Draft Genome Sequence of Mycobacterium houstonense Strain ATCC 49403^T

Abstract

Mycobacterium houstonense is a nontuberculous species rarely responsible for human infection. The draft genome of M. houstonense ATCC 49403^T comprises 6,451,020 bp, exhibiting a 66.96% G+C content, 5,881 protein-coding genes, and 65 predicted RNA genes.

GENOME ANNOUNCEMENT

Refined taxonomic evaluation of the Mycobacterium fortuitum third biovariant complex led to the delineation of Mycobacterium houstonense as a new species (1). M. houstonense is represented by only two isolates from human sources in the United States, including one face wound isolate from Houston, TX, hence the name given to this species (1). Accordingly, the sources of human infection remain unknown, but sequences closely related to those of M. houstonense have been detected in consumed fishes (2). Moreover, M. houstonense is one of the nontuberculous Mycobacterium species containing the erm gene, which supports resistance to macrolides (3). It is therefore of medical and general interest to further describe the genome of this species, and we performed whole-genome sequencing of the M. houstonense ATCC 49403^T strain. Genomic DNA was isolated from M. houstonense grown in MGIT Middlebrook liquid culture (Becton, Dickinson, Le Pont-de-Claix, France) at 37°C in a 5% CO₂ atmosphere. M. houstonense genomic DNA was then sequenced in 2 Illumina MiSeq runs (Illumina, Inc., San Diego, CA) using a 6.9-kb mate-paired library. Reads from Illumina were trimmed using Trimmomatic (4) and assembled using Velvet (version 1.2.03) (5). Contigs were combined together by SSPACE version 2 (6), Opera version 2 (7) helped by GapFiller version 1.10 (8), and homemade tools in Python to refine the set. Finally, the draft genome of M. houstonense strain ATCC 49403^T consists of 27 scaffolds and 197 contigs containing 6,451,020 bp. The G+C content of this genome is 66.96%. Noncoding genes and miscellaneous features were predicted using RNAmmer (9), ARAGORN (10), Rfam (11), PFAM (12), and Infernal (13). Coding DNA sequences (CDSs) were predicted using Prodigal (14), and functional annotation was achieved using BLAST+ (15) and HMMER3 (16) against the UniProtKB database (17). The genome was shown to contain at least 5,946 predicted RNAs, including 6 rRNAs (2 genes are 5S rRNA, 3 genes are 16S rRNA, and 1 gene is 23S rRNA) and 59 tRNAs. A total of 5,881 identified genes yielded a coding capacity of 5,222,064 bp (coding percentage, 80.94%). Among these genes, 4,813 (81.84%) were found to be putative proteins, and 1,068 (18.16%) were assigned as hypothetical proteins. Moreover, 3,338 genes matched a least one sequence in the Clusters of Orthologous Groups database (18, 19) with BLASTP default parameters. In silico DNA-DNA hybridization (DDH) (20) was performed with 23 reference genomes selected on the basis of their 16S rRNA gene proximity with M. houstonense. The M. houstonense genome was locally aligned 2-by-2 using the BLAT algorithm (21, 22) against each one of the 23 selected genomes, and DDH values were estimated from a generalized linear model (23). The DDH (value, ≥25%) was 29.9% (±2.40%) for Mycobacterium fortuitum CT6, 29.7% (±2.45%) for Mycobacterium nonchromogenicum, and 24.8% (±2.40%) for Mycobacterium mageritense.

Nucleotide sequence accession number.

The M. houstonense ATCC 49403^T strain genome sequence has been deposited at EMBL under the accession no. FJVO00000000.