Genome Sequence of Bacillus glycinifermentans TH008, Isolated from Ohio Soil.

The genome sequence of an Ohio soil isolate, TH008, was determined. The sequence reveals a close relationship between TH008 and domesticated Bacillus glycinifermentans strains found in a traditional Korean fermented soybean food.

GENOME ANNOUNCEMENT

Recently, there has been a burgeoning interest in traditional fermented foods for their potential health and other societal benefits (1, 2). One such food is the Korean fermented soybean paste, cheonggukjang. Two spore-forming bacterial isolates from this food were found to comprise a novel species denoted Bacillus glycinifermentans (3). Interestingly, a survey of Ohio soil bacteria capable of growing at 55°C in NaCl 5% (wt/vol) revealed a relatively high frequency of isolates that clustered with B. glycinifermentans in multilocus sequence typing analysis (not shown). To determine whether these isolates could be considered “wild” relatives of domesticated B. glycinifermentans strains, one typical isolate, strain TH008, was selected for analysis. TH008 had been isolated from a moist soil sample collected on the banks of the Olentangy River in Columbus, Ohio (40.0024 N, 83.0219 W).

To determine the genome sequence for TH008, an Illumina paired-end library (2 × 150 bp) was sequenced. Prior to assembly, 4 million read pairs were selected and sequence adapter trimming was performed followed by digital normalization (4). De novo assembly was performed on the resulting normalized paired-end reads with the Velvet assembler (5), using a minimum contig length cutoff of 1,000 bp and k-mer length 77. The assembly had 111 contigs with coverage 39.0; the N50 was 97,587, and the largest contig was 225,240, and the total size was 4.459 Mb. Genome annotation was performed by the NCBI Prokaryotic Genome Annotation Pipeline (http://www.ncbi.nlm.nih.gov/genome/annotation_prok/). Clusters predicted to encode secondary metabolites were identified by antiSMASH 3.0 (6). A total of 4,522 genes were predicted, including 4,294 coding sequences (CDS), 241 pseudo genes, 41 tRNAs, 5 rRNA genes, and 114 frameshifted genes. The average DNA G+C content was 46.1%.

In silico DNA-DNA hybridization (7) with formula 2 of the GGDH 2.0 online application (http://ggdc.dsmz.de/distcalc2.php) estimated the hybridization of TH008 with the known cheonggukjang isolates of B. glycinifermentans, GO-13T (GenBank LECW01) and KJ-17 (GenBank LECV01) as 67.60% ± 2.91 and 67.20% ± 2.91, respectively. This result indicates a high likelihood (73.57% and 72.63%, respectively) that TH008 shares a common species identity with these isolates. Ohio soils therefore appear to contain bacteria closely related to flora found in a Korean fermented food.

Nucleotide sequence accession numbers.

This whole-genome shotgun project has been deposited in DDBJ/ENA/GenBank under the accession no. JZBT00000000. The version described in this paper is the first version, JZBT01000000.