Introduction

Blackleg disease of canola and other Brassica crops is caused by the haploid ascomycete pathogen Leptosphaeria maculans (Sun et al. 2007; Fitt et al. 2006). The pathogen has been reported in North America (Canada, USA and Mexico), most countries in Europe, Australia, Brazil, Argentina, and a few countries in Africa (Fitt et al. 2006). Despite ongoing resistance breeding, financial losses to this disease range from 22 to 300 million dollars per year in canola-producing regions in the world (Fitt et al. 2006).

Two types of genetic resistance to L. maculans have been described in Brassica species; qualitative resistance (that is usually assayed at the seedling stage) and quantitative resistance (that is usually studied in adult-plants) (Ansan-Melayah et al. 1998; Pilet et al. 1998; Dion et al. 1995; Ferreria et al. 1995; Rimmer 2006). Qualitative or race-specific resistance follows the gene-for-gene model for plant–pathogen interactions (Flor 1971) and includes the genes Rlm1, Rlm2 and Rlm4 (Ansan-Melayah et al. 1998), Rlm3, Rlm5, Rlm6, and Rlm8 (Balesdent et al. 2002), Rlm7 and Rlm9 (Delorume et al. 2004), LepR1 and LepR2 (Yu et al. 2005), and LepR3 (Li and Cowling 2003). The genetic control of quantitative resistance is currently more poorly defined but appears to play a significant role in the protection of Brassica crops (Rimmer 2005).

Genetic studies in L. maculans have identified nine avirulence gene specificities, AvrLm1 to 9, corresponding to nine resistance genes of Brassica hosts. Most of these genes are contained within two genetic clusters, the AvrLm1-2-6 cluster (Balesdent et al. 2002) and the AvrLm3-4-7-9 cluster (Balesdent et al. 2005) while AvrLm5 and AvrLm8 are not linked to any of the other genes (Balesdent et al. 2002). Three of the avirulence genes have been cloned; AvrLm1 (Gout et al. 2006), AvrLm6 (Fudal et al. 2007) and the dual avirulence gene, AvrLm4-7 (Parlange et al. 2009), that is recognized by two distinct resistance genes from Brassica napus, Rlm4 and Rlm7. The proteins predicted from the coding sequences of all three genes are small and secreted. The products of AvrLm6 (Fudal et al. 2007) and AvrLm4-7 (Parlange et al. 2009) are both cysteine-rich, while the protein encoded by the AvrLm1 gene is predicted to have a paucity of disulphide bridges (Gout et al. 2006).

The LepR1 gene for resistance to L. maculans was derived from Brassica rapa subsp. sylvestris and was transferred to canola using two independent crossing strategies as bridges; resynthesis of B. napus (Crouch 1994) and the formation of AAC allotriploids with backcrossing to B. napus (S. R. Rimmer, unpublished). LepR1 controls race-specific interactions with L. maculans and was effective against most isolates tested, including ones collected from Canada, Australia and Europe (Yu et al. 2005). Yu et al. (2005) used a population of B. napus double haploid (DH) lines segregating for LepR1 to map the gene to chromosome 2 of the Brassica A-genome (B. napus linkage group N2 corresponds to Brassica chromosome A2).

A rare L. maculans isolate, 87-41, that is virulent on B. napus plants carrying LepR1 was identified by Yu et al. (2005). This made it theoretically possible to determine the genetic location in the L. maculans genome of AvrLepR1, the avirulence gene with which LepR1 interacts. In this report, we describe the genetic mapping of AvrLepR1 relative to marker-loci defined by sequence-related amplified polymorphism (SRAP) markers (Li and Quiros 2001) and the specificity testing of AvrLepR1 to establish its relationship to the AvrLm3, AvrLm4, AvrLm7 and AvrLm9 genes which all mapped to a similar region of the L. maculans genome.

Materials and methods

B. napus lines and L. maculans isolates

The B. napus line ddm-12-6s-1 (S. R. Rimmer, unpublished), carries LepR1 and is a BC2S2 selection from a cross between double haploid B. napus DH12075 (Mayerhofer et al. 2005) and B. rapa subsp. sylvestris (Crouch 1994), backcrossed to N–o-1, a DH line derived from ‘Westar’ (Sharpe et al. 1995). A distinct selection from the ‘Westar’ canola variety, which was described by Chen and Fernando (2006), was used as a positive control for infection by L. maculans. The L. maculans isolate, 87-41 (Yu et al. 2005), was expected to be virulent on both ddm-12-6s-1 and ‘Westar’, while isolates WA-74, 99-49 and 99-56 (Yu et al. 2005) were expected to be avirulent on ddm-12-6s-1 but virulent on ‘Westar’. All four isolates were obtained from the collection of Dr. S. R. Rimmer. The following B. napus differential lines and cultivars, 22-1-1, ‘Falcon’, 23-1-1 and ‘Darmor’, which harbored Rlm3, Rlm4, Rlm7 and Rlm9, respectively (Balesdent et al. 2005), were used to differentiate the corresponding Avr genes in the L. maculans parental isolates and progeny.

Inoculum preparation and virulence phenotyping

The preparation of pycnidiospores and the inoculation of B. napus cotyledons were performed as described previously (Ferreria et al. 1995). Each L. maculans isolate was tested on 12 seedlings of the ddm-12-6s-1 tester line and 12 seedlings of ‘Westar’ as a highly susceptible control. The disease reactions were scored 12, 14 and 21 days after inoculation and rated using the 0–9 scale described by Williams (1985).

Crossing isolates and tetrad analysis

L. maculans isolates were crossed by pairing single-pycnidiospore isolates as described by Mengistu et al. (1995) and tetrad analysis was performed following the protocol of Gall et al. (1994). Toothpicks bearing pseudothecia were stored in vials with silica gel after air-drying.

Sequence-related amplified polymorphism markers

Sequence-related amplified polymorphism (SRAP) markers were generated using the methods described by Li and Quiros (2001). Each amplification reaction used L. maculans template DNA, one fluorescently labeled primer and one unlabelled primer. PCR amplifications were carried out using a programmable thermal controller (BIO RAD thermal cycler gradient, BIO RAD, California, USA). Four reaction products, each using a distinct fluorescent dye, were typically pooled before size separation and detection of the fluorescently labeled DNA amplification products using an ABI 3100 Genetic Analyzer (ABI, California). The four distinguishable dyes used to label samples were FC1 (Blue), PM88 (Red), SA7 (Green) and SA12 (Yellow). The LIZ 500 (Orange) fluorescent dye was used to label the DNA fragments of the internal size standard. All five dyes were supplied by ABI (ABI, California). Genomic DNA was extracted from parents and F1 progeny as described by Pongam et al. (1998).

Molecular markers for the L. maculans mating type locus and the avirulence genes AvrLm1 and AvrLm4-7

The primer pairs used to amplify the two allelic variants of the mating type locus were described by Cozijnsen and Howlett (2003). The diagnostic primer pairs 1/2 and 3/4 (Table 1) were designed from the AvrLm1 gene sequence (EMBL database, accession number AM084345) and the BAC 28G8 sequence (EMBL database, accession number AM998637), respectively. These sequences were positioned on the same chromosome as the AvrLm4-7 gene, ~15 cM distant from it (Parlange et al. 2009). The AvrLm4-7 locus of L. maculans was amplified using the AVRint-UP/-Lo primer pair described by Parlange et al. (2009). The PCR-amplification products of AvrLm4-7 variants were cloned using TOPO TA cloning kits (Invitrogen, USA) and submitted for contract sequencing. The primer 5/6 amplified the 99-56 allele at nucleotide 212 of the AvrLm4-7 gene (SNP212) while the primer pair 7/8 amplified the 87-41 allele at nucleotide 241 (SNP241) (Table 1).

Table 1 Nucleotide sequences of primer pairs used to amplify fragments of AvrLm1 (1/2), a region of the L. maculans genome 15 cM removed from AvrLm4-7 (3/4) and SNPs primers (5/6 and 7/8) designed to preferentially amplify specific alleles at the AvrLm4-7 locus
Full size table

Genetic map construction

Linkage analysis was performed using the computer program JoinMap version 3.0 (Van Ooijen and Voorrips 2001). Minimum LOD (logs of the odds ratios of linkage vs. no linkage) scores of 4.0 (maximum recombination fraction of 0.4) were used to group loci. Recombination frequencies were converted to centi-Morgan (cM) map distances using Kosambi’s mapping function (Kosambi 1944).

Results

Selecting the parental isolates for mapping AvrLepR1

An experimental system for the genetic mapping of AvrLepR1 required a pair of L. maculans isolates exhibiting contrasting abilities to infect a B. napus line carrying LepR1. B. napus lines ddm-12-6s-1, carrying LepR1, and ‘Westar’ were used to test the virulence of the four L. maculans isolates. As expected, isolate 87-41 was virulent on both genotypes (Fig. 1; Table 2). Of the three isolates expected to be avirulent, WA-74 and 99-43 produced intermediate reactions on ddm-12-6s-1 while 99-56 was obviously avirulent (Fig. 1; Table 2). All four L. maculans isolates were highly virulent on the ‘Westar’ positive control (Table 2). Repeating the whole experiment yielded essentially identical results (Table 2). The four L. maculans isolates were tested with primers diagnostic for the mating type locus. Isolates 87-41, WA-74 and 99-43 yielded the amplification products expected for the MAT1-1 mating type while 99-56 yielded those expected for the MAT1-2 mating type.

Fig. 1
figure 1

The aerial portions of four seedlings of B. napus line ddm-12-6s-1, which carries the LepR1 gene for resistance to L. maculans (the fungus that causes blackleg disease in canola). One cotyledon of each seedling has been wounded and inoculated with L. maculans pycnidiospores as described by Williams (1985). The two seedlings on the right were inoculated with the avirulent isolate 99-56 which carries the common, active form of the AvrLepR1 gene and which elicited a rapid resistance reaction from line ddm-12-6s-1, corresponding to a disease rating of 1 (Williams 1985). The two seedlings on the left were inoculated with the virulent isolate 87-41 which carries the uncommon, inactive form of the avirulence gene (avrLepR1) and which produced a conspicuously successful infection on line ddm-12-6s-1, corresponding to a disease rating of 8 (Williams 1985)

Full size image
Table 2 Virulence/avirulence phenotypes of four L. maculans isolates (87-41, WA-74, 99-43 and 99-56) used to inoculate B. napus cotyledons of a susceptible control line (‘Westar’) and a plant line carrying LepR1 (ddm-12-6s-1)
Full size table

L. maculans mating and virulence/avirulence phenotyping of the resulting recombinants

Based on the above results, 87-41 and 99-56 were crossed to produce a population segregating for AvrLepR1. A total of 94 single-ascospore progenies were isolated from the L. maculans cross and each isolate was tested for virulence on B. napus line ddm-12-6s-1 (that carries LepR1) and ‘Westar’ (the susceptible control). The population of L. maculans isolates segregated for virulence (50 isolates) and avirulence (44 isolates) and the segregation ratio approximated to the 1:1 ratio (χ2 = 0.38, P = 0.66) expected for genetic control of the phenotype by a single gene (Fig. 2). The 1:1 segregation ratio was supported by the analysis of complete sets of ascospores from two asci, each segregated 1:1 avirulent/virulent. All 94 progenies were virulent on ‘Westar’.

Fig. 2
figure 2

The frequency distribution of disease ratings within the L. maculans population of 94 single-ascospore isolates derived from the 87-41 (virulent) × 99-56 (avirulent) cross. The population was segregating for virulence on B. napus ddm-12-6s-1, controlled by AvrLepR1, and the disease ratings were scored 14 days post-inoculation using the 0–9 rating scale described by Williams (1985)

Full size image

Identifying SRAP markers informative for the L. maculans cross

SRAP primer pairs were screened using DNA of the two parental isolates (87-41 and 99-56) separately. Informative primer pairs that amplified one or more DNA fragments that were specific to one or the other of the parental isolates were selected for further use. In addition, bulked segregant analysis (BSA; Michelmore et al. 1991) employing contrasting bulked samples, one containing equal amounts of DNA from 12 avirulent isolates and the other containing equal amounts of DNA from 12 virulent isolates, was used to identify SRAP markers closely linked to AvrLepR1. Most primer pairs produced 30-40 amplification products with lengths ranging from 50 to 600 base pairs (bp). The average number of informative amplification products per primer pair was 1.1, with the most informative primer pair producing six products that were polymorphic between the two parents.

The segregation patterns of 283 informative SRAP polymorphisms were scored in the 94 isolates of the L. maculans mapping population. The informative DNA fragments were scored as either present or absent. Of the 283 loci assayed with SRAP markers, 254 loci (~90%) had balanced 1:1 segregation ratios (P > 0.05) while the remaining 29 loci (~10%) exhibited distorted segregation ratios (P < 0.05). This indicated that the 94 single-ascospore progenies represented a fairly balanced population with only an approximately twofold over representation of loci with distorted segregation ratios. Of the 29 distorted loci (P < 0.05), only four were highly distorted (P < 0.01) and of these only one had a probability <0.001.

Genetic linkage analysis and mapping the AvrLepR1 avirulence gene of L. maculans

Genetic linkage analysis was carried out on the 283 SRAP-defined marker-loci. This established 36 linkage groups each containing at least two loci and together representing 259 loci (Fig. 3) with 24 marker-loci remaining unlinked. The linkage groups spanned 2076.4 cM of L. maculans genome with individual linkage groups ranging from 13.1 to 164.8 cM in length (Fig. 3).

Fig. 3
figure 3

Genetic map of L. maculans based on 264 loci associated into 36 linkage groups (LG’s). The majority of loci, 259, were detected using sequence-related amplified polymorphism (SRAP) markers. The mating type locus (MAT1) on LG7, AvrLm1 on LG11 and AvrLm4-7 on LG4, were each assayed using specially designed PCR-based markers. The AvrLepR1 locus was mapped on LG4. The locus names are shown to the right of each LG and genetic distances from the top in cM are indicated on the left of each LG

Full size image

The segregation pattern of the molecular markers specific for the L. maculans mating type gene positioned the mating type locus (MAT) 15.8 cM from the top of linkage group 7 (Fig. 3). PCR amplifications with primers specific for AvrLm1 (Table 1) yielded a 333 bp fragment using DNA from isolate 87-41 and produced no amplification product from isolate 99-56 DNA confirming the presence of AvrLm1 in isolate 87-41 and its absence from isolate 99-56. Segregation for the AvrLm1 marker positioned the gene 61.3 cM from the top of linkage group 13 (Fig. 3).

Segregation for virulence/avirulence on B. napus line ddm-12-6s-1 identified the L. maculans locus (AvrLepR1) that controls avirulence against the host’s LepR1 gene. The L. maculans AvrLepR1 gene was located 10.1 cM from the top of linkage group 4 in the 13.1 cM interval flanked by, SBG49-110 and FT161-223 (Fig. 3). Linkage group 4 was 164.8 cM in length and carried 21 SRAP-defined loci. It also carried the 28G8-490 and P1314-300 marker-loci that are diagnostic of the chromosome carrying the AvrLm4-7 gene. This suggested that AvrLepR1 could be a new member of the AvrLm3-4-7-9 genetic cluster (Balesdent et al. 2005; Van de Wouw et al. 2008) or even a new allele of the AvrLm4-7 gene. Both the parental isolates (87-41 and 99-56) were avirulent on B. napus plants carrying either Rlm4 or Rlm7, indicating that each isolate carried a functional AvrLm4-7 gene and that the AvrLm4-7 locus could not be mapped in the segregating population based on the segregation of differential disease reactions. In order to test the possibility that the (functionally indistinguishable) alleles of AvrLm4-7 in isolates 87-41 and 99-56 could be distinguished at the level of precise nucleotide sequence, the AvrLm4-7 locus of each parental isolate was amplified (using the AVRint-UP/-Lo primer pair: Parlange et al. 2009), cloned and sequenced with fivefold replication. A comparison of the sequences of the two parental alleles and the published sequence of AvrLm4-7 gene (Parlange et al. 2009) revealed that the 99-56 allele carried SNP212 (a substitution of A for C at nucleotide 212 of the amplification product) while the 87-41 allele carried SNP241 (a substitution of C for G at nucleotide 241 of the amplification product). Genetic markers based on the differential PCR-amplification of these SNP’s using primer pairs 5/6 and 7/8 (Table 1) allowed segregation at the AvrLm4-7 locus to be scored in the mapping population and the AvrLm4-7 locus to be positioned 6.8 cM below the AvrLepR1 locus on linkage group 4 (Fig. 4) and in the 5.4 cM interval flanked by FT161-223 and P1314-300. To confirm this result, the AvrLm4-7 gene was amplified, cloned and sequenced in six recombinant isolates carrying crossovers between the FT161-223 and P1314-300 marker-loci flanking to AvrLm4-7. The AvrLm4-7 gene regions from six non-recombinant isolates, three avirulent and three virulent, were also cloned and sequenced as controls. In every case, the gene-sequence analysis of the AvrLm4-7 alleles confirmed the earlier analysis based on SNP markers.

Fig. 4
figure 4

A large-scale view of the top of linkage group 4 showing the relative positions of AvrLepR1 and AvrLm4-7. The locus names are shown to the right of the LG and the genetic distances from the top in cM are indicated on the left of the LG

Full size image

AvrLepR1 has a specificity distinct from those of AvrLm3, AvrLm4, AvrLm7 and AvrLm9

With AvrLepR1 possibly mapping to the AvrLm3-4-7-9 genetic cluster, it was considered prudent to investigate whether AvrLepR1 interacted with Rlm3, Rlm4, Rlm7 or Rlm9. The interactions of each parent of the L. maculans mapping population (87-41 and 99-56), together with a random sample of 30 of the single-ascospore progenies, were tested with a set of six B. napus lines. The lines 22-1-1 (Rlm3), cv. ‘Falcon’ (Rlm4), 23-1-1 (Rlm7) and cv. ‘Darmor’ (Rlm9) were each reported to carry a single distinct gene for resistance to L. maculans, while ddm-12-6s-1 carried LepR1 and cv. ‘Westar’ was used as the susceptible control. The results are summarized in Table 3. The two parental isolates and the 30 progenies all interacted identically with line 22-1-1, cv. ‘Falcon’, line 23-1-1, cv. ‘Darmor’ and ‘Westar’ (the isolates were either uniformly virulent or uniformly avirulent: Table 3). Only on ddm-12-6s-1 did the L. maculans isolates exhibit variation. Isolate 87-41 and 15 of the recombinants were virulent on ddm-12-6s-1, while isolate 99-56 and the remaining 15 recombinants were avirulent on this line. These results indicated that the LepR1—AvrLepR1 gene interaction was distinct from those involving Rlm3, Rlm4, Rlm7 or Rlm9.

Table 3 Interaction phenotypes of the L. maculans parental isolates and of a set of 30 single-spore progenies all tested on the same panel of B. napus lines, each carrying a defined resistance gene
Full size table

Discussion

LepR1 is a new gene for resistance to L. maculans that was introduced into amphidiploid B. napus (canola) recently from the related diploid species, B. rapa, and that is harbored by the B. napus line ddm-12-6s-1 (Yu et al. 2005). Several gene-for-gene interactions involving specific L. maculans avirulence genes and the corresponding resistance genes of the B. napus host have already been described (Ansan-Melayah et al. 1995; Balesdent et al. 2001, 2005). The genetic mapping of AvrLepR1 positioned the gene on linkage group 4 close to, or part of, the AvrLm3-4-7-9 cluster reported by Balesdent et al. (2005).

A new genetic map of L. maculans, based on a population of 94 single-ascospore progenies and segregating for virulence on B. napus line ddm-126s-1, was developed. The population was assayed with a set of informative PCR-based markers (predominantly SRAP markers) and 288 polymorphic loci were identified. Genetic linkage analysis established 36 linkage groups spanning a total of 2,076.4 cM (Fig. 3) with 24 marker-loci remaining unlinked. The relatively large number of linkage groups, 36, compared to 21 groups in the L. maculans genetic map developed by Cozijnsen et al. (2000) and 15 groups (containing 5 or more markers) reported by Kuhn et al. (2006) and the high proportion of unlinked loci (24 out of 288, 8% of the loci) suggest that the marker coverage of the L. maculans genome was incomplete and that some chromosomes are probably represented by more than one linkage group. The L. maculans map described by Cozijnsen et al. (2000) was based on 159 markers mapped on 21 linkage groups with 18 loci remaining unlinked and those of Kuhn et al. (2006) were based on between 177 and 443 markers mapped on 15 major groups. Unfortunately, the use of a novel set of genetic markers makes a detailed comparison with other maps of L. maculans impossible.

Overall, SRAP markers provided an inexpensive system for genetic mapping in L. maculans, that required very little development. SRAPs are a molecular-marker system with some advantages over RAPD, RFLP, and AFLP markers (Li and Quiros 2001). They have been used for genetic mapping in plants (Li and Quiros 2001; Ferriol et al. 2003; Sun et al. 2007; Dusabenyagasani and Fernando 2008) and fungi (Dodds et al. 2004) and also in genetic diversity studies in a range of organisms (Fernando et al. 2006; Sun et al. 2006; Zhang et al. 2005). The clustering of SRAP markers (Fig. 3) was one reason for the marker coverage of the genome being incomplete despite the large number of marker-loci assayed. Marker-dense regions of the L. maculans genome were also reported by Kuhn et al. (2006). This marker clustering is a general feature of many PCR-based genetic markers and has been noted in genetic maps of a range of fungi, including, Mycosphaerella graminicola (Kema et al. 2002), Cochliobolus sativus (Zhong et al. 2002) and Blumeria graminis (Pedersen et al. 2002).

Twenty-nine or 10% of the SRAP marker-loci exhibited segregation distortion to the degree expected to occur naturally in only 5% of the markers (χ2 = 16.9, P < 0.001). The elevated level of segregation distortion observed in the SRAP marker-loci could have been the result of a small degree of inadvertent selection having been imposed on the population. Alternatively, it could have been caused by systematic scoring errors at a small number of loci (~14). The SRAP markers are dominant and thus prone to errors caused by failed amplification reactions. Further interrogation of the scoring and mapping data revealed that the 29 loci exhibiting segregation distortion (P < 0.05) were not randomly distributed but were rather clustered on 14 of the 36 linkage group and that only 2 of the distorted loci were not linked to another marker (LOD score <4.0).

The established physical size of the L. maculans genome is 45 Mbp (Rouxel et al. 2011). The 36 linkage groups of the new genetic map span a total of 2,076.4 cM suggesting a physical to genetic ratio of 22 kb per cM. However, the map is incomplete so this initial figure is certainly an over-estimate of the average physical size per cM for the genome. An earlier genetic investigation of L. maculans based on 443 loci produced a map that encompassed only 2,129 cM (Kuhn et al. 2006). Since this earlier map might be expected to represent a more complete coverage of the genome, it is possible that the meiotic divisions that produced the single-spore isolates on which the new map is based experienced an elevated recombination frequency relative to those which gave rise to the earlier population.

The current investigation located the AvrLepR1 locus on linkage group 4 of the new genetic map of the L. maculans genome. With a map composed largely of SRAP markers, it was not possible to use comparative mapping to position AvrLepR1 relative to other avirulence genes. A polymorphic molecular marker specific for AvrLm1 allowed this gene to be positioned on linkage group 13, well removed from the AvrLepR1 locus on linkage group 4. New molecular markers specific for the DNA sequence variants of the AvrLm4-7 gene resident in the two parental isolates, 87-41 and 99-56, allowed the segregation at the AvrLm4-7 locus to be scored in the mapping population and the positioning of the AvrLm4-7 locus relative to the AvrLepR1 locus (Fig. 4). The obvious recombination between AvrLepR1 and AvrLm4-7 demonstrated that they are distinct genes but their close linkage suggests that AvrLepR1 is probably a new member of the L. maculans AvrLm3-4-7-9 genetic cluster, identified previously by Balesdent et al. (2005). The reaction phenotypes of isolates 99-56 (that carries AvrLepR1) and 87-41 (that does not carry AvrLepR1) (Table 3) demonstrated that AvrLepR1 has a distinct specificity from those of AvrLm3, AvrLm4, AvrLm7 and AvrLm9. However, because both isolates were virulent on host plants carrying Rlm3 or Rlm9 (Table 3) it was not possible to map any of the corresponding L. maculans avirulence genes (AvrLm3 or AvrLm9) based of segregation for the virulence/avirulence phenotypes in the new L. maculans mapping population. Following the finding the AvrLm4 and AvrLm7 are two distinct specificities of the same avirulent gene (Parlange et al. 2009), the possibility that AvrLm3 and/or AvrLm9 are allelic variants of AvrLepR1 cannot be dismissed.

Nine L. maculans avirulence genes, each conferring specificity to one or two distinct Brassica resistance genes, have been analyzed genetically (Balesdent et al. 2005; Gout et al. 2006; Parlange et al. 2009). Three of these genes have been cloned; AvrLm1 (Gout et al. 2006), AvrLm6 (Fudal et al. 2007) and AvrLm4-7 (Parlange et al. 2009). We have now defined the reaction phenotype and identified the locus encoding a novel avirulence gene, AvrLepR1, which specifies incompatibility towards the LepR1 resistance gene carried by B. napus line ddm-12-6s-1. The detailed characterization of isolate 87-41 (avrLepR1, AvrLm1) and isolate 99-56 (AvrLepR1, avrLm1) has established this pair of isolates as ideal for surveying B. napus germplasm for the presence of the LepR1 and Rlm1 resistance genes. SRAP markers tightly linked to AvrLepR1 have been identified. These results are the first steps towards cloning AvrLepR1, uncovering its molecular function and defining its molecular interaction with LepR1.