Description

Background: Leptosphaeria maculans is an ascomycete phytopathogen of Brassica napus (oilseed rape, canola). The first genome sequence of this fungus (JN3), obtained in the 2000s using Sanger sequencing showed an unusual compartmentalization between AT-rich and GC-equilibrated blocks termed isochores. The location of effector genes in the plastic, AT-rich compartments of the genome was instrumental in establishing the "two-speed fungal genome" paradigm. According to this paradigm, phytopathogen genomes encompass a plastic compartment enriched in genes involved in niche adaptation. Similar genome organisation was later on found to be a commonality in numerous filamentous phytopathogens, but its identification was slowed down by the use of short-reads sequencing that tended to misrepresent the dispensable genome in the assemblies. Thus the genome sequencing of L. biglobosa, a species related to L. maculans resulted in a poor-quality assembly with complete absence of a putative dispensable genome.

Findings: JN3 genome assembly was improved and reduced to 33 scaffolds with a scaffold N50 of 2.4Mb. L. maculans NZ-T4 and L. biglobosa G12-14 assemblies were generated de novo using Oxford Nanopore MinION reads. Their scaffold N50 were of 383 kb and 462 kb respectively. Their genome annotation was supported by integrating conserved proteins and RNA sequencing from Leptosphaeria-infected samples. It resulted in 13,047, 14,026 and 12,678 predicted protein-coding genes for JN3, NZ-T4 and G12-14, respectively. Genome completeness was estimated using BUSCO and reached 98%, 96% and 96%, respectively.

Conclusion: The newly produced high-quality assemblies and annotations of those three Leptosphaeria genomes will allow further studies, notably focused on the interaction between Leptosphaeria maculans and canola. The discovery of as yet unknown effectors will notably allow progress in B. napus breeding towards L. maculans resistance.