeed, an ~7-kb fragment, including CNL-B1, shares ≥95% nucleotide identity with the region surrounding CNL-B6, suggesting that these two segments in ContB1 result from a recent DNA duplication (Figure 2AFigure 2. ). This 7-kb fragment is also conserved in the region surrounding CNL-J1 (nucleotide identity ≥90%). Between CNL-B1 and CNL-J1, the size of the conserved region is even bigger and extends ~14 kb (nucleotide identity ≥90%) (Figure 2Figure 2. ). Between ContB2 and ContJ2, two CNL, CNL-B18 and CNL-JA71, also appear to be more closely related than average since they present 95% of nucleotide identity (supplemental Table 2). A conserved 6-kb fragment including an intergenic region surrounding CNL-B18 and CNL-JA71 was also identified (nucleotide identity ≥90%) (Figure 2BFigure 2. ). ContB2 and ContJ2 also harbor in their 3′-end the ATPase-encoding sequences, sharing 97% of nucleotide identity between ContB2 and ContJ2.
The high sequence conservation, including intergenic regions between ContB1 and ContJ1, and the fact that these regions map at the same genetic position (GP 3) strongly suggest that these contigs are issued from orthologous regions from BAT93 and JaloEEP558. Similar conclusions can be made for ContB2 and ContJ2.
Evolutionary rates of CNL sequence divergence from BAT93 and JaloEEP558:
The rates of substitution within and between genotypes BAT93 and JaloEEP558 were studied to characterize the evolutionary force operating on the CNL gene family (Table 1). When complete CNL sequences from BAT93 or JaloEEP558 are inspected, similar rates of substitution are observed within and between CNL-B and CNL-J sequences. The same conclusion was obtained when the NBS domain or the LRR domain were analyzed separately. The rates of substitution are also nearly the same at a given GP (GP 1, GP 2, GP 3, or GP 4) for complete CNL sequence comparisons within and between genotypes. The fact that similar rates of substitution are observed within and between genotypes whatever the analyzed level (complete CNL, NBS, or LRR domain, four different GP) suggests that the CNL sequences did not homogenize within a haplotype because frequent occurrence of recombination events (unequal crossing over or gene conversion) is expected to give smaller values of substitution rates within each haplotype than between haplotypes. The LRR domain presents a rate of substitution nearly twice the one observed for the NBS domain. Finally, the rate of substitution at GP 1 is nearly half the one observed for the other GP (GP 2, GP 3, GP 4).
TABLE 1
TABLE 1
Evolutionary distances within and between Andean (JaloEEP558) and Mesoamerican (BAT93) B4-CNL sequences
CNL sequences from BAT93 (CNL-B) and JaloEEP558 (CNL-J) located at a same genetic position are often more similar than paralogs:
Phylogenetic trees were built with the 29 complete nucleotidic CNL sequences as well as on separate domains (NBS, LRR) (Figure 4Figure 4. ). Well-supported clades were reproducible on phylogenetic trees obtained with the three different methods (data not shown) and only ML trees are presented (Figure 4Figure 4. ).
Figure 4.
Figure 4.
Figure 4.—
Phylogenetic trees built using the ML method showing the relationships between complete CNL sequences (A), region B (NBS domain) (B), and region D (LRR domain) (C). Regions A, B, C, D, and E of the CNL sequences are defined in Figure 3Figure 3. . The soybean NBS–LRR (more ...)
CNL sequences from BAT93 and JaloEEP55 do not group into distinct Mesoamerican and Andean clades. On the contrary, the evolutionary history of CNL sequences from BAT93 and JaloEEP558 is interrelated (Figure 4Figure 4. ). CNL sequences from BAT93 and JaloEEP558 located at the same genetic position are often more closely related than the paralogs from the same genotype. For example, five well-supported clades corresponding to these criteria are found on phylogenetic trees based on complete CNL, NBS, and LRR domains (Figure 4Figure 4. ). They group as CNL-B19, CNL-BA8, CNL-BA11, and CNL-JA78 (GP 1); CNL-JA71 with CNL-B18 (GP 2); CNL-B6*, CNL-J1, and CNL-B1* (GP 3); CNL-B11 with CNL-J4* (GP 3); and CNL-B23 with CNL-J11* (GP 4). This is also illustrated in supplemental Table 2, showing, for example, that both CNL-B11 and CNL-J4* mapped at GP 3 present 99% of nucleotide identity. On the phylogenetic trees, paralogs that are closely linked physically are sometimes distantly related. This is illustrated by CNL-B23 and CNL-B22, located <20 kb apart on the same phage insert and mapped at GP 4 (supplemental Figure 1). CNL-B23 (GP 4) groups with the Andean CNL-J11* (GP 4), while its paralog, CNL-B22 (GP 4), is distantly related on the phylogenetic trees (Figure 3Figure 3. ). Altogether, these results suggest that significant sequence homogenization did not occur within either genotype.
Likelihood-based tests of topology were performed because inspecting only robust nodes does not prove congruence of the topology. Significant incongruence was detected between the NBS and LRR trees (Figure 4, B and CFigure 4. ). Consequently, NBS and LRR domains seem to have different evolutionary histories, suggesting the occurrence of recombination events. These tree topology tests were extended to all possible combinations of domains (A, B, C, D, and E). In all cases, significant incongruence was detected. Combined with the results from the previous paragraph, our data suggest that recombination events have occurred, but not at sufficient frequency to lead to sequence homogenization within a haplotype.
Positively selected residues were identified mainly on “x” residues predicted to be solvent exposed in the LRR domain:
A strong preference for nonsynonymous compared to synonymous codon substitutions has been detected in many plant R gene families (Ellis et al. 2000; Jones and Dangl 2006; Bent and Mackey 2007). To examine whether B4-CNL members are subject to diversifying selection, the program PAML was used on the 18 full-length B4-CNL (12 from BAT93 and 6 from JaloEEP558) (Yang 1997). Sites under diversifying selection were investigated using the M8 and M7 codon substitution models. The likelihood-ratio test for comparing M8 with M7 is 2Δ ln L = 938, which is much greater than the χ2 critical value (9.21 at the 1% significance level, with d.f. 2). This shows that our data were significantly more likely under the M8 model with positive selection on some codons than under the M7 model, which does not allow the presence of sites with ω > 1. A total of 72 sites under diversifying selection were identified with a posterior probability > 99% (Figure 3Figure 3. ). An important fraction of the positive sites (53 of 72; 74%) was located in the LRR region (region D) (Figure 3Figure 3. ). Two positive sites were also identified in region A (before the NBS domain), 6 in region B, 8 in region C, and 3 in region E. Of the 53 positive sites located in the LRR region, 37 (70%) were confined on the “x” residues of the “xxLxLxx” motif (x, any amino acid, L, leucine or isoleucine) predicted to be solvent exposed.
The B4 R gene cluster is localized in a subtelomeric region of the short arm of chromosome 4 and is adjacent to heterochromatic regions:
To localize the B4 R gene cluster on chromosomes of the BAT93 bean genotype, four phages (λB35, λB10, λB62, and λB61 mapped at GP 1, GP 2, GP 3, and GP 4, respectively; Figure 1BFigure 1. ) were used as probes in FISH experiments. λB35, λB61, and λB62 showed clear and reproducible signals in the terminal region of the short arm of a chromosome pair (Figure 5, A and BFigure 5. ). Double hybridization of λB61 and λB62 (Figure 5BFigure 5. ) and of λB61 and λB35 (data not shown) on mitotic chromosomes revealed that λB35, λB61, and λB62 localized in the same chromosome region. This chromosome region appeared more intensely stained compared to other regions in some FISH experiments, suggesting that it is a region of constitutive heterochromatin (A. Pedrosa-Harand, unpublished data). λB35, λB61, and λB62 hybridization signals placed all those clones adjacent to the terminal heterochromatic block in the short arm, but their position relative to each other could not be unambiguously determined (Figure 5CFigure 5. ). This chromosome could be identified as chromosome 4 through hybridization with specific markers, in agreement with the mapping of those phages on LG B4 (A. Pedrosa-Harand, unpublished data). Chromosome 4 was previously named 10 (Pedrosa et al. 2003), but all chromosomes have been recently renamed according to the linkage group names of the core linkage map of common bean (Freyre et al. 1998; Pedrosa-Harand et al. 2008).
Figure 5.
Figure 5.
Figure 5.—
Localization of phages λB10, λB35, λB61, and λB62 on BAT93 chromosomes. (A–E) λB35 (red, A), λB61 (red, B and D), λB62 (green, B) and λB10 (red, E) on mitotic metaphase chromosomes (more ...)
The phage λB10 labeled multiple chromosome ends, a pattern that indicated the presence of one or more repetitive sequences in its insert. This or these sequences seem to be particularly abundant around the B4 R gene cluster, where it generated strong hybridization signals apparently colocalizing with λB61 (Figure 5, D and EFigure 5. ) and the terminal heterochromatic block. Since λB10 was genetically mapped between λB35 and λB62 (see Figure 5CFigure 5. ), it is likely that the repetitive sequence that it contains is interspersed with CNL sequences in the B4 R gene cluster.
To confirm the position of λB35 and λB61 relative to each other and to the heterochromatic block, these two phages were also localized on meiotic pachytene chromosomes. The signals generated from both probes greatly overlapped (Figure 5, G and HFigure 5. ), suggesting that these clones are similar enough to cross-hybridize and detect the whole R gene cluster under the stringent conditions used in these experiments (72%). This is in agreement with sequence analysis showing that B4-CNL sequences exhibit at least 80% nucleotide identity (supplemental Table 2). The localization of both clones on pachytene chromosomes demonstrated that the B4 R gene cluster has a subtelomeric localization and confirmed that B4-CNL are adjacent to a large, terminal heterochromatic block (knob) (arrow in Figure 5FFigure 5. ). Furthermore, B4-CNL members are present on both sides of, and are included in, a smaller heterochromatic region (arrowhead in Figure 5FFigure 5. ), detected only at this level of resolution.
DISCUSSION
In this report, two genotypes representative of the two common bean gene pools, BAT93 (Mesoamerican) and JaloEEP558 (Andean), were selected to investigate the genetic events involved in the evolution of the B4 R gene cluster since the divergence of these two gene pools. We gained insight into the organization and the evolution of the B4 R gene cluster by using a combination of approaches, including sequence analysis of 29 CNL from both bean genotypes, FISH experiments, and genetic mapping based on PFGE separation.
Genetic mapping based on hybridization with NBS–PRLJ1 after PFGE separation confirmed that the B4-CNL members are specific to the end of bean LG B4. It has been demonstrated that they are divided into two distinct blocks: block 1 spanning 2.7 cM at the end of LG B4 (corresponding to GP 1, GP 2, GP 3) and separated by 26.4 cM from the second block (block 2). In agreement with their genetic location on the bean genetic map, FISH experiments proved that the B4-CNL sequences are present only at the end of the short arm of chromosome 4. In the common bean genome, one other CNL cluster, the Co-2 cluster, has been identified at the end of LG B11 (Geffroy et al. 1998; Creusot et al. 1999). Co-2-CNL sequences, presenting ~65% of nucleotide identity with PRLJ1, have not been picked up under the hybridization conditions used in this article. FISH mapping of B4-CNL sequences to meiotic pachytene chromosomes provides additional information: the B4-CNL sequences are located only in the subtelomeric region of the short arm of chromosome 4 on both sides of a minor heterochromatic block (knob). Furthermore, a second knob, referred to as “major,” has also been identified in a more distal position. The organization “major knob/B4-CNL sequences/minor knob/B4-CNL sequences” suggests that chromosome inversion might have split an ancestral knob into two knobs. The major knob is unambiguously placed in a distal position, while the minor knob is tentatively placed between GP 3 and GP 4, since these two genetic positions are separated by 26.4 cM and since λB35 (mapped at GP 1) and λB61 (mapped at GP 4) labeled on both sides of the minor knob (Figure 5CFigure 5. ). FISH mapping on pachytene chromosomes with probes specific to each GP would be necessary to confirm this hypothesis.
Our analysis indicates that the B4 R gene cluster is a large cluster in terms of its physical size as well as in terms of the number of CNL that it encodes. Indeed, the physical size of block 1 was estimated to be at least 1000 kb in BAT93 and 700 kb in JaloEEP558. Furthermore, a minimal number of 17 and 12 B4-CNL are present in BAT93 and JaloEEP558, respectively. Thus, the B4 cluster is relatively large compared to other plant R gene clusters. For example, the tomato I2 locus encodes 7 members in 90 kb (Simons et al. 1998), the maize rp3 locus encodes 5 members in 140 kb (Webb et al. 2002), and the lettuce Dm3 locus encodes 32 members within at least 3000 kb (Kuang et al. 2004). Large variation in gene copy number among different genotypes has been observed, such as for the maize Rp1 cluster, in which a range of 1–52 copies has been estimated (Smith et al. 2004). In our analysis, no such dramatic differences in B4-CNL copy number were observed between BAT93 (17 CNL) and JaloEEP558 (12 CNL). However, hybridization experiments with PRLJ1 suggest a higher level of complexity in BAT93 than in JaloEEP558, since BAT93 exhibits a higher number of strongly hybridizing bands than JaloEEP558 (Figure 1AFigure 1. ). Sequencing of phage contigs (ContB1, ContJ1, ContB2, ContJ2) suggests that the B4-CNL are not uniformly distributed but rather clustered in a CNL-rich region, with an average density of one CNL every 4 kb (Figure 2Figure 2. ). This conclusion is also supported by PFGE data revealing that the intensity of the P-PRLJ1 bands is highly variable, suggesting that several SalI fragments are extremely rich in CNL members compared to others. The B4 R gene cluster is composed of homogeneous CNL sequences since all the 29 B4-CNL sequences show >80% nucleotide identity, suggesting that these CNL sequences derive from local tandem duplications. This, however, is not a universal phenomenon since several “heterogeneous” R clusters, presenting highly divergent sequences, have been reported (Hulbert et al. 2001; Leister 2004). The most striking example is the Mla powdery mildew resistance locus of barley harboring three different families of NBS–LRR-encoding genes with a sequence identity of <33% (Wei et al. 1999).
The genomic organization of the B4-CNL sequences into two blocks mirrors what is observed for resistance specificities since two distinct groups of resistance specificities and resistance QTL have been described: one group (containing Co-9, Co-y, Co-z) is located in a distal position (in block 1 in Figure 1BFigure 1. ) while another group (containing Co-RVI) is tentatively located in a more proximal region on linkage group B4. Indeed, on the basis of comparative mapping studies, the region containing P-PRLJ1.a and -f (block 2 on Figure 1BFigure 1. ) is known to harbor the Co-RVI anthracnose resistance gene (Adam-Blondon et al. 1994; Freyre et al. 1998). Altogether, the end of the short arm of chromosome 4 carries an important collection of specific R genes and R QTL effective against diverse pathogens [fungi C. lindemuthianum and Uromyces appendiculatus, bacteria Pseudomonas syringae pv phaseolicola, and bean golden yellow mosaic virus (BGYMV)] (Lopez et al. 2003; Miklas et al. 2006; Rodriguez-Suarez et al. 2008). However, their location relative to each other is not precisely known because they are often present in different genotypes. On the basis of our current and previous studies, the GP 1 region appears to be particularly interesting. First, it is the most distal genetic position. Second, it is the only genetic position where functional specificities from BAT93 (Co-9) and JaloEEP558 (Co-y) have been mapped. And third, the rate of substitution at GP 1 is nearly half that observed for the other GP (GP 2, GP 3, and GP 4). Several CNL are candidate genes for the Co-y and Co-9 specificities: two full-length CNL from JaloEEP558 (CNL-J9 and CNL-JA78) colocalized with Co-y (present in JaloEEP558) and three full-length CNL from BAT93 (CNL-B19, CNL-BA8, and CNL-BA11) colocalized with Co-9 (present in BAT93).
One important feature of heterochromatic knobs, described in maize by McClintock et al. (1981), is that they are highly unstable regions. This instability is exemplified by results from A. thaliana where the heterochromatic knob located on chromosome 4 close to the centromere is present in ecotypes Col and Ws but absent in Ler (Fransz et al. 2000). The molecular mechanism underlying this instability is not known. In A. thaliana, suppression of recombination in the region containing the knob is observed in a cross between Col and Ler (Drouaud et al. 2006). This suppression of recombination probably reflects the absence of the knob in Ler rather than an intrinsic feature of heterochromatic blocks. In this study, suppression of recombination is not observed in a cross between BAT93 and JaloEEP558, since genetic distance between GP 1 and GP 4, containing the minor knob, was estimated to be 29.1 cM. On the contrary, increased levels of recombination are suspected. Indeed, it is only recently that GP 1, GP 2, and GP 3 have been linked to one end of LG B4. In a previous version of the bean linkage map, they were considered as a separate small linkage group referred to as LG B14 (Freyre et al. 1998; Geffroy et al. 1999). The difficulty in linking LG B14 to LG B4 could be explained either by an increased level of recombination in the region containing the minor knob or by the existence of a low level of polymorphism between BAT93 and JaloEEP558 in that region. An alternative hypothesis would be that the absence of the minor knob in JaloEEP558 results in a polymorphism between BAT93 and JaloEEP558 and is responsible for the difficulty in linking LG B14 to LG B4. Similarly, in Lotus japonicus, cytogenetic analyses revealed that an inversion in the region between LG 1A and LG 1B was unexpectedly responsible for the difficulty of linking these two linkage groups because this polymorphism would have led, in principle, to suppression of recombination (Pedrosa et al. 2002). Cytogenetic analyses of pachytene chromosomes from JaloEEP558 are currently underway to study the existence of the minor knob in JaloEEP558.
Our results provide two leads to explain the origin of the important proliferation of CNL sequences at the B4-R gene cluster: (i) its proximity to heterochromatin blocks (knobs) and (ii) its subtelomeric localization. Concerning the proximity to heterochromatin, the tomato Mi-1 R gene cluster has also been located near heterochromatin, more precisely in the junction of euchromatin and pericentromeric heterochromatin on chromosome 6 (Zhong et al. 1999; Seah et al. 2007). This region in tomato is highly enriched in R genes effective against diverse pathogens as well as in resistance gene analog sequences with unidentified function (Seah et al. 2007). In our study, the B4-CNL sequences also seem to be located at a junction of euchromatin and heterochromatin. Consequently, it is tempting to speculate that some features of the chromosome structure in these euchromatin/heterochromatin junction regions may provide a favorable environment for R gene proliferation. We hypothesize that it could be due to some form of gene silencing. Indeed, genes inserted in proximity to heterochromatin are often silenced as in the case of position-effect variegation in Drosophila, and chromatin structure is known to play a major role in epigenetic regulation of the genome (Schotta et al. 2003; Lippman and Martienssen 2004). The importance of RNA silencing in maintaining low levels of expression at plant disease resistance gene clusters was recently identified by Yi and Richards (2007) for the Arabidopsis RPP5 locus. They further hypothesize that the disruption of RNA silencing by pathogens, as occurs in infections by viruses and Agrobacterium (Dunoyer et al. 2006; Ding and Voinnet 2007), may play an important role in optimizing plant response to pathogen attack by increasing the expression of R genes.
On the other hand, the subtelomeric location of the B4 R gene cluster is another feature that might favor R gene proliferation. Indeed, subtelomeres of such diverse organisms as human, yeast, and trypanosomes are dynamic and variable mosaics of multichromosomal blocks of sequence, resulting from the fact that subtelomeres are hotspots of interchromosomal recombination (Mefford and Trask 2002; Linardopoulou et al. 2005). It has been proposed that the processes acting on subtelomeric regions may have a role in diversifying gene families (Mefford and Trask 2002). Furthermore, in several organisms such as Saccharomyces cerevisiae, genes in close proximity to telomeres undergo transcriptional silencing, a phenomenon known as “telomere position effect” (Mefford and Trask 2002). This putative silencing effect could have a beneficial effect on R gene proliferation in a similar way to what we proposed above for heterochromatic silencing. In the common bean genome, most of the well-characterized large R gene clusters are located at the end of linkage groups, suggesting that the location at the end of a linkage group, and by inference a subtelomeric location, is favorable for R gene proliferation. For example, the Co-x, I, and Co-2 loci have been mapped to the ends of LG B1, LG B2, and LG B11, respectively (Geffroy et al. 1998, 2008; Creusot et al. 1999; Vallejos et al. 2006). Furthermore, the Co-4 anthracnose resistance locus was also confirmed to be in a subtelomeric position of bean chromosome 8 (Melotto et al. 2004). In other plant species, large R gene clusters, such as the maize Rp1 cluster and barley Mla cluster, are also located at the end of linkage groups (Pryor and Ellis 1993; Wei et al. 1999). However, systematic localization of large R gene clusters in subtelomeric regions is not always observed, for example, in the A. thaliana genome (Meyers et al. 2003). In addition, subtelomeric regions in A. thaliana do not share extensive similarity among most nonhomologous chromosomes, such as seen in yeast and humans (Kuo et al. 2006). This suggests that additional data are needed to understand the behavior of subtelomeric regions in plants (Fan et al. 2008) and that A. thaliana may not be representative of all plant species. In conclusion, in common bean, the localization of the B4 R gene cluster in regions with high genome plasticity (proximity to knobs and/or subtelomeric localization) is likely to increase genetic and epigenetic variations, which may in turn result in the accelerated evolution of R gene specificities.
In plants and animals, several multigene families, such as rRNA genes, are subject to concerted evolution, in which family members share greater sequence identity within a species than between species. This homogenization of sequence within a given species is probably the result of repeated occurrences of unequal crossing over and/or gene conversion (Nei and Rooney 2005). Sequencing of 29 B4-CNL from two bean genotypes representative of the two gene pools of P. vulgaris gave us the opportunity to compare paralogs and orthologs. No sequence homogenization of the B4-CNL within a haplotype was observed. Indeed, similar substitution rates were observed for CNL sequences within and between haplotypes. Furthermore, phylogenetic analyses revealed that Andean and Mesoamerican B4-CNL do not form two separate clades, but that both genotypes possess an assortment of different CNL. Indeed, sequences from BAT93 and JaloEEP558 located at the same GP are often, but not always, more similar than paralogs. This pattern of evolution suggests that the B4-CNL multigene family is not subject to concerted evolution but rather follows the “birth-and-death” model of evolution (Nei et al. 1997). This model was first proposed to explain the evolution of MHC genes in mammals (Nei et al. 1997; Nei and Rooney 2005) and has been subsequently adapted for R gene cluster evolution in plants (Michelmore and Meyers 1998). As expected under the birth and death model (Nei and Rooney 2005), the B4 R gene cluster contains a large number of CNL corresponding to pseudogenes (in JaloEEP558, 6 pseudogenes of 12 B4-CNL; in BAT93, 5 pseudogenes of 17 B4-CNL). However, even if recombination events were not frequent enough to lead to sequence homogenization within one haplotype, two types of evidence suggest that recombination occurs occasionally in the B4 R gene cluster. First, significant incongruence was identified between trees built on different domains (A, B, C, D, and E) of the B4-CNL sequences. Second, a duplicated 7-kb region of highly conserved sequences (>95% nucleotide identity), including the noncoding region, was identified in BAT93 (green region on ContB1 in Figure 2AFigure 2. ). Blocks of homology, including noncoding sequence, have also been observed in other R gene clusters, as, for example, in the tomato Mi-1 gene cluster (Seah et al. 2007) or the Rph7 locus in barley (Scherrer et al. 2005). This indicates the occurrence of sequence exchanges, even if the molecular mechanism underlying these sequence exchanges cannot be deduced from these analyses. As already observed for other R gene clusters (Zhou et al. 2007), the B4-CNL polymorphism appears to be primarily generated by variation in LRR copy number (from 19 to 22 LRRs), nucleotide substitution, and diversifying selection, especially on the x residues predicted to solvent expose in the LRR domain (Figure 3Figure 3. ). Variation at these x residues appears to be evolutionarily advantageous, suggesting that they may be involved in determining pathogen recognition specificity. Together, these results suggest that the host–pathogen coevolution process identified at the level of the centers of diversity of P. vulgaris (Geffroy et al. 1999) could be governed mainly by minor molecular changes on these x residues. In other words, on the basis of a complete sequence of a given B4-CNL, it is impossible to predict if it has an Andean or a Mesoamerican origin, while at the phenotypic level a strong differentiation for resistance has been identified with respect to the centers of diversity (Geffroy et al. 1999).
In conclusion, the subtelomeric location of the B4-CNL sequences combined with B4-CNL's proximity to knobs might confer to this cluster an unusual potential for adaptation to new strains of an ever-changing array of pathogens, as testified by the numerous CNL, specific R genes, and resistance QTL mapped at the B4 R gene cluster. Phage sequencing revealed that B4-CNLs are tightly clustered at CNL-rich regions with an average density of one CNL sequence every 4 kb. PFGE analysis revealed that the B4-CNL sequences are present on at least 1000 kb in BAT93 for block 1. This large physical size prompted us to use a more suitable genomic library to study the B4 R gene cluster. We are currently using a BAC library from genotype BAT93, presenting an average insert size of 125 kb (Kami et al. 2006) to establish the complete sequence of the B4 R gene cluster to understand the factors involved in its large expansion.
Acknowledgments
We thank Peter Moffett and Nicolas Chen for helpful discussions and critical reading of the manuscript. We thank Anne-Valérie Gendrel for stimulating discussions. We also thank the two anonymous reviewers for their valuable advice. We thank Dieter Schweizer (Gregor Mendel Institute of Molecular Plant Biology) and Austrian Science Fund for supporting the participation of A.P.-H. in this project. The research was supported by the Institut National de la Recherche Agronomique, the Centre National de la Recherche Scientifique, and the Ministère de la Recherche.
References
  • Adam-Blondon, A. F., M. Sévignac, H. Bannerot and M. Dron, 1994. SCAR, RAPD and RFLP markers linked to a dominant gene (Are) conferring resistance to anthracnose in common bean. Theor. Appl. Genet. 88 865–870.
  • Ameline-Torregrosa, C., B. B. Wang, M. S. O'Bleness, S. Deshpande, H. Y. Zhu et al., 2008. Identification and characterization of nucleotide-binding site-leucine-rich repeat genes in the model plant Medicago truncatula. Plant Physiol. 146 5–21. [PubMed]
  • Ashfield, T., L. E. Ong, K. Nobuta, C. M. Schneider and R. W. Innes, 2004. Convergent evolution of disease resistance gene specificity in two flowering plant families. Plant Cell 16 309–318. [PubMed]
  • Bai, J. F., L. A. Pennill, J. C. Ning, S. W. Lee, J. Ramalingam et al., 2002. Diversity in nucleotide binding site-leucine-rich repeat genes in cereals. Genome Res. 12 1871–1884. [PubMed]
  • Belkhadir, Y., R. Subramaniam and J. L. Dangl, 2004. Plant disease resistance protein signaling: NBS-LRR proteins and their partners. Curr. Opin. Plant Biol. 7 391–399. [PubMed]
  • Bent, A. F., and D. Mackey, 2007. Elicitors, effectors, and R genes: the new paradigm and a lifetime supply of questions. Annu. Rev. Phytopathol. 45 399–436. [PubMed]
  • Borst, P., and D. R. Greaves, 1987. Programmed gene rearrangements altering expression. Science 235 658–667. [PubMed]
  • Broughton, W. J., G. Hernandez, M. Blair, S. Beebe, P. Gepts et al., 2003. Beans (Phaseolus spp.): model food legumes. Plant Soil 252 55–128.
  • Chacon, M. I., B. Pickersgill and D. G. Debouck, 2005. Domestication patterns in common bean (Phaseolus vulgaris L.) and the origin of the Mesoamerican and Andean cultivated races. Theor. Appl. Genet. 110 432–444. [PubMed]
  • Chacon, M. I., B. Pickersgill, D. G. Debouck and J. S. Arias, 2007. Phylogeographic analysis of the chloroplast DNA variation in wild common bean (Phaseolus vulgaris L.) in the Americas. Plant Syst. Evol. 266 175–195.
  • Chin, D. B., R. Arroyo Garcia, O. E. Ochoa, R. V. Kesseli, D. O. Lavelle et al., 2001. Recombination and spontaneous mutation at the major cluster of resistance genes in lettuce (Lactuca sativa). Genetics 157 831–849. [PubMed]
  • Chisholm, S. T., G. Coaker, B. Day and B. J. Staskawicz, 2006. Host-microbe interactions: shaping the evolution of the plant immune response. Cell 124 803–814. [PubMed]
  • Clark, R. M., G. Schweikert, C. Toomajian, S. Ossowski, G. Zeller et al., 2007. Common sequence polymorphisms shaping genetic diversity in Arabidopsis thaliana. Science 317 338–342. [PubMed]
  • Creusot, F., M. Sévignac and M. Dron, 1992. Large-scale mapping using high-molecular-weight DNA from P. vulgaris. Plant Mol. Biol. Rep. 10 207–218.
  • Creusot, F., C. Macadré, E. Ferrier Cana, C. Riou, V. Geffroy et al., 1999. Cloning and molecular characterization of three members of the NBS-LRR subfamily located in the vicinity of the Co-2 locus for anthracnose resistance in Phaseolus vulgaris. Genome 42 254–264. [PubMed]
  • Dangl, J. L., and J. D. G. Jones, 2001. Plant pathogens and integrated defence responses to infection. Nature 411 826–833. [PubMed]
  • Ding, S. W., and O. Voinnet, 2007. Antiviral immunity directed by small RNAs. Cell 130 413–426. [PubMed]
  • Drouaud, J., C. Camilleri, P. Y. Bourguignon, A. Canaguier, A. Berard et al., 2006. Variation in crossing-over rates across chromosome 4 of Arabidopsis thaliana reveals the presence of meiotic recombination “hot spots.” Genome Res. 16 106–114. [PubMed]
  • Dunoyer, P., C. Himber and O. Voinnet, 2006. Induction, suppression and requirement of RNA silencing pathways in virulent Agrobacterium tumefaciens infections. Nat. Genet. 38 258–263. [PubMed]
  • Ellis, J., P. Dodds and T. Pryor, 2000. The generation of plant disease resistance gene specificities. Trends Plant Sci. 5 373–379. [PubMed]
  • Fan, C., Y. Zhang, Y. Yu, S. Rounsley, M. Long et al., 2008. The subtelomere of Oryza sativa chromosome 3 short arm as a hot bed of new gene origination in rice. Mol. Plant 1 839–850.
  • Ferrier Cana, E., V. Geffroy, C. Macadre, F. Creusot, P. Imbert Bollore et al., 2003. Characterization of expressed NBS-LRR resistance gene candidates from common bean. Theor. Appl. Genet. 106 251–261. [PubMed]
  • Ferrier Cana, E., C. Macadre, M. Sevignac, P. David, T. Langin et al., 2005. Distinct post-transcriptional modifications result in seven alternative transcripts of the CC-NBS-LRR gene JA1tr of Phaseolus vulgaris. Theor. Appl. Genet. 110 895–905. [PubMed]
  • Flor, H. H., 1955. Host-parasite interaction in flax rust—its genetics and other implications. Phytopathology 45 680–685.
  • Fransz, P. F., S. Armstrong, J. H. de Jong, L. D. Parnell, G. van Drunen et al., 2000. Integrated cytogenetic map of chromosome arm 4S of A. thaliana: structural organization of heterochromatic knob and centromere region. Cell 100 367–376. [PubMed]
  • Freyre, R., P. 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.
  • Gascuel, O., 1997. BIONJ: an improved version of the NJ algorithm based on a simple model of sequence data. Mol. Biol. Evol. 14 685–695. [PubMed]
  • Geffroy, V., F. Creusot, J. Falquet, M. Sevignac, A. F. Adam-Blondon et al., 1998. A family of LRR sequences in the vicinity of the Co-2 locus for anthracnose resistance in Phaseolus vulgaris and its potential use in marker-assisted selection. Theor. Appl. Genet. 96 494–502.
  • Geffroy, V., D. Sicard, J. C. F. de Oliveira, M. Sevignac, S. Cohen et al., 1999. Identification of an ancestral resistance gene cluster involved in the coevolution process between Phaseolus vulgaris and its fungal pathogen Colletotrichum lindemuthianum. Mol. Plant Microbe Interact. 12 774–784. [PubMed]
  • Geffroy, V., M. Sevignac, J. C. F. De Oliveira, G. Fouilloux, P. Skroch et al., 2000. Inheritance of partial resistance against Colletotrichum lindemuthianum in Phaseolus vulgaris and co-localization of quantitative trait loci with genes involved in specific resistance. Mol. Plant Microbe Interact. 13 287–296. [PubMed]
  • Geffroy, V., M. Sevignac, P. Billant, M. Dron and T. Langin, 2008. Resistance to Colletotrichum lindemuthianum in Phaseolus vulgaris: a case study for mapping two independent genes. Theor. Appl. Genet. 116 407–415. [PubMed]
  • Goldman, N., J. P. Anderson and A. G. Rodrigo, 2000. Likelihood-based tests of topologies in phylogenetics. Syst. Biol. 49 652–670. [PubMed]
  • Guindon, S., and O. Gascuel, 2003. A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Syst. Biol. 52 696–704. [PubMed]
  • Hammond-Kosack, K. E., and J. D. G. Jones, 1997. Plant disease resistance genes. Annu. Rev. Plant Physiol. Plant Mol. Biol. 48 575–607. [PubMed]
  • Hammond-Kosack, K. E., and J. E. Parker, 2003. Deciphering plant-pathogen communication: fresh perspectives for molecular resistance breeding. Curr. Opin. Biotechnol. 14 177–193. [PubMed]
  • Hayes, A. J., S. C. Jeong, M. A. Gore, Y. G. Yu, G. R. Buss et al., 2004. Recombination within a nucleotide-binding-site/leucine-rich-repeat gene cluster produces new variants conditioning resistance to soybean mosaic virus in soybeans. Genetics 166 493–503. [PubMed]
  • Heslop-Harrison, J. S., T. Schwarzacher, K. Anamthawat-Jónsson, A. R. Leitch, M. Shi et al., 1991. In situ hybridization with automated chromosome denaturation. Technique 3 109–115.
  • Heslop-Harrison, J. S., G. E. Harrison and I. J. Leitch, 1992. Reprobing of DNA: DNA in situ hybridization preparations. Trends Genet. 8 372–373. [PubMed]
  • Hulbert, S. H., C. A. Webb, S. M. Smith and Q. Sun, 2001. Resistance gene complexes: evolution and utilization. Annu. Rev. Phytopathol. 39 285–312. [PubMed]
  • Jones, D. A., and J. D. G. Jones, 1997. The role of leucine-rich repeat proteins in plant defences, pp. 89–167 in Advances in Botanical Research Incorporating Advances in Plant Pathology, Vol 24. Academic Press, London.
  • Jones, J. D. G., and J. L. Dangl, 2006. The plant immune system. Nature 444 323–329. [PubMed]
  • Kami, J., V. B. Velasquez, D. G. Debouck and P. Gepts, 1995. Identification of presumed ancestral DNA sequences of phaseolin in Phaseolus vulgaris. Proc. Natl. Acad. Sci. USA 92 1101–1104. [PubMed]
  • Kami, J., V. Poncet, V. Geffroy and P. Gepts, 2006. Development of four phylogenetically-arrayed BAC libraries and sequence of the APA locus in Phaseolus vulgaris. Theor. Appl. Genet. 112 987–998. [PubMed]
  • Kishino, H., and M. Hasegawa, 1989. Evaluation of the maximum likelihood estimate of the evolutionary tree topologies from DNA sequence data, and the branching order in Hominoidea. J. Mol. Evol. 29 170–179. [PubMed]
  • Kohler, A., C. Rinaldi, S. Duplessis, M. Baucher, D. Geelen et al., 2008. Genome-wide identification of NBS resistance genes in Populus trichocarpa. Plant Mol. Biol. 66 619–636. [PubMed]
  • Kuang, H., S. S. Woo, B. C. Meyers, E. Nevo and R. W. Michelmore, 2004. Multiple genetic processes result in heterogeneous rates of evolution within the major cluster disease resistance genes in lettuce. Plant Cell 16 2870–2894. [PubMed]
  • Kuo, H. F., K. M. Olsen and E. J. Richards, 2006. Natural variation in a subtelomeric region of Arabidopsis: implications for the genomic dynamics of a chromosome end. Genetics 173 401–417. [PubMed]
  • Lander, E. S., P. Green, J. Abrahamson, A. Barlow, M. J. Daly et al., 1987. MAPMAKER: an interactive computer package for constructing primary genetic linkage maps of experimental and natural populations. Genomics 1 174–181. [PubMed]
  • Leister, D., 2004. Tandem and segmental gene duplication and recombination in the evolution of plant disease resistance genes. Trends Genet. 20 116–122. [PubMed]
  • Linardopoulou, E. V., E. M. Williams, Y. X. Fan, C. Friedman, J. M. Young et al., 2005. Human subtelomeres are hot spots of interchromosomal recombination and segmental duplication. Nature 437 94–100. [PubMed]
  • Lippman, Z., and R. Martienssen, 2004. The role of RNA interference in heterochromatic silencing. Nature 431 364–370. [PubMed]
  • Lopez, C. E., I. F. Acosta, C. Jara, F. Pedraza, E. Gaitan-Solis et al., 2003. Identifying resistance gene analogs associated with resistances to different pathogens in common bean. Phytopathology 93 88–95. [PubMed]
  • Lupas, A., M. Vandyke and J. Stock, 1991. Predicting coiled coils from protein sequences. Science 252 1162–1164.
  • Madden, L. V., and M. Wheelis, 2003. The threat of plant pathogens as weapons against US crops. Annu. Rev. Phytopathol. 41 155–176. [PubMed]
  • Martin, G. B., A. J. Bogdanove and G. Sessa, 2003. Understanding the functions of plant disease resistance proteins. Annu. Rev. Plant Biol. 54 23–61. [PubMed]
  • Mauricio, R., E. A. Stahl, T. Korves, D. C. Tian, M. Kreitman et al., 2003. Natural selection for polymorphism in the disease resistance gene Rps2 of Arabidopsis thaliana. Genetics 163 735–746. [PubMed]
  • McClintock, B., T. A. Kato and A. Bluemenshein, 1981. Chromosome Constitution of Races in Maize. Colegio de Postgraduados, Chapingo, Mexico.
  • McDowell, J. M., and S. A. Simon, 2006. Recent insights into R gene evolution. Mol. Plant Pathol. 7 437–448.
  • McHale, L., X. P. Tan, P. Koehl and R. W. Michelmore, 2006. Plant NBS-LRR proteins: adaptable guards. Genome Biol. 7 212. [PubMed]
  • Mefford, H. C., and B. J. Trask, 2002. The complex structure and dynamic evolution of human subtelomeres. Nat. Rev. Genet. 3 91–102. [PubMed]
  • Melotto, M., M. F. Coelho, A. Pedrosa-Harand, J. D. Kelly and L. E. A. Camargo, 2004. The anthracnose resistance locus Co-4 of common bean is located on chromosome 3 and contains putative disease resistance-related genes. Theor. Appl. Genet. 109 690–699. [PubMed]
  • Meyers, B. C., A. W. Dickerman, R. W. Michelmore, S. Sivaramakrishnan, B. W. Sobral et al., 1999. Plant disease resistance genes encode members of an ancient and diverse protein family within the nucleotide-binding superfamily. Plant J. 20 317–332. [PubMed]
  • Meyers, B. C., A. Kozik, A. Griego, H. H. Kuang and R. W. Michelmore, 2003. Genome-wide analysis of NBS-LRR-encoding genes in Arabidopsis. Plant Cell 15 809–834. [PubMed]
  • Meyers, B. C., S. Kaushik and R. S. Nandety, 2005. Evolving disease resistance genes. Curr. Opin. Plant Biol. 8 129–134. [PubMed]
  • Michelmore, R. W., 2003. The impact zone: genomics and breeding for durable disease resistance. Curr. Opin. Plant Biol. 6 397–404. [PubMed]
  • Michelmore, R., and B. C. Meyers, 1998. Clusters of resistance genes in plants evolve by divergent selection and a birth and death process. Genome Res. 8 1113–1130. [PubMed]
  • Miklas, P. N., J. D. Kelly, S. E. Beebe and M. W. Blair, 2006. Common bean breeding for resistance against biotic and abiotic stresses: from classical to MAS breeding. Euphytica 147 105–131.
  • Ming, R., S. B. Hou, Y. Feng, Q. Y. Yu, A. Dionne-Laporte et al., 2008. The draft genome of the transgenic tropical fruit tree papaya (Carica papaya Linnaeus). Nature 452 991–997. [PubMed]
  • Nei, M., and A. P. Rooney, 2005. Concerted and birth-and-death evolution of multigene families. Annu. Rev. Genet. 39 121–152. [PubMed]
  • Nei, M., X. Gu and T. Sitnikova, 1997. Evolution by the birth-and-death process in multigene families of the vertebrate immune system. Proc. Natl. Acad. Sci. USA 94 7799–7806. [PubMed]
  • Pan, Q. L., J. Wendel and R. Fluhr, 2000. Divergent evolution of plant NBS-LRR resistance gene homologues in dicot and cereal genomes. J. Mol. Evol. 50 203–213. [PubMed]
  • Pedrosa, A., M. F. Jantsch, E. A. Moscone, P. F. Ambros and D. Schweizer, 2001. Characterisation of pericentromeric and sticky intercalary heterochromatin in Ornithogalum longibracteatum (Hyacinthaceae). Chromosoma 110 203–213. [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. [PubMed]
  • Pedrosa, A., C. E. Vallejos, A. Bachmair and D. Schweizer, 2003. Integration of common bean (Phaseolus vulgaris L.) linkage and chromosomal maps. Theor. Appl. Genet. 106 205–212. [PubMed]
  • Pedrosa-Harand, A., C. C. S. de Almeida, M. Mosiolek, M. Blair, D. Schweizer et al., 2006. Extensive ribosomal DNA amplification during Andean common bean (Phaseolus vulgaris L.) evolution. Theor. Appl. Genet. 112 924–933. [PubMed]
  • Pedrosa-Harand, A., T. Porch and P. Gepts, 2008. Standard nomenclature for common bean chromosomes and linkage groups. Annu. Rep. Bean Improv. Coop. 51 106–107.
  • Posada, D., and K. A. Crandall, 2001. Selecting the best-fit model of nucleotide substitution. Syst. Biol. 50 580–601. [PubMed]
  • Pryor, T., and J. Ellis, 1993. The genetic complexity of fungal resistance genes in plants. Adv. Plant Pathol. 10 281–305.
  • Rairdan, G., and P. Moffett, 2007. Brothers in arms? Common and contrasting themes in pathogen perception by plant NB-LRR and animal NACHT-LRR proteins. Microbes Infect. 9 677–686. [PubMed]
  • Rairdan, G. J., S. M. Collier, M. A. Sacco, T. T. Baldwin, T. Boettrich et al., 2008. The coiled-coil and nucleotide binding domains of the potato Rx disease resistance protein function in pathogen recognition and signaling. Plant Cell 20 739–751. [PubMed]
  • Richter, T. E., T. J. Pryor, J. L. Bennetzen and S. H. Hulbert, 1995. New rust resistance specificities associated with recombination in the Rp1 complex in maize. Genetics 141 373–381. [PubMed]
  • Rodriguez, F., J. L. Oliver, A. Marin and J. R. Medina, 1990. The general stochastic model of nucleotide substitution. J. Theor. Biol. 142 485–501. [PubMed]
  • Rodriguez-Suarez, C., J. J. Ferreira, A. Campa, A. Paneda and R. Giraldez, 2008. Molecular mapping and intra-cluster recombination between anthracnose race-specific resistance genes in the common bean differential cultivars Mexico 222 and Widusa. Theor. Appl. Genet. 116 807–814. [PubMed]
  • Scherrer, B., E. Isidore, P. Klein, J. S. Kim, A. Bellec et al., 2005. Large intraspecific haplotype variability at the Rph7 locus results from rapid and recent divergence in the barley genome. Plant Cell 17 361–374. [PubMed]
  • Schotta, G., A. Ebert, R. Dorn and G. Reuter, 2003. Position-effect variegation and the genetic dissection of chromatin regulation in Drosophila. Semin. Cell Dev. Biol. 14 67–75. [PubMed]
  • Seah, S., A. C. Telleen and V. M. Williamson, 2007. Introgressed and endogenous Mi-1 gene clusters in tomato differ by complex rearrangements in flanking sequences and show sequence exchange and diversifying selection among homologues. Theor. Appl. Genet. 114 1289–1302. [PubMed]
  • Shimodaira, H., 2002. An approximately unbiased test of phylogenetic tree selection. Syst. Biol. 51 492–508. [PubMed]
  • Simons, G., J. Groenendijk, J. Wijbrandi, M. Reijans, J. Groenen et al., 1998. Dissection of the fusarium I2 gene cluster in tomato reveals six homologs and one active gene copy. Plant Cell 10 1055–1068. [PubMed]
  • Smith, S. M., A. J. Pryor and S. H. Hulbert, 2004. Allelic and haplotypic diversity at the Rp1 rust resistance locus of maize. Genetics 167 1939–1947. [PubMed]
  • Stahl, E. A., G. Dwyer, R. Mauricio, M. Kreitman and J. Bergelson, 1999. Dynamics of disease resistance polymorphism at the Rpm1 locus of Arabidopsis. Nature 400 667–671. [PubMed]
  • Strimmer, K., and A. Rambaut, 2002. Inferring confidence sets of possibly misspecified gene trees. Proc. R. Soc. Lond. Ser. B Biol. Sci. 269 137–142.
  • Sudupak, M. A., J. L. Bennetzen and S. H. Hulbert, 1993. Unequal exchange and meiotic instability of disease-resistance genes in the Rp1 region of maize. Genetics 133 119–125. [PubMed]
  • Takken, F. L. W., M. Albrecht and W. I. L. Tameling, 2006. Resistance proteins: molecular switches of plant defence. Curr. Opin. Plant Biol. 9 383–390. [PubMed]
  • Tamura, K., and M. Nei, 1993. Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Mol. Biol. Evol. 10 512–526. [PubMed]
  • Tian, D. C., H. Araki, E. Stahl, J. Bergelson and M. Kreitman, 2002. Signature of balancing selection in Arabidopsis. Proc. Natl. Acad. Sci. USA 99 11525–11530. [PubMed]
  • Tuskan, G. A., S. DiFazio, S. Jansson, J. Bohlmann, I. Grigoriev et al., 2006. The genome of black cottonwood, Populus trichocarpa (Torr. & Gray). Science 313 1596–1604. [PubMed]
  • Vallejos, C. E., G. Astua-Monge, V. Jones, T. R. Plyler, N. S. Sakiyama et al., 2006. Genetic and molecular characterization of the I locus of Phaseolus vulgaris. Genetics 172 1229–1242. [PubMed]
  • Velasco, R., A. Zharkikh, M. Troggio, D. A. Cartwright, A. Cestaro et al., 2007. A high quality draft consensus sequence of the genome of a heterozygous grapevine variety. PLoS ONE 2 e1326. [PubMed]
  • Webb, C. A., T. E. Richter, N. C. Collins, M. Nicolas, H. N. Trick et al., 2002. Genetic and molecular characterization of the maize rp3 rust resistance locus. Genetics 162 381–394. [PubMed]
  • Wei, F. S., K. Gobelman Werner, S. M. Morroll, J. Kurth, L. Mao et al., 1999. The Mla (powdery mildew) resistance cluster is associated with three NBS-LRR gene families and suppressed recombination within a 240-kb DNA interval on chromosome 5S (1HS) of barley. Genetics 153 1929–1948. [PubMed]
  • Wickler, T., N. Yahiaoui and B. Keller, 2007. Illegitimate recombination is a major evolutionary mechanism for initiating size variation in plant resistance genes. Plant J. 51 631–641. [PubMed]
  • Yang, Z. H., 1997. PAML: a program package for phylogenetic analysis by maximum likelihood. Comput. Appl. Biosci. 13 555–556. [PubMed]
  • Yi, H., and E. J. Richards, 2007. A cluster of disease resistance genes in Arabidopsis is coordinately regulated by transcriptional activation and RNA silencing. Plant Cell 19 2929–2939. [PubMed]
  • Zhong, X. B., J. Bodeau, P. F. Fransz, V. M. Williamson, A. van Kammen et al., 1999. FISH to meiotic pachytene chromosomes of tomato locates the root knot nematode resistance gene Mi-1 and the acid phosphatase gene Aps-1 near the junction of euchromatin and pericentromeric heterochromatin of chromosome arms 6S and 6L, respectively. Theor. Appl. Genet. 98 365–370.
  • Zhou, B., M. Dolan, H. Sakai and G. L. Wang, 2007. The genomic dynamics and evolutionary mechanism of the Pi2/9 locus in rice. Mol. Plant Microbe Interact. 20 63–71. [PubMed]
  • Zhou, T., Y. Wang, J. Q. Chen, H. Araki, Z. Jing et al., 2004. Genome-wide identification of NBS genes in japonica rice reveals significant expansion of divergent non-TIR NBS-LRR genes. Mol. Genet. Genomics 271 402–415. [PubMed]
Articles from Genetics are provided here courtesy of
Genetics Society of America

PubMed articles by these authors
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph