Identification of novel effectors from Albugo candida to reveal new components of plant defence
Final Report Abstract
Albugo sp. are obligate biotroph, filamentous, oomycetes pathogens. They infect a broad range of hosts including important economic Brassicacea species like Brassica rapa, B. oleracea and B. juncea. A remarkable feature of Albugo sp. is their capacity to impose on host plants susceptibility not only to Albugo, but also to other parasites to which the host is resistant in the absence of Albugo infection (Cooper et al. 2002). These observations indicate that Albugo has powerful mechanisms for suppression of host defenses. Within the genus Albugo two major clades can be distinguished: A. candida and A. laibachii clade. Since the A. laibachii clade contains mainly species selfing and therefore exhibits less heterozygosity, the A. laibachii isolate Nc14 was sequenced as a reference genome first. 6.6 Gbp was generated using the Illumina sequencing platform and a pipeline was established to improve genome assemblies using short read data only. Therefore VELVET was used as a primary assembler and MINIMUS as a secondary meta assembler. 35.8Mbp of an estimated 38Mbp were assembled. The discrepancy between estimated genome size and assembled genome could be shown to be due to repetitive regions, which were not resolved. This is reflected by a high coverage of short contigs. Analysing the repeat content within the resolved genome revealed about 18% of the assembly to be repeats with DNA- and LTR-transposons being the most frequent. Based on the telomeric repeats the chromosome number could be estimated 11 to 12. 20% of repeats resolved within the genome shows the advantage of using the established assembly pipeline. 8.4 Gbp was generated from normalized cDNA libraries of different stages of infection. Aligning these reads to the assembly shows that ~50% of the genome is covered at least 2x. Only very few reads match to the repetitive part of the genome showing that major coding regions are within the longer, non repetitive contigs and are therefore resolved. Gene content and continuity of the genome was first assessed using CEGMA. In terms of core eukaryotic genes, 93.6% of a selected set of 248 could be detected. For group 2, 3 and 4 of the core eukaryotic genes, all genes predicted where present in full length, none of the genes was split over contigs, further validating the quality of the assembly. For gene predictions a pipeline was set up combining trained-, ab-initio- and consensus-gene predictions. 13,398 genes could be predicted. Testing these gene models against a set of annotated core eukaryotic orthologous groups indicate that 75% of these groups are present in the current annotation. For comparison, 78% of KOGs were present in P. infestans. Functional annotations were performed using ASGARD in combination with manual annotation. 902 gene models could be functionally annotated. Surprising was that Nc14 has possibly all major enzymes involved in brassinosteroid biosyntheses or major enzymes to synthesize lipopolysaccharides. The major goal of the project was to identify effectors capable of suppressing defence. We therefore studied the secretome in greater depth. Genetically identified oomycete avirulence (AVR) proteins have been described to be secreted proteins that have signal peptide and RXLR motifs. In most published oomycete genomes so far, the RxLR motif is over-represented and positionally constrained in their secretome (Win et al. 2006). 547 secreted proteins without a transmembrane domain were detected including 19 RXLRs and 16 RXLQs. Shuffling experiments indicated, that proteins carrying these motifs were not over-represented in A. laibachii and are therefore unlikely to be the major class of effectors. Analysing the CRN class of effectors known from P. infestans revealed only two homologs and no family expansion. To identify new classes of effectors in the Albugo clade, we selected all secreted proteins either showing heterozygosity or showing divergence between Nc14 and Alem1, a closely related A. laibachii isolate. Using MEME and Motif search we identified a new class of potential effectors carrying a “CHXC” motif. 32 CHXC candidates could be identified, showing significant overrepresentation of the motif after permutation tests. Using the permutation test it could be shown that CHXC over-representation is restricted to the Albuginales. Since this unique appearance of a new effector class in a single clade wasn’t described before it is necessary to study how this class phylogenetically evolved. Phylogenetic analyses revealed, that CHXCs are a unique class only distantly related to an A. thaliana CHXC identified as an allinase homolog. It is therefore attempting to speculate that CHXCs might be the cause of a horizontal gene transfer between host and pathogen that has to be shown in future studies. In addition to the informatics approach, using the P.capsici heterologous test system could be used to validate that the N-terminal domain of CHXCs is capable of enabling transfer into the host cell. It was verified in addition that CHXCs could enhance virulence when delivered from P. syringae via the Type III secretion system. In summary, using a short read assembly approach, the genome of A.laibachii Nc14 could be assembled and annotated establishing new advanced pipelines. The quality of the genome and its annotation was carefully assessed. The high quality enabled to identify a new class of effectors unique to Albuginacea. These results open up a new field in the research of effector evolution and effector function.
Publications
- (2009) A draft genome sequence of Pseudomonas syringae pv. tomato T1 reveals a type III effector repertoire significantly divergent from that of Pseudomonas syringae pv. tomato DC3000. Mol Plant Microbe Interact. Jan;22(1):52-62
Almeida NF, Yan S, Lindeberg M, Studholme DJ, Schneider DJ, Condon B, Liu H, Viana CJ, Warren A, Evans C, Kemen E, Maclean D, Angot A, Martin GB, Jones JD, Collmer A, Setubal JC, Vinatzer BA
- (2009) De novo assembly of the Pseudomonas syringae pv. syringae B728a genome using Illumina/Solexa short sequence reads. FEMS Microbiol Lett. 291(1):103-11
Farrer RA, Kemen E, Jones JD, Studholme DJ
