Genetic Variation at the Tomato Cf-4/Cf-9 Locus Induced by EMS Mutagenesis and Intralocus Recombination

Corresponding author: Jonathan D. G. Jones, John Innes Centre, Norwich, NR4 7UH, United Kingdom., jonathan.jones{at}sainsbury-laboratory.ac.uk (E-mail)

Communicating editor: D. CHARLESWORTH

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The interaction between tomato (Lycopersicon esculentum) and the leaf mold pathogen Cladosporium fulvum is an excellent model for investigating disease resistance gene evolution. The interaction is controlled in a gene-for-gene manner by Cf genes that encode type I transmembrane extracellular leucine-rich repeat glycoproteins that recognize their cognate fungal avirulence (Avr) proteins. Cf-4 from L. hirsutum and Cf-9 from L. pimpinellifolium are located at the same locus on the short arm of tomato chromosome 1 in an array of five paralogs. Molecular analysis has shown that one mechanism for generating sequence variation in Cf genes is intragenic sequence exchange through unequal crossing over or gene conversion. To investigate this we used a facile genetic selection to identify novel haplotypes in the progeny of Cf-4/Cf-9 trans-heterozygotes that lacked Cf-4 and Cf-9. This selection is based on the ability of Avr4 and Avr9 to induce Cf-4- or Cf-9-dependent seedling death. The crossovers were localized to the same intergenic region defining a recombination hotspot in this cross. As part of a structure-function analysis of Cf-9 and Cf-4, nine EMS-induced mutant alleles have been characterized. Most mutations result in single-amino-acid substitutions in their C terminus at residues that are conserved in other Cf proteins.

TOMATO Cf genes confer resistance to the leaf mold pathogen Cladosporium fulvum through recognition of distinct fungal-encoded avirulence (Avr) peptides secreted into the leaf apoplast during infection. The interaction between tomato and C. fulvum is an attractive model system to study the molecular basis of recognition specificity in plant disease resistance (R) proteins (VAN DER HOORN et al. 2001A ; WULFF et al. 2001 ), R-protein-mediated defense responses (RIVAS and THOMAS 2002 ), and R gene evolution (PARNISKE et al. 1997 ; VAN DER HOORN et al. 2001B ). Cf genes encode type I transmembrane glycoproteins composed predominantly of extracytoplasmic leucine-rich repeats (LRRs), a membrane-spanning region, and a short cytoplasmic domain that lacks an obvious signaling function (RIVAS and THOMAS 2002 ). Cf proteins, in common with many R proteins, activate a hypersensitive cell death response (HR) upon recognition of their cognate Avr proteins.

The tomato Cf-4 and Cf-9 genes were introgressed into the cultivated tomato Lycopersicon esculentum cv. "Moneymaker" (Cf0), from wild relatives. Their protein products activate a plant defense response upon recognition of the C. fulvum Avr4 and Avr9 proteins, respectively. Cf-4 (from L. hirsutum) and Cf-9 (from L. pimpinellifolium) each reside in an array of five paralogs at the Milky Way (MW) locus on the short arm of chromosome 1 (PARNISKE et al. 1997 ; THOMAS et al. 1997 ). Sequence analysis of the Cf4 and Cf9 haplotypes, together with the corresponding locus from the disease-sensitive Cf0 line (PARNISKE et al. 1997 ), revealed extensive polymorphism within Hcr9's (Homologues of C. fulvum resistance gene 9). It was concluded that extensive intergenic and intragenic sequence exchange has occurred between paralogs. Selection for nonsynonymous nucleotide substitutions in sequences encoding the putative solvent-exposed residues of a conserved LRR structural motif also contributes to Hcr9 sequence variation (PARNISKE et al. 1997 ). Other comparisons of R gene families have also revealed a higher rate of nonsynonymous to synonymous substitution within the putative solvent-exposed residues of the ß-strand/ß-turn structural motif of LRRs (MCDOWELL et al. 1998 ; MEYERS et al. 1998 ; ELLIS et al. 1999 ; NOEL et al. 1999 ). This observation is consistent with the proposed role for LRRs in conferring Avr recognition specificity (DANGL and JONES 2001 ).

As a result of studies on several R gene families in Arabidopsis, flax, maize, rice, lettuce, and tomato, two models for R gene evolution have been proposed. On the basis of molecular analysis of the tomato Pto and the lettuce Dm3 loci, MICHELMORE and MEYERS 1998 proposed the "birth and death" model, similar to the evolution of MHC alleles in mammals. In this model, R genes are "born" through duplication events, evolve in isolation, and "die" due to deletion or mutation. Sequence exchange is proposed to occur between orthologs, but is rare between paralogs, and point mutation acting in concert with divergent selection appears to be the primary process generating novelty. The analysis of other R gene loci resulted in the so-called "permutation model" for R gene evolution (DODDS et al. 2001A ), in which sequence exchange occurs between paralogs as well as orthologs. This model was deduced from analysis of the Arabidopsis RPP5, maize Rp1, flax N and P, and tomato Cf-4/Cf-9 loci (PARNISKE et al. 1997 ; NOEL et al. 1999 ; DODDS et al. 2001A , DODDS et al. 2001B ; SUN et al. 2001 ).

Comparative sequencing and the analysis of genetic stability at R gene loci have revealed key molecular mechanisms affecting R gene evolution such as inter- and intragenic recombination, gene conversion, mutation, and transposon insertion (RICHTER et al. 1995 ; PARNISKE et al. 1997 ; LUCK et al. 1998 , LUCK et al. 2000 ; COLLINS et al. 1999 ; NOEL et al. 1999 ; CHIN et al. 2001 ; SUN et al. 2001 ). The analysis of Cf-4/Cf-9 recombinants could provide useful insights into gene evolution at this locus. Recombinants lacking Cf-4 and Cf-9 were originally identified in a genetic screen after inoculation with C. fulvum race 5 (that expresses Avr4 and Avr9). However, a number of additional Hcr9's have since been shown to confer resistance through recognition of different Avr determinants (PARNISKE et al. 1997 ; LAUGE et al. 1998A ; TAKKEN et al. 1999 ; PANTER et al. 2002 ), and it is possible that some unequal crossovers would not have been recovered. In an attempt to overcome these limitations we developed a facile genetic selection to identify recombinants that lack Cf-4 and Cf-9. This selection is based on the ability of Avr4 and Avr9 to induce Cf-4- and Cf-9-dependent seedling death (HAMMOND-KOSACK et al. 1994A ; THOMAS et al. 1997 ). As part of a structure-function analysis of Cf proteins, we have also characterized eight loss-of-function ethyl methanesulfonate-induced mutant alleles of Cf-9 and one of Cf-4.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Details of tomato lines used:
Near-isogenic lines (NILs) of L. esculentum cv. Moneymaker containing Cf-4 (Cf4), Cf-9 (Cf9), or no genes for resistance to C. fulvum (Cf0) were maintained as described previously (DIXON et al. 1996 , DIXON et al. 2000 ).

Mutagenesis with EMS:
The mutagenesis procedure was similar to that described previously (HAMMOND-KOSACK et al. 1994B ). Approximately 3000 Cf9 seeds and 1000 Cf4 seeds were imbibed for 24 hr at room temperature in water. The seeds were incubated in a solution containing 60 mM EMS for 24 hr at 22°. After extensive washing the seeds were sown in batches of 200 in seed flats. Viable seedlings were transferred to individual pots and treated with a 10^–5 M solution of the gibberellic acid biosynthesis inhibitor paclobutrazol. This helped to maintain the large number of M₁ plants in a semi-dwarf state. The first fruit truss was removed since it is more likely to contain chimeric material. Selfed seed were collected from the second fruit truss of M₁ plants. The effectiveness of the mutagenic procedure was assessed by scoring several phenotypic characters in M₂ families as described by HAMMOND-KOSACK et al. (1994b). The frequency of visible mutations was similar to that reported previously (results not shown).

Mutant identification:
Mutants compromised in their ability to induce an Avr4- or Avr9-dependent hypersensitive response were identified by inoculating M₂ families with recombinant potato virus X (PVX) expressing the C. fulvum Avr4 (PVX:Avr4) or Avr9 (PVX:Avr9) genes fused to the plant Pr1A signal peptide sequence for targeting to the leaf apoplast. In contrast to a previous report (THOMAS et al. 2000 ) that involved inoculating recombinant virus particles passaged through Nicotiana clevelandii, recombinant PVX was cloned into the T-DNA vector pGREEN (HELLENS et al. 2000 ) under control of the cauliflower mosaic virus 35S promoter. Recombinant clones were electroporated into Agrobacterium tumefaciens strain GV3101 containing the helper plasmid pSOUP. Stationary phase cultures were resuspended at an OD₆₀₀ of 0.5 in Murashige and Skoog salts containing 2-[N-Morpholino]ethanesulfonic acid (MES) adjusted to pH 5.6 with KOH, 2% w/v sucrose, and 10 µM acetosyringone. After 3 hr at room temperature bacterial suspensions were infiltrated into the cotyledons of M₂ seedlings. In Cf4 and Cf9 control plants a systemic Avr-dependent hypersensitive response could be observed after 7 days.

Kanamycin resistance assays:
Resistance to kanamycin in tomato seedlings was determined by germination on Murashige and Skoog media supplemented with 300 µg ml^–1 kanamycin.

Generation of a line expressing Avr4 and Avr9:
A line expressing Avr4 and Avr9 transgenes was constructed by intercrossing a kanamycin-resistant transgenic line expressing Avr9 (SLJ6021 A; HAMMOND-KOSACK et al. 1994B ) with four independent kanamycin-resistant transgenic lines expressing Avr4 (Avr4 7291 N, J, L, and M; THOMAS et al. 1997 ). The F₁ plants were self-pollinated and the segregation for resistance to kanamycin was tested in F₂ progeny. The ratio of kanamycin-resistant to -sensitive progeny in the cross SLJ6021 A x Avr47291 J was close to the predicted 15:1 ratio for independently assorting genes, and these progeny were analyzed further. PCR analysis was used to identify kanamycin-resistant plants containing Avr4 and Avr9. Twenty-four individuals were then analyzed by DNA gel blots, using probes to the Avr4 and Avr9 transgenes and an endogenous tomato gene. Putative homozygotes for the Avr4 and Avr9 transgenes were identified on the basis of their relative hybridization intensity on blots. The genotypes of three individuals were confirmed in testcrosses to Cf4 and Cf9 plants. One line, designated Cf0 Avr4, Avr9, was used in testcrosses to identify individuals lacking Cf-4 and Cf-9 function.

Generation of molecular markers at the Milky Way locus:
A cleaved amplified polymorphic sequence (CAPS) marker named MW3, located 3' of the LoxR gene in the MW locus, was used to distinguish the Cf0, Cf4, and Cf9 haplotypes (Fig 1). Amplification with the primer combination LoxR-F1 and LoxR-R1 (Table S1, supplementary information at http://www.genetics.org/supplemental/) yields a 496-bp product from Cf0, a 408-bp product from Cf4, and a 496-bp product from Cf9. Digestion with HincII generates 262- and 234-bp fragments derived from the Cf0 allele and 262 and 146 bp from the Cf4 allele, while the Cf9 allele is not cut with this enzyme. The CAPS marker MW5 was used to distinguish the Cf4 and Cf9 haplotypes at the 5' end of the MW locus (Fig 1). Amplification with the primers 99A8, 99A10R, and 99A9R generates a 601-bp product from Cf4, which is not cleaved by RsaI, while a 611-bp product is generated from Cf9, which is cleaved by RsaI to produce fragments of 502 and 109 bp. This primer combination does not amplify a product from the Cf0 haplotype. The CAPS marker MW1 was used to delimit recombination events in tract II (Fig 1), which includes the Hcr9-4C and Hcr9-9E 5' flanking regions (Fig 4B). Amplification with 99DE1F and 99DE2R (Table S1) gave products of 1.512 and 1.467 kbp from Cf4 and Cf9, respectively, but only the Cf9 allele can be digested by AvaI to generate fragments of 502 and 965 bp.

View larger version (27K): In this window In a new window Download PPT slide   Figure 1. Schematic of Hcr9's in the Cf0, Cf4, and Cf9 haplotypes to summarize sequence homologies in the intergenic regions. Tracts of near-identical intergenic sequences are depicted in the same colors. The protein-coding regions of Hcr9's are colored according to the haplotype from which they originated, i.e., black, Cf0 from L. esculentum; blue, Cf4 from L. hirsutum; and red, Cf9 from L. pimpinellifolium. The regions of maximum intergenic sequence homology are indicated by the gray tracts labeled I–V. Lipoxygenase exons are indicated by green arrows and boxes. The locations of two polymorphic regions for which the CAPS markers MW3 and MW5 were developed at the 3' and 5' end of the Milky Way locus are indicated. Functional resistance genes with known specificities are marked with an "R," or "adult R" in the case of Hcr9-9B and Hcr9-9E, which confer resistance in adult plants (PARNISKE et al. 1997 ; PANTER et al. 2002 ).

View larger version (49K): In this window In a new window Download PPT slide   Figure 2. The predicted amino acid sequence of Cf-9 is shown with structural domains indicated on the left (RIVAS and THOMAS 2002 ). The locations of six EMS-derived mutations that abolish Cf-9 function and a single mutation that abolishes Cf-4 function are highlighted in black. The predicted amino acid changes are shown on the right of the protein sequence. The L457F change in which the F from Cf-4 replaces the L from Cf-9 abolishes Cf-9 function in Agrobacterium-mediated transient assays in mature tobacco leaves and is highlighted in gray (VAN DER HOORN et al. 2001A ; WULFF et al. 2001 ). Sequences that form part of the putative ß-strand/ß-turn conserved structural motif in LRR proteins (xxLxLxx, where L is leucine and x is any amino acid) are shown delimited by the dashed box. Putative N-linked glycosylation sites (NxS/T) are shown underlined.

View larger version (18K): In this window In a new window Download PPT slide   Figure 3. Crossing strategies used in this study. (A) The meiotic stability of Cf-4 was measured by testcrossing to the Cf0 Avr4, Avr9 line. Most of the progeny display the seedling lethal phenotype as a result of Avr4-induced seedling death. Survivors (S) can be recovered when Cf-4 is mutated or lost due to unequal crossing over or gene conversion. (B) Crossing strategy to identify loss-of-function alleles or recombinants lacking Cf-4 or Cf-9 function in a Cf-4/Cf-9 trans-heterozygote. (C) Crossing strategy to recover functional alleles of EMS-derived Cf-9 mutants.

View larger version (62K): In this window In a new window Download PPT slide   Figure 4. (A) Analysis of Cf4 and Cf9 NILs and S2 survivors. DNA gel blots of F2 plants homozygous for the recombinant chromosome, plus DNA from three plants that define three recombinant classes, class I (V514), class IIa (V517), and class IIb (V408; PARNISKE et al. 1997 ; THOMAS et al. 1997 ), are shown. DNA was digested with BglII and hybridized with a 1.6-kbp probe from the 5' end of Cf-9. The sizes of BglII DNA fragments in kilobase pairs that originate from specific Hcr9's are indicated on the left and the right. (B) Physical organization of the Cf4 (light shading) and Cf9 (dark shading) haplotypes. BglII restriction sites are represented by vertical lines and the sizes of fragments revealed upon hybridization to the Cf-9 probe are indicated in base pairs. Hcr9 protein-coding regions are indicated by horizontal arrows. The crossovers in previously characterized recombinants (V408, V512, V514, V516, and V517; PARNISKE et al. 1997 ) and recombinants analyzed in this study are indicated by solid triangles. On the basis of the comparison between the restriction patterns of the previously characterized Cf-4/Cf-9 disease-sensitive recombinants (THOMAS et al. 1997 ) and the recombinants reported here, the crossovers in S1 survivors were localized to a 3.468-kbp region delimited by the CAPS marker MWI and a deletion of 353 bp in Hcr9-4C (class IIa) or a contiguous fragment of 2.289 kbp (class IIb). (C) Expanded view of crossovers in the Hcr9-4C/Hcr9-9E 5' flanking regions. Crossovers are indicated by darkly shaded bands. The lightly shaded background represents regions with >98.5% sequence identity. The 5' ends of the coding regions of Hcr9-9E (9E) and Hcr9-4C (4C) are indicated. ATn and ATTn microsatellite repeats are indicated by lollipops and fragments of DNA with homology to a dispersed repeat element (CR-1, 92% identity in 38 nucleotides) associated with the Cab-1 locus of tomato and auxin-responsive promoter element (aux, 90% identity in 40 nucleotides) from tobacco (BERNATZKY et al. 1988 ; TAKAHASHI et al. 1990 ) are indicated by open and solid boxes, respectively.

DNA gel blot analysis:
Tomato genomic DNA was prepared as previously described (THOMAS et al. 1997 ) and probed with a 5' or 3' fragment of the Cf-9 gene prepared by PCR amplification of plasmid pCDNAL9 (JONES et al. 1994 ), using the primer combinations F10/F5 or F6/Cf4-16, respectively (Fig 4B, Table S1).

DNA sequence analysis:
Mutant Cf-4 and Cf-9 alleles were amplified from genomic DNA using gene-specific primers. The primer combination F10/F5 (Table S1) amplifies a 1.377-kbp fragment from the 5' half of Cf-4 and a 1.568-kbp fragment from the 5' half of Cf-9. The primer combination F6/Cf4-16 (Table S1) amplifies a 1.500-kbp fragment from the 3' half of Cf-4 or Cf-9. The amplified gene fragments were directly sequenced on both strands. The location of mutations was verified by direct sequencing of a PCR product encompassing the modified nucleotide. Hcr9 promoter sequences were aligned and analyzed as previously described (PARNISKE et al. 1997 ; PARNISKE and JONES 1999 ). The Lasergene package (DNASTAR, Madison, WI) was used for DNA sequence assembly and primer selection.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Isolation and characterization of EMS-induced Cf-9 and Cf-4 mutants:
As part of a structure-function analysis of the Cf-9 and Cf-4 proteins, and to identify extragenic suppressors of Cf gene function (DIXON et al. 2000 ), we mutagenized the Cf9 and Cf4 NILs. Approximately 2000 batches of 25 M₂ seed collected from individual Cf9 M₁ plants were screened for suppressors of PVX:Avr9-induced plant death (HAMMOND-KOSACK et al. 1995 ). Also, 600 M₂ families, derived from individual Cf4 M₁ plants, were screened for suppression of PVX:Avr4-induced plant death (THOMAS et al. 1997 ).

Four Cf9 M₂ families (S568, S315, S87, and F965) in which 25% of the seedlings did not exhibit plant death were identified (Table 1). Only one Cf4 M₂ family (H249) containing 25% survivors was identified (Table 1). These mutants were investigated genetically by testcrossing to Cf0 and backcrossing to Cf9 or Cf4 (Table 1). Progeny of the four Cf9 mutants backcrossed to Cf9 and the single Cf4 mutant backcrossed to Cf4 were infected with the appropriate PVX:Avr construct. All of the progeny exhibited an Avr-dependent necrotic phenotype (Table 1). This analysis suggests that all five are recessive loss-of-function mutants (Table 1). None of the progeny from testcrosses between the mutants and Cf0 exhibited plant death after infection with recombinant PVX (Table 1). Therefore all five mutations map to the introgressed region within the Cf4 and Cf9 NILs that contain the corresponding Cf genes. The Cf-4 and Cf-9 alleles from each mutant were characterized by PCR amplification and DNA sequence analysis (Table 1).

Four recessive Cf-9 mutants isolated in a previous study were also characterized (HAMMOND-KOSACK et al. 1994B ). All nine mutants contained single-nucleotide substitutions within Cf-4 or Cf-9 or deletions encompassing Cf-9 (Table 1). Of the nine mutations characterized five would result in single-amino-acid changes in Cf-9 (Cf-9^D365N, Cf-9^D508N, Cf-9^S675L, and Cf-9^G825R) or Cf-4 (Cf-4^E662V, see Table 1 and Fig 2). Two Cf-9 mutant alleles could not be amplified by PCR (Cf-9¹ and Cf-9²) and were analyzed on DNA gel blots (results not shown). Both mutants originated from sequence deletions at the Cf-9 locus. In the case of Cf-9² a deletion of 300 bp that included the 3' terminal sequences of Cf-9 occurred. In Cf-9¹ a more extensive deletion that included Cf-9 and the adjacent paralog Hcr9-9D was observed (results not shown).

Two Cf-9 partial loss-of-function mutants that were reported previously were also investigated (Cf-9^M164 and Cf-9^M525; HAMMOND-KOSACK et al. 1994B ). However, no nucleotide mutations in the Cf-9 coding sequences, or within 1.5 kbp of their 5' and 3' flanking regions, were detected.

Meiotic stability of Cf-4:
Cf-9 has been shown to be meiotically stable (PARNISKE et al. 1997 ). Cf-4 stability was determined by testcrossing Cf4 (as female parent) to the Cf0 Avr4, Avr9 line (Fig 3A). Eleven survivors (S₁ seedlings) were recovered from 3847 testcross progeny and allowed to self-pollinate. None of the S₂ progeny were resistant to kanamycin, suggesting they lacked both Avr transgenes. This was confirmed by backcrossing S₂ plants to Cf4. None of the testcross progeny exhibited the seedling lethal phenotype (results not shown). These survivors were possibly due to self contaminants or gynogenesis (i.e., the development of an embryo from an egg cell without fertilization, see below).

Cf-4 stability was measured previously in an experiment to isolate Cf-4 by transposon tagging (TAKKEN et al. 1998 ). This was analyzed in a line containing the maize transposon Dissociation located 3 cM proximal to Cf-4 (TAKKEN et al. 1998 ). Three disease-sensitive mutants in 20,000 progeny were attributed to deletions or unequal crossing over at the Cf-4 locus. These events were most likely due to the presence of the transposon. Therefore, it appears that Cf-4 and Cf-9 (PARNISKE et al. 1997 ) are meiotically stable in homozygous lines.

Genetic selection for novel haplotypes at the Cf-4/Cf-9 locus:
We used a facile genetic selection to identify recombinants lacking Cf-4 and Cf-9 on the basis of an Avr4- and Avr9-dependent seedling lethal assay (HAMMOND-KOSACK et al. 1994A ; THOMAS et al. 1997 ). Cf-4/Cf-9 trans-heterozygotes were testcrossed to the Cf0 Avr4, Avr9 line (Fig 3B). In most testcrosses Cf-4/Cf-9 trans-heterozygotes were used as female parents since recombination is higher in the female gametophyte than in the male gametophyte (DE VICENTE and TANKSLEY 1991 ). In total, 42 S₁ survivors were recovered from 17,601 testcross progeny. These plants were characterized genetically.

The progeny of seven S₁ plants (H822, H881, H888, H889, H891, H896, and J091) from crosses where Cf-4/Cf-9 trans-heterozygotes were used as female parents segregated 15:1 for kanamycin resistance, suggesting they arose from a testcross. These S₁ plants were tested for functional Avr4 and Avr9 transgenes by testcrossing to Cf4 and Cf9 NILs. The progeny segregated 1:1 for wild-type and seedling lethal phenotypes, confirming that each S₁ parent contained a functional Avr4 and Avr9 transgene (Table S3 at http://www.genetics.org/supplemental/).

Two additional S₁ survivors (G640 and G641) arose from crosses where Cf-4/Cf-9 trans-heterozygotes were used as male parents. The S₂ progeny of these plants also segregated 15:1 for resistance to kanamycin (Table S2 at http://www.genetics.org/supplemental/). To confirm that these testcross survivors were not due to silencing of the Avr4 or Avr9 transgenes, S₂ seedlings were infected with PVX:Avr4 and PVX:Avr9 (Table S4 at http://www.genetics.org/supplemental/). No necrotic S₂ seedlings were observed, whereas Cf4 and Cf9 controls died, demonstrating that G640 and G641 must also lack Cf-4 and Cf-9.

The progeny of 26 S₁ survivors were all sensitive to kanamycin and further genetic tests showed they lacked functional Avr4 and Avr9 transgenes (results not shown). These plants arose in experiments where Cf-4/Cf-9 trans-heterozygotes were used as female parents and were most probably due to self-pollination.

Seven other S₁ plants were completely sterile and did not produce S₂ seed. Molecular analysis and functional NPT assays suggested that one survivor may be due to androgenesis in one case (i.e., the development of an embryo containing only paternal chromosomes) and gynogenesis in the others (results not shown). The rate of gynogenesis for tomato was previously estimated as 1.2–4.2 x 10^–4 (ECOCHARD et al. 1969 ; KOORNEEF et al. 1989 ; HAMZA et al. 1993 ), which is similar to the rate observed here (2 x 10^–4). Also, the sterility observed in this class of survivors has also been attributed to gynogenesis (ECOCHARD et al. 1969 ; IVANOVA et al. 2000 ).

In summary, of 42 S₁ plants recovered from 17,601 Cf-4/Cf-9 testcross progeny, only 9 originated from controlled crosses (H822, H881, H888, H889, H891, H896, J091, G640, and G641), retained functional Avr4 and Avr9 transgenes, and lacked Cf-4 and Cf-9 (Tables S2–S4).

Molecular characterization of novel haplotypes:
The nine S₂ populations were analyzed with CAPS markers to determine if loss of Cf-4 and Cf-9 was associated with flanking marker exchange. The CAPS markers MW3 and MW5 were developed to distinguish the Cf0, Cf4, and Cf9 haplotypes at the 3' and 5' ends of the MW locus, respectively (Fig 1 and MATERIALS AND METHODS). All survivors contained sequences from the Cf4 haplotype at the 5' end of the MW locus and sequences from the Cf9 haplotype at the 3' end of the MW locus (Table 2). The simplest explanation for these data is that loss of Cf-4 and Cf-9 resulted from crossing over at the Cf-4/Cf-9 locus. The analysis enabled identification of individuals homozygous for the recombinant chromosome that could be used to determine their Hcr9 composition.

DNA gel blots, PCR amplification, and DNA sequencing were used to locate the crossovers in the homozygotes. Genomic DNA was digested with BglII, blotted, and hybridized with a probe from the 5' half of Cf-9. All nine recombinants lacked BglII fragments from Cf-4 and Cf-9 (Fig 4A) and a 6.6-kbp BglII fragment derived from Hcr9-4C. All retained a 4.1-kbp BglII fragment derived from Hcr9-4B and BglII fragments derived from Hcr9-9E (Fig 4A and Fig B). Therefore, the crossovers in the new recombinants appear to be located within the Hcr9-4B to Hcr9-4C region within the Cf4 haplotype and in the Hcr9-9D to Hcr9-9E region within the Cf9 haplotype. This is a region of highly homologous DNA sequence that contains the 5' flanking region and coding sequences of Hcr9-4C and Hcr9-9E (Fig 1 and Fig 4B).

Two distinct classes of recombinants were discerned on the basis of their characteristic Hcr9-9E-derived BglII fragments. Five recombinants (H882, H888, H891, H896, and J091) retained a 3.6-kbp BglII Hcr9-9E fragment present in the Cf9 haplotype (Fig 4A). The remaining recombinants (G640, G641, H881, and H889) contained a novel Hcr9-9E BglII fragment of 3.2 kbp (Fig 4A). This difference is due to a 353-bp deletion in the Hcr9-4C 5' flanking DNA compared to the corresponding Hcr9-9E flanking region (Fig 4B). Crossovers located downstream of this deletion contain a chimeric BglII fragment of 3.2 kbp containing the Hcr9-9E coding region (Fig 4B). The crossovers in these latter recombinants could thus be delimited to a 2.289-kbp region downstream of the deletion in the Hcr9-4C 5' flanking DNA and upstream of the Hcr9-9E coding sequence (Fig 4B).

Crossovers in the other recombinants (H882, H888, H891, H896, and J091) were delimited to a region 3' to the Hcr9-4B gene in the Cf4 haplotype. The CAPS marker MW1 (Fig 4B) was used to further delimit the crossovers in these recombinants. All five recombinants retained MW sequences derived from the Cf4 haplotype (Table 2 and Fig 4). This delimited the crossovers to a 3.6-kbp region between MW1 and the 353-bp insertion in the Hcr9-9E 5' flanking region within the Cf9 haplotype (Fig 4B). DNA fragments spanning these regions were PCR amplified and sequenced. The crossovers were then localized to a region between the two closest flanking polymorphic nucleotides (Table 2). Some crossovers in class II recombinants are coincident with AT (V408, H881) and ATT (V517, V512, H891, and J091) microsatellite repeats (Fig 4C).

Of the nine recombinants reported here five were similar to the previously reported class IIa intergenic recombinants (H822, H888, H891, H896, and J091) and four were similar to the previously reported class IIb intergenic recombinants (G640, G641, H881, and H889—see Table 2; PARNISKE et al. 1997 ; THOMAS et al. 1997 ). The crossovers in G640 and G641 were each delimited to a 192-bp region that extended 178 bp into the 5' region of the Hcr9-4C and Hcr9-9E open reading frames. However, this did not result in any nucleotide substitutions in the Hcr9-9E gene. When the reciprocal products of the class II recombinants are considered (i.e., the haplotypes that carry both Cf-4 and Cf-9 and five additional Hcr9's) and the number of progeny screened, the recombination frequency in the 3.6-kbp region delimited by these recombinants is 35 kbp/cM. This number is 20 times higher than the genome average for tomato of 740 kbp/cM (TANKSLEY et al. 1992 ) and therefore represents a recombination "hotspot" in this cross.

In this genetic selection, the loss of Cf-4 or Cf-9 gene function as a result of unequal crossing over was 1 in 1955, which is not significantly different from the rate of 1 in 1500 (² = 0.64) obtained in the screen for disease sensitivity using C. fulvum race 5 (PARNISKE et al. 1997 ; THOMAS et al. 1997 ).

Measuring Cf-9 intragenic recombination in Cf-4/Cf-9 trans-heterozygotes:
No Hcr9 intragenic recombination events have been observed in the progeny of Cf-4/Cf-9 trans-heterozygotes. We used a genetic screen to identify intragenic recombinants in trans-heterozygous plants containing Cf-4 and EMS-induced mutant alleles of Cf-9. The trans-heterozygous plants were testcrossed to a Cf0 Avr9 line (Fig 3C). Intragenic recombinants between Cf-9 and another Hcr9 (such as Cf-4) could reconstitute a functional Cf-9 gene that would be identified as a necrotic individual. Cf-4 and Cf-9 contain several kilobase pairs of homologous 5' and 3' flanking DNA (tract III in Fig 1). As Cf-4 and Cf-9 encode the same amino acid sequence in the last 1 kbp of their 3' halves, in vivo intragenic recombinants in this region would be expected to be functional. Before commencing the screen, it was established that sufficient DNA could be isolated for functional characterization of the chimeric gene from seedlings showing early signs of necrosis.

To identify intragenic recombinants Cf4 was crossed to three different Cf-9 mutants (Cf-9^D508N, Cf-9^R582Ter, and Cf-9^S567L—see Table 1 and Fig 2). All of the mutations in Cf-9 are located downstream of the amino acid L457. This residue is the most C-terminal variant amino acid that distinguishes Cf-9 from Cf-4 that is required for Avr9 recognition (VAN DER HOORN et al. 2001A ; WULFF et al. 2001 ). The trans-heterozygotes were testcrossed to Cf0 Avr9 plants and the progeny were inspected for gain-of-function alleles (Table 3). No functional Cf-9 alleles were recovered in the 15,652 progeny tested (Table 3).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Characterization of EMS-induced mutants of Cf-9 and Cf-4:
Six EMS-induced point mutants of Cf-9 and one Cf-4 mutant were characterized at the molecular level. Two mutants affecting LRRs 21 and 26 (Cf-9^R582Ter and Cf-9^Q734Ter) would result in the production of truncated Cf-9 proteins lacking the C-terminal domains D, E, F, and G (Fig 2). These domains are essential for Cf-9 function (THOMAS et al. 2000 ). The Cf-4^E662V mutant would result in the substitution of a putative solvent-exposed amino acid in the xxLxLxx motif within LRR 26 (Fig 2).

The Cf-9^G825R mutant would result in substitution of an aliphatic amino acid with a positively charged amino acid in the hydrophobic -helical transmembrane domain. The G825 residue forms part of a tri-tandem GXXXG motif (see Fig 2) that is required for homo- and heterodimerization of membrane proteins in vivo (GERBER and SHAI 2001 ). Therefore, the mutant protein may not localize in the membrane or may be impaired in its interaction with another protein required for Cf-9-dependent signaling.

The other mutations (Cf-9^D365N, Cf-9^D508N, and Cf-9^S675L) would result in substitution of single putative solvent-exposed amino acids within the xxLxLxx motif in Cf-9 C-terminal LRRs (Table 1 and Fig 2). These mutations could also alter the glycosylation pattern within Cf-9 (Fig 2). The protein encoded by the Cf-9^D365N mutant allele would contain a substituted amino acid within the two additional N-terminal LRRs of Cf-9 (compared to Cf-4) that are required to induce an Avr9-dependent hypersensitive response (VAN DER HOORN et al. 2001A , VAN DER HOORN et al. 2001B ; WULFF et al. 2001 ). With the exception of the L457 residue in Cf-9 LRR 16, this is the only other example of a single-amino-acid substitution in the N-terminal LRRs that completely abolishes Cf-9 function (VAN DER HOORN et al. 2001A ; WULFF et al. 2001 ). However, it is not yet known if any of these variant alleles encode stable proteins in planta. The Cf-9^D508N, Cf-9^S675L, Cf-9^G825R, and Cf-4^E662V mutations affect amino acids that are conserved in the C-terminal LRRs of all predicted Hcr9 and Hcr2 proteins analyzed (DIXON et al. 1996 , DIXON et al. 1998 ; PARNISKE et al. 1997 ; PARNISKE and JONES 1999 ). These residues may be important for interacting with signaling partner proteins (DIXON et al. 1996 ), or they may affect protein stability.

No extragenic suppressors of Cf-9 or Cf-4 function were identified, in contrast to a previous screen (HAMMOND-KOSACK et al. 1994B ) and mutagenesis of the Cf2 line, which identified Rcr3 (DIXON et al. 2000 ). The high doses of Avr4 and Avr9 elicitor delivered by recombinant PVX may preclude the identification of mutations that partially compromise Cf-9 or Cf-4 function. The identification of extragenic suppressors of Cf-4 and Cf-9 may require more sensitive screens that could reveal partial loss-of-function alleles.

A recombination hotspot in Cf-4/Cf-9 trans-heterozygotes:
This selection identified recombinant Hcr9 haplotypes similar to our previous genetic screen (PARNISKE et al. 1997 ). Molecular analysis revealed flanking marker exchange in the progeny lacking Cf-4 and Cf-9. Conceivably, this was due to gene conversion and the conversion tracts extended >12.5–15.9 kbp beyond the 5' marker or 4.2–8.0 kbp beyond the 3' marker (Fig 1). However, this is unlikely since the ratio of gene conversions to crossovers is low in plants (DOONER and MARTINEZ-FEREZ 1997 ; SCHNABLE et al. 1998 ; CHIN et al. 2001 ). In plants, homologous but unequal crossing over does not appear to be associated with gene conversion, and this is analogous to the situation in humans (METZENBERG et al. 1991 ).

Thirteen of the 14 crossovers were localized within tract II, in the 5' flanking regions of Hcr9-4C and Hcr9-9E, while only one was located in tract I (Fig 1 and Fig 4B). The 13 crossovers in tract II are clustered in two groups within a 3.6-kbp region (Fig 4B and Fig C). The apparent preference for tract II over tract I may be due to the fact that it is significantly longer. This would be consistent with results from other studies in humans and plants (METZENBERG et al. 1991 ; DOONER and MARTINEZ-FEREZ 1997 ).

Recombination hotspots in plants usually occur within genes (BROWN and SUNDARESAN 1991 ; EGGLESTON et al. 1995 ; BUSCHGES et al. 1997 ; DOONER and MARTINEZ-FEREZ 1997 ; OKAGAKI and CLIFFORD 1997 ), but an intergenic recombination hotspot has been reported in the multigenic a1-sh2 interval of maize (YAO et al. 2002 ). The recombinants in this study define a hotspot in an Hcr9 intergenic region where the recombination frequency is 20 times the genome average (35 kbp/cM). In tomato, recombination frequencies on the short arm of chromosome 1 can vary by one order of magnitude in different interspecific crosses (BONNEMA et al. 1997 ).

The most extensive regions of sequence homology at MW are in tracts II, III, and V (Fig 1). However, a large number of the predicted recombinant classes would not be identified using the current genetic selection; e.g., crossing over within tract V would not result in recombinant chromosomes lacking Cf-4 and Cf-9. The most extensive tract of sequence homology is tract III that includes the 5' and 3' flanking sequences of Cf-4 and Cf-9 (Fig 1). Only intragenic crossovers that involved the 5' coding sequences of Cf-4 and Cf-9 might generate a chimeric Hcr9 that lacked Avr4 or Avr9 recognition specificity (VAN DER HOORN et al. 2001A , VAN DER HOORN et al. 2001B ; WULFF et al. 2001 ). In conclusion, only a fraction of all predicted crossovers could have been identified in our selection.

Intragenic recombination at the Cf-4/Cf-9 locus:
We attempted to measure the frequency of intragenic recombination in three crosses containing Cf-9 mutant alleles and Hcr9-4's (Fig 2 and Fig 3C). No gain-of-function Cf-9 alleles were detected in 15,600 testcross progeny (Table 3). Functional analysis of Cf-9 has shown that L457 in LRR 16 is essential for Avr9 recognition (Fig 2; VAN DER HOORN et al. 2001A ; WULFF et al. 2001 ). In our model the most likely Cf-9 partner for intragenic recombination would be Cf-4 (through pairing in tract III, see Fig 1), but crossovers involving other Hcr9's could also create a functional allele. The regions of sequence between nucleotides encoding L457 and the lesions in the three Cf-9 mutant alleles used in our analysis are 150, 372, or 654 bp (Table 3). Considering the interval within which crossing over would have had to occur to reconstitute a functional Cf-9 allele, the number of progeny analyzed was too low. Even if the recombination frequency in tract III were comparable to that in tract II only one gain-of-function allele would be expected in 15,600 progeny. Therefore no clear evidence for or against intragenic crossing over in Hcr9's has been provided by this study.

Comparative sequencing of the Cf-4/Cf-9 locus suggested that gene conversion and intragenic recombination have played significant roles in generating sequence variation (PARNISKE et al. 1997 ). The crossovers in two Cf-4/Cf-9 recombinants described here (G640 and G641—see Fig 4) included the 5' coding sequences of Hcr9-9E and Hcr9-4C, but these did not result in genes encoding novel Hcr9's. However, natural populations of tomato, and particularly L. pimpinellifolium, appear to be extremely polymorphic at Cf gene loci on the short arm of chromosome 1 (LAUGE et al. 1998B , LAUGE et al. 2000 ). Also, the role of intragenic recombination in Hcr9 evolution was demonstrated by analysis of natural variants of Cf-9. One allele that most likely arose by intragenic recombination between Cf-9 and Hcr9-9D was characterized (VAN DER HOORN et al. 2001B ).

It has been proposed that sequence differences in the Hcr9 intergenic regions suppress mispairing between Hcr9 paralogs, thereby preventing sequence homogenization of the gene cluster, while facilitating intragenic exchange between orthologs (PARNISKE et al. 1997 ). This would account for the apparent meiotic stability of Cf-9 (PARNISKE et al. 1997 ) and Cf-4 (this study). Our detailed analysis of the 5' flanking regions of Hcr9's at the MW locus shows that sequence duplications and insertions have been important in the evolution of the Hcr9 5' flanking DNAs, as indicated by the juxtaposition of numerous blocks of near-identical DNA sequences (Fig 5).

View larger version (19K): In this window In a new window Download PPT slide   Figure 5. Sequence relationships of Hcr9 intergenic regions at the MW locus. Blocks of color indicate stretches of near-identical DNA sequence. Most Hcr9 5' flanking sequences, except Hcr9-4C and Hcr9-9E, are highly homologous, suggesting a common evolutionary origin. Hcr9-4C and Hcr9-9E share only a short region of 5' flanking sequence homology with the other Hcr9's. The extant versions have most likely evolved by deletions, duplications, and insertions; e.g., a deletion associated with a small direct repeat (pink triangles) is indicated by a vertical arrow. No sequence inversions are apparent as indicated by the polarity of the triangles. The two horizontal right-hand arrows represent a duplicated sequence. The locations of crossovers in Cf-4/Cf-9 trans-heterozygotes are delimited by horizontal double-headed arrows relative to the ATG of Hcr9-9E (plants G640, G641, H889, H881, and V408) and Hcr9-9D (plant V514).

In the case of Cf-4/Cf-9 trans-heterozygotes, where the genes originated from different Lycopersicon species, crossing over may be restricted to the regions of maximum sequence homology. High levels of recombination are observed in this cross, but it is not clear why this is apparently restricted to the intergenic regions. It is possible that the frequency of Hcr9 intragenic recombination in this cross is below the level that could be detected by our analysis. One way to analyze the rate of intragenic recombination at this locus would be to determine the gain-of-function frequency in Cf-9 trans-heterozygotes containing distinct mutant alleles (e.g., excision alleles of Dissociation-tagged Cf-9 mutants; JONES et al. 1994 ).

Clearly, other strategies will need to be devised to determine the molecular mechanisms of Hcr9 evolution at this locus. Identifying the full spectrum of recombinant chromosomes will require high-throughput genetic or molecular screens that can reveal an exchange of flanking markers at the Cf-4/Cf-9 locus.

	FOOTNOTES

¹ These authors contributed equally to this work.

	ACKNOWLEDGMENTS

We are grateful to Edward Nicholls for help with screening EMS-mutagenized tomato families; Matthew Smoker for help with the NPT enzyme assays; Sara Perkins, Mike Hills, and Damian Alger for horticultural assistance; and David Baker and Patrick Bovill for DNA sequencing. The work with recombinant PVX was performed under the conditions of Department for Environment Food and Rural Affairs license PHL 161A/4357/(12/2002). The Sainsbury Laboratory is funded by the Gatsby Charitable Foundation. B.B.H.W. received a postgraduate training grant from the Danish Research Academy.

Manuscript received July 22, 2003; Accepted for publication January 16, 2004.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

BERNATZKY, R., E. PICHERSKY, V. S. MALIK, and S. D. TANKSLEY, 1988  CR1—a dispersed repeated element associated with the Cab-1 locus in tomato. Plant Mol. Biol. 10:423-433.[CrossRef]

BONNEMA, G., D. SCHIPPER, S. VAN HEUSDEN, P. ZABEL, and P. LINDHOUT, 1997  Tomato chromosome 1: high resolution genetic and physical mapping of the short arm in an interspecific Lycopersicon esculentum x L. peruvianum cross. Mol. Gen. Genet. 253:455-462.[CrossRef][Medline]

BROWN, J. and V. SUNDARESAN, 1991  A recombinational hotspot in the maize A1 intragenic region. Theor. Appl. Genet. 81:185-188.[CrossRef]

BÜSCHGES, R., K. HOLLRICHER, R. PANSTRUGA, G. SIMONS, and M. WOLTER et al., 1997  The barley Mlo gene: a novel control element of plant pathogen resistance. Cell 88:695-705.[CrossRef][Medline]

CHIN, D. B., R. ARROYO-GARCIA, O. E. OCHOA, R. V. KESSELI, and 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.[Abstract/Free Full Text]

COLLINS, N., J. DRAKE, M. AYLIFFE, Q. SUN, and J. ELLIS et al., 1999  Molecular characterization of the maize Rp1-D rust resistance haplotype and its mutants. Plant Cell 11:1365-1376.[Abstract/Free Full Text]

DANGL, J. L. and J. D. G. JONES, 2001  Plant pathogens and integrated defence responses. Nature 411:826-833.[CrossRef][Medline]

DE VICENTE, M. C. and S. D. TANKSLEY, 1991  Genome-wide reduction in recombination of backcross progeny derived from male versus female gametes in an interspecific cross of tomato. Theor. Appl. Genet. 83:173-178.[CrossRef]

DIXON, M. S., D. A. JONES, J. S. KEDDIE, C. M. THOMAS, and K. HARRISON et al., 1996  The tomato Cf-2 disease resistance locus comprises two functional genes encoding leucine-rich repeat proteins. Cell 84:451-459.[CrossRef][Medline]

DIXON, M. S., K. HATZIXANTHIS, D. A. JONES, K. HARRISON, and J. D. G. JONES, 1998  The tomato Cf-5 disease resistance gene and six homologs show pronounced allelic variation in leucine-rich repeat copy number. Plant Cell 10:1915-1925.[Abstract/Free Full Text]

DIXON, M. S., C. GOLSTEIN, C. M. THOMAS, E. A. VAN DER BIEZEN, and J. D. G. JONES, 2000  Genetic complexity of pathogen perception by plants: the example of Rcr3, a tomato gene required specifically by Cf-2.. Proc. Natl. Acad. Sci. USA 97:8807-8814.[Abstract/Free Full Text]

DODDS, P. N., G. J. LAWRENCE, and J. G. ELLIS, 2001a  Contrasting modes of evolution acting on the complex N locus for rust resistance in flax. Plant J. 27:439-453.[CrossRef][Medline]

DODDS, P. N., G. J. LAWRENCE, and J. G. ELLIS, 2001b  Six amino acid changes confined to the leucine-rich repeat beta-strand/beta-turn motif determine the difference between the P and P2 rust resistance specificities in flax. Plant Cell 13:163-178.[Abstract/Free Full Text]

DOONER, H. K. and I. M. MARTÍNEZ-FÉREZ, 1997  Recombination occurs uniformly within the bronze gene, a meiotic recombination hotspot in the maize genome. Plant Cell 9:1633-1646.[Abstract]

ECOCHARD, R., M. S. RAMANNA, and D. DE NETTANCOURT, 1969  Detection and cytological analysis of tomato haploids. Genetica 40:181-190.[CrossRef]

EGGLESTON, W. B., M. ALLEMAN, and J. L. KERMICLE, 1995  Molecular organization and germinal instability of R-stippled maize. Genetics 141:347-360.[Abstract]

ELLIS, J. G., G. J. LAWRENCE, J. E. LUCK, and P. N. DODDS, 1999  Identification of regions in alleles of the flax rust resistance gene L that determine differences in gene-for-gene specificity. Plant Cell 11:495-506.[Abstract/Free Full Text]

GERBER, D. and Y. SHAI, 2001  In vivo detection of hetero-association of glycophorin-A and its mutants within the membrane. J. Biol. Chem. 276:31229-31232.[Abstract/Free Full Text]

HAMMOND-KOSACK, K. E., K. HARRISON, and J. D. G. JONES, 1994a  Developmentally regulated cell death on expression of the fungal avirulence gene Avr9 in tomato seedlings carrying the disease-resitance gene Cf-9.. Proc. Natl. Acad. Sci. USA 91:10445-10449.[Abstract/Free Full Text]

HAMMOND-KOSACK, K. E., K. HARRISON, and J. D. G. JONES, 1994b  Identification of two genes required in tomato for full Cf-9-dependent resistance to Cladosporium fulvum. Plant Cell 6:361-374.[Abstract]

HAMMOND-KOSACK, K. E., B. J. STASKAWICZ, J. D. G. JONES, and D. C. BAULCOMBE, 1995  Functional expression of a fungal avirulence gene from a modified potato virus X genome. Mol. Plant-Microbe Interact. 8:181-185.

HAMZA, S., C. CAMILLERI, J.-M. POLLIEN, H. VAUCHERET, and J.-P. BOURGIN et al., 1993  Selection for spontaneous tomato haploids using a conditional lethal marker. Theor. Appl. Genet. 86:657-664.

HELLENS, R. P., E. A. EDWARDS, N. R. LEYLAND, S. BEAN, and P. M. MULLINEAUX, 2000  pGreen: a versatile and flexible binary Ti vector for Agrobacterium-mediated plant transformation. Plant Mol. Biol. 42:819-832.[CrossRef][Medline]

IVANOVA, S. V., L. I. DOLGODVOROVA, G. I. KARLOV, and E. V. KUCHKOVSKAJA, 2000  Morphometric and cytogenetic characteristics of haploid tomato plants. Russ. J. Genet. 36:41-50.

JONES, D. A., C. M. THOMAS, K. E. HAMMOND-KOSACK, P. J. BALINT-KURTI, and J. D. G. JONES, 1994  Isolation of the tomato Cf-9 gene for resistance to Cladosporium fulvum by transposon tagging. Science 266:789-793.[Abstract/Free Full Text]

KOORNEEF, M., J. A. M. VAN DIEPEN, C. J. HANHART, A. C. KIEBOOM-DE WAART, and L. MARINELLI et al., 1989  Chromosomal instability in cell- and tissue cultures of tomato haploids and diploids. Euphytica 43:179-186.[CrossRef]

LAUGÉ, R., A. P. DMITRIEV, M. H. A. J. JOOSTEN, and P. J. G. M. DE WIT, 1998a  Additional resistance gene(s) against Cladosporium fulvum on the Cf-9 introgression segment are associated with strong PR accumulation. Mol. Plant-Microbe Interact. 11:301-308.

LAUGÉ, R., M. H. A. J. JOOSTEN, J. P. W. HAANSTRA, P. H. GOODWIN, and P. LINDHOUT et al., 1998b  Successful search for a resistance gene in tomato targeted against a virulence factor of a fungal protein. Proc. Natl. Acad. Sci. USA 95:9014-9018.[Abstract/Free Full Text]

LAUGÉ, R., P. H. GOODWIN, P. J. G. M. DE WIT, and M. H. A. J. JOOSTEN, 2000  Specific HR-associated recognition of secreted proteins from Cladosporium fulvum occurs in both host and non-host plants. Plant J. 23:735-745.[CrossRef][Medline]

LUCK, J. E., G. J. LAWRENCE, E. J. FINNEGAN, D. A. JONES, and J. G. ELLIS, 1998  A flax transposon identified in two spontaneous mutant alleles of the L6 rust resistance gene. Plant J. 16:365-369.[CrossRef][Medline]

LUCK, J. E., G. J. LAWRENCE, P. N. DODDS, K. W. SHEPHERD, and J. G. ELLIS, 2000  Regions outside of the leucine-rich repeats of flax rust resistance proteins play a role in specificity determination. Plant Cell 12:1367-1377.[Abstract/Free Full Text]

MCDOWELL, J. M., M. DHANDAYDHAM, T. A. LONG, M. G. M. AARTS, and S. GOFF et al., 1998  Intragenic recombination and diversifying selection contribute to the evolution of the downy mildew resistance at the RPP8 locus of Arabidopsis.. Plant Cell 10:1861-1874.[Abstract/Free Full Text]

METZENBERG, A. B., G. WURZER, T. H. J. HUISMAN, and O. SMITHIES, 1991  Homology requirements for unequal crossing over in humans. Genetics 128:143-161.[Abstract]

MEYERS, B. C., D. B. CHIN, K. A. SHEN, S. SIVARAMAKRISHNAN, and D. O. LAVELLE et al., 1998  The major resistance gene cluster in lettuce is highly duplicated and spans several megabases. Plant Cell 10:1817-1832.[Abstract/Free Full Text]

MICHELMORE, R. W. 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.[Abstract/Free Full Text]

NOËL, L., T. L. MOORES, E. A. VAN DER BIEZEN, M. PARNISKE, and M. J. DANIELS et al., 1999  Pronounced intraspecific haplotype divergence at the RPP5 complex disease resistance locus of Arabidopsis. Plant Cell 11:2099-2111.[Abstract/Free Full Text]

OKAGAKI, R. J. and F. W. CLIFFORD, 1997  Analysis of recombination sites within the maize waxy locus. Genetics 147:815-821.[Abstract]

PANTER, S. N., K. E. HAMMOND-KOSACK, K. HARRISON, J. D. G. JONES, and D. A. JONES, 2002  Developmental control of promoter activity is not responsible for mature onset of Cf-9B-mediated resistance to leaf mold in tomato. Mol. Plant-Microbe Interact. 15:1099-1107.[Medline]

PARNISKE, M. and J. D. G. JONES, 1999  Recombination between diverged clusters of the tomato Cf-9 plant disease resistance gene family. Proc. Natl. Acad. Sci. USA 96:5850-5855.[Abstract/Free Full Text]

PARNISKE, M., K. E. HAMMOND-KOSACK, C. GOLSTEIN, C. M. THOMAS, and D. A. JONES et al., 1997  Novel disease resistance specificities result from sequence exchange between tandemly repeated genes at the Cf-4/9 locus of tomato. Cell 91:821-832.[CrossRef][Medline]

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.[Abstract]

RIVAS, S. and C. M. THOMAS, 2002  Recent advances in the study of tomato Cf resistance genes. Mol. Plant Pathol. 3:277-282.

SCHNABLE, P. S., A.-P. HSIA, and B. NIKOLAU, 1998  Genetic recombination in plants. Curr. Opin. Plant Biol. 1:123-129.[CrossRef][Medline]

SUN, Q., N. C. COLLINS, M. AYLIFFE, S. M. SMITH, and J. DRAKE et al., 2001  Recombination between paralogues at the rp1 rust resistance locus in maize. Genetics 158:423-438.[Abstract/Free Full Text]

TAKAHASHI, Y., Y. NIWA, Y. MACHIDA, and T. NAGATA, 1990  Location of the cis-acting auxin-responsive region in the promoter of the par gene from tobacco mesophyll protoplasts. Proc. Natl. Acad. Sci. USA 87:8013-8016.[Abstract/Free Full Text]

TAKKEN, F. L. W., D. SCHIPPER, H. J. J. NIJKAMP, and J. HILLE, 1998  Identification and Ds-tagged isolation of a new gene at the Cf-4 locus of tomato involved in disease resistance to Cladosporium fulvum race 5. Plant J. 20:279-288.[CrossRef]

TAKKEN, F. L. W., C. M. THOMAS, M. H. A. J. JOOSTEN, C. GOLSTEIN, and N. WESTERINK et al., 1999  A second gene at the tomato Cf-4 locus confers resistance to Cladosporium fulvum through recognition of a novel avirulence determinant. Plant J. 20:279-288.[Medline]

TANKSLEY, S. D., M. W. GANAL, J. P. PRINCE, M. C. DEVICENTE, and M. W. BONIERBALE et al., 1992  High-density molecular linkage maps of the tomato and potato genomes. Genetics 132:1141-1160.[Abstract]

THOMAS, C. M., D. A. JONES, M. PARNISKE, K. HARRISON, and P. J. BALINT-KURTI et al., 1997  Characterization of the tomato Cf-4 gene for resistance to Cladosporium fulvum identifies sequences that determine recognitional specificity in Cf-4 and Cf-9. Plant Cell 9:2209-2224.[Abstract]

THOMAS, C. M., S. J. TANG, K. HAMMOND-KOSACK, and J. D. G. JONES, 2000  Comparison of the hypersensitive response induced by the tomato Cf-4 and Cf-9 genes in Nicotiana spp.. Mol. Plant-Microbe Interact. 13:465-469.[Medline]

VAN DER HOORN, R. A. L., R. ROTH, and P. J. G. M. DE WIT, 2001a  Identification of distinct specificity determinants in resistance protein Cf-4 allows construction of a Cf-9 mutant that confers recognition of avirulence protein AVR4. Plant Cell 13:273-285.[Abstract/Free Full Text]

VAN DER HOORN, R. A. L., M. KRUIJT, R. ROTH, B. F. BRANDWAGT, and M. H. A. J. JOOSTEN et al., 2001b  Intragenic recombination generated two distinct Cf genes that mediate AVR9 recognition in the natural population of Lycopersicon pimpinellifolium.. Proc. Natl. Acad. Sci. USA 98:10493-10498.[Abstract/Free Full Text]

WULFF, B. B. H., C. M. THOMAS, M. SMOKER, M. GRANT, and J. D. G. JONES, 2001  Domain swapping and gene shuffling identify sequences required for induction of an Avr-dependent hypersensitive response by the tomato Cf-4 and Cf-9 proteins. Plant Cell 13:255-272.[Abstract/Free Full Text]

YAO, H., Q. ZHOU, J. LI, H. SMITH, and M. YANDEAU et al., 2002  Molecular characterisation of meiotic recombination across the 140-kb multigenic a1-sh2 interval of maize. Proc. Natl. Acad. Sci. USA 99:6157-6162.[Abstract/Free Full Text]

This article has been cited by other articles:

				X. Sun, Y. Zhang, S. Yang, J.-Q. Chen, B. Hohn, and D. Tian Insertion DNA Promotes Ectopic Recombination during Meiosis in Arabidopsis Mol. Biol. Evol., October 1, 2008; 25(10): 2079 - 2083. [Abstract] [Full Text] [PDF]

				N. C. Collins, N. J. Shirley, M. Saeed, M. Pallotta, and J. P. Gustafson An ALMT1 Gene Cluster Controlling Aluminum Tolerance at the Alt4 Locus of Rye (Secale cereale L.) Genetics, May 1, 2008; 179(1): 669 - 682. [Abstract] [Full Text] [PDF]

				R. A.L. van der Hoorn, B. B.H. Wulff, S. Rivas, M. C. Durrant, A. van der Ploeg, P. J.G.M. de Wit, and J. D.G. Jones Structure-Function Analysis of Cf-9, a Receptor-Like Protein with Extracytoplasmic Leucine-Rich Repeats PLANT CELL, March 1, 2005; 17(3): 1000 - 1015. [Abstract] [Full Text] [PDF]

				E. A. Kellogg and J. L. Bennetzen The evolution of nuclear genome structure in seed plants Am. J. Botany, October 1, 2004; 91(10): 1709 - 1725. [Abstract] [Full Text] [PDF]