Genome Sequence Analysis of Mycoplasma sp. HU2014, Isolated from Tissue Culture

Abstract

The draft genome sequence of a novel Mycoplasma strain, designated Mycoplasma sp. HU2014, has been determined. The genome comprises 1,084,927 nucleotides and was obtained from a mycoplasma-infected culture of chicken DT40 cells. Phylogenetic analysis places this taxon in a group comprising the closely related species Mycoplasma yeatsii and Mycoplasma cottewii.

GENOME ANNOUNCEMENT

Mycoplasmas, members of the wall-less Mollicutes (1), include multiple species that are important pathogens of humans and animals (2). In addition, several species are notorious contaminants of tissue culture (3), necessitating rigorous testing for inadvertent growth that may confound biomedical research or adversely impact biomanufacturing (4). Commonly reported estimates of contamination rates range from 15 to 35%, although incidences of 65 to 80% have been reported for these parasitic blights (5). Consistent with such values, a recent study revealed a high incidence of mycoplasma-derived reads in the NCBI mammalian RNA-seq archive (6).

As part of our ongoing mutagenesis studies in the chicken DT40 bursal lymphoma cell-line (7), whole-genome sequencing disclosed an inadvertent Mycoplasma coculture with a DT40 clone. The strain, designated Mycoplasma sp. HU2014, was almost identical to Mycoplasma yeatsii and Mycoplasma cottewii based on 16S rRNA analysis. These serologically distinct species (8) have almost identical 16S rRNA sequences, differing at 4 single nucleotide polymorphisms (9). Contamination with M. yeatsii, but not M. cottewii, has been reported once (10), and so is extremely rare in comparison with other mycoplasmas. Despite M. yeatsii and M. cottewii being considered caprine commensals or opportunists (11), the most likely source of the contamination was fetal bovine serum.

To gain further insight into this taxon, Illumina 150-bp paired-end reads were assembled de novo (12) and screened by Mega BLASTn to retrieve contigs that matched the complete genomes of M. yeatsii or Mycoplasma putrefaciens. The resulting 75 contigs were auto-annotated (13) and then manually curated based on the M. yeatsii GM274B genome features (14). The final data set comprised 1,084,927 nucleotides (25.4% GC; >6,800-fold average coverage depth), which is larger than that of M. yeatsii GM274B (895 kb), indicating that most, if not all, of the genome had been retrieved. A total of 1,080 genes were annotated (including partial genes at contig termini), including 910 open reading frames, 30 tRNAs, and 3 small RNAs.

Based on the available sequence data and published molecular typing schemes, it was not possible to unambiguously assign HU2014 to a known species. Both 16S rRNA-based and multilocus sequence typing indicate very close relationship between M. cottewii and M. yeatsii (15). The 5 multilocus sequence type targets from HU2014 exhibited 93 to 98% identity to their respective orthologs from these two species. For rpoB and lepA, the M. cottewii and M. yeatsii sequences were more similar to each other (98% identity) than either was to HU2014 (95 to 96%), but this pattern did not comport for the other targets. Further comparative analysis is warranted to delineate the fine structure of this cluster of related taxa.

These data provide evidence for a distinct evolutionary history for the HU2014 lineage and suggest future taxonomic refinement to accommodate the spectrum of sequence variation encompassed within this clade. As HU2014 is cytotoxic to DT40 cells, it will be of interest to determine the molecular attributes responsible for pathogenesis. These data also serve as a further precautionary note on the difficulties associated with mycoplasma prevention in eukaryotic culture systems.

Nucleotide sequence accession number.

This whole-genome shotgun project has been deposited at DDBJ/EMBL/GenBank under accession number LFYS00000000.