Draft Genome Sequence of the Animal and Human Pathogen Malassezia pachydermatis Strain CBS 1879

Abstract

Malassezia pachydermatis is a basidiomycetous yeast that causes infections in humans and animals. Here, we report the genome sequence of Malassezia pachydermatis strain CBS 1879, which will facilitate the study of mechanisms underlying pathogenicity of the only non-lipid-dependent Malasezzia species.

GENOME ANNOUNCEMENT

M. pachydermatis is the only non-lipid-dependent species of the genus Malasezzia. All other 13 species (1) of this genus are obligate lipophilic and require fatty acids for growth. This is due to the lack of a fungal type fatty acid synthase (2). M. pachydermatis is able to assimilate fatty acids from the growth medium and can thus be considered a facultative lipophilic species (3). The molecular mechanisms underpinning this behavior are not yet clear.

M. pachydermatis is a member of the microbiota of animals. It is an opportunistic pathogen of dogs causing dermatitis and otitis externa. M. pachydermatis has also been implicated in human bloodstream infections (4, 5). Three genomes of the obligate lipophilic species M. globosa, M. restricta, and M. sympodiales have been reported (2, 6). Our goal is to understand the facultative lipophilic nature of M. pachydermatis.

M. pachydermatis genomic DNA was extracted as previously described (7). The DNA was sequenced with the Illumina HiSeq 2000 platform at ServiceXS (Leiden, the Netherlands). Two runs with 120-bp paired-end reads on 250-bp fragments were performed following standard Illumina protocols with a 280-fold genome coverage. Reads were quality controlled with FastQC (8) and trimmed using Flexbar (9). De novo assembly was performed using CLC Assembly Cell (CLC bio, Denmark). The resulting contigs were scaffolded using SSPACE_Basic (10), and gaps were filled with GapFiller (11). The final assembly consisted of 148 contigs that were linked by pair-end reads into 91 scaffolds, 28 of which were longer than 1 kb. The maximum contig and scaffold length were 1,466,538 and 1,489,072 bp, respectively, and the N₅₀ was 0.64 Mbp and 1.3 Mbp, respectively. The genome size was 8.15 Mbp with a G+C content of 55.17%.

The genome was annotated using Maker2 (12) and we made use of a set of 109,264 previously reported Ustilaginomycotina proteins, 1,413 ESTs from Malassezia spp., and CEGMA (13). The homology-based predictor GeneMark and the ab-initio predictors SNAP (14) and Augustus (15) were used to predict genes. In order to train Augustus and SNAP we ran MAKER two consecutive times; the initial annotation output from MAKER was converted into a model for SNAP and a training set for Augustus, which was used in the subsequent run. Functional annotation of the predicted genes was performed by Blast2GO (16), which involved Blast and InterProScan annotation (17, 18).

A total of 4,202 protein-coding genes were predicted with an average size of 1,581 bp. The coding regions corresponded to 81% of the genome. In addition, CEGMA showed that 97.18% of the eukaryotic core genome was present in the genome (13).

Lipid degrading enzymes play an important role in the host invasion process of M. pachydermatis (19, 20). A total of 50 lipid degrading enzymes were identified in the genome, including 35 lipases and 15 esterases. Most interestingly, a typical fungal fatty acid synthase was not detected in the genome. Instead a polyketide synthase, homologous to fatty acid synthases (21), was detected that showed 75% identity with its bidirectional homologue of M. sympodialis. How M. pachydermatis is able to grow in the absence of fatty acids is a subject for future research.

Nucleotide sequence accession numbers.

This whole-genome shotgun project has been deposited at DDBJ/EMBL/GenBank under the accession number LGAV00000000. The version described in this paper is LGAV01000000.