The race structure of Leptosphaeria maculans (phoma stem canker) in the United Kingdom

Phoma stem canker (Leptosphaeria maculans) is the main disease problem on winter oilseed rape (Brassica napus) in the UK, causing losses of > £100M p.a., despite use of fungicides. Two types of resistance to L. maculans have been identified in B. napus. These are major resistance (R) gene-mediated qualitative resistance that operates in cotyledons and leaves in autumn (Balesdent et al., 2001) and quantitative resistance that operates in leaf stalk and stem tissues, after initial leaf infection until harvest in summer (Huang et al., 2009). R gene-mediated resistance to L. maculans is associated with a gene-for-gene interaction in which the product of a pathogen effector (Avr) gene is recognised by the product of a host R gene so that the pathogen is unable to infect the host. Therefore, R gene-mediated resistance to L. maculans is effective in protecting plants only if the corresponding avirulent allele is predominant in the local L. maculans population (Balesdent et al., 2001; Rouxel et al., 2003). In Brassica spp., 15 R genes (Rlm1-Rlm10, RlmS, LepR1-LepR3, BLRM1-2) that confer resistance against L. maculans have been identified. It is important to monitor the races in L. maculans populations since this monitoring can provide information to guide deployment of cultivars with different R genes to increase their durability.

Winter oilseed rape field experiments at eight different sites in the UK with the cultivar Drakkar (no R gene, used as trap for all pathogen races) were used to sample the races of L. maculans at those sites. Leaves with phoma leaf spots were collected in autumn/winter 2010/2011 and single pycnidial isolates were obtained. From each site, 15 to 30 isolates were collected, with 296 isolates in total. To determine the race of each of the isolates, a set of differential cultivars/lines with known R genes was used for cotyledon inoculation tests. To date, 57 isolates have been characterised by cotyledon inoculation. The results show that the avirulent allele of AvrLm7 is predominant at all the sites sampled, with a mean frequency of 100%. This indicates that the corresponding resistance gene Rlm7 is effective. The frequencies of AvrLm5 (98%) and AvrLm6 (93%) avirulent alleles were greater than those of AvrLm4 (53%) and AvrLm1 (23%). The populations of L. maculans were 100% virulent at AvrLm2, AvrLm3 and AvrLm9 loci.
There were differences between sites in frequency of Avr alleles. For example, frequencies of AvrLm4 in Norfolk and Shropshire were 100% and 90%, respectively, which were greater than those at other locations (mean frequency 30% at Rothwell, Stockbridge and Cowlinge).
In addition, single pycnidial isolates of L. maculans were obtained from cultivars with quantitative resistance (with no known R gene) (Es-Astrid and NK Grandia). The frequencies of Avr alleles were investigated by cotyledon inoculation. This was done to investigate whether there are effects of host background (quantitative) resistance on selection for Avr alleles. It was found that there are some host selection effects; for example the frequency of the avirulent allele AvrLm7 was 100% in populations obtained from Drakkar but 70% in populations obtained from cultivars with quantitative resistance. For further analysis, the isolates were cultured to harvest mycelium for DNA extraction and molecular characterisation.