Draft Genome Sequence of the Bacterium Comamonas aquatica CJG.

A Gram-negative bacterial strain, Comamonas aquatica CJG, absorbs low-density lipoprotein but not high-density lipoprotein in serum. Here, we report its draft genomic sequence of 3,764,434 bp, containing total 3,425 genes, 27% of which encode proteins for metabolism and energy conversion, and it is 30% identical to the genome of Comamonas testosteroni.

GENOME ANNOUNCEMENT

Through our effort to examine possible pathogens in water sources, a Gram-negative bacterial strain, Comamonas aquatica CJG, was isolated from Daguan River in Kuming, China, in 1963 (1, 2). C. aquatica is like a slightly curved rod, with two to eight flagella. Its liquid culture grows at 37°C with light-red color and a foul smell. It is resistant to several antibiotics, including penicillin, cephalosporin, and bacitracin, but it is highly sensitive to chloramphenicol and tetracycline (1). Interestingly, C. aquatica can absorb serum low-density lipoprotein (LDL) and very low-density lipoprotein (VLDL) but not high-density lipoprotein (HDL) (1, 2).

This strain was originally named Pseudomonas sp. 16Zu, since it was very close to the genus Pseudomonas biochemically and morphologically. Based on 16S rRNA gene sequence, we found that this strain is in fact quite different from Pseudomonas but rather is 99% identical to a lesser-known species, Comamonas aquatica (3, 4). It is also 96% identical to Comamonas testosteroni (5, 6). We therefore renamed this strain Comamonas aquatica CJG.

We then sequenced its whole genome by Illumina HiSeq 2000 using its DNA sample prepared by a commercial bacterial genomic extraction kit. To build libraries, the genomic DNA was sonicated with Covaris or Bioruptor to produce fragments of <500 bp in size, filled up and extended by Klenow DNA polymerase, and ligated to T-vector. Sequencing of this library with inserts of 150 to 500 bp produced 781 Mb paired-end reads (~200-fold coverage), which were filtered by the NGS QC toolkit (7) to generate clean data of 700 Mb. The clean data combined with another data set generated similarly using a library of 2,000 bp (350 Mb) were assembled in SOAPdenovo (8). The final assembly has 136 scaffolds, with an N50 length of 284,677 bp and maximum length of 669,486 bp. The genome was annotated using the databases of COG, Swiss-Prot, KEGG, and TrEMBL (9–12).

The draft genome of the C. aquatica CJG strain is 3,764,434 bp. The G+C content is 64.4%. A total 3,425 genes of about 87% of the genome are annotated, encoding 3,125 proteins. Those responsible for metabolism and energy conversion account for 27% of the total genes, and about 470 genes are related to replication, transcription, translation, and trafficking. The bacterial strain also harbors 111 genes, functioning in protein posttranslational folding, modification, and homeostasis. About 145 genes encode signal transduction components and virulence factors, including EnvZ, KdpE, ComE, and CheA. Interestingly, the strain contains 209 functional unknown proteins. Whether these contribute to LDL absorbance remains to be determined.

A comparison of the C. aquatica CJG genome with the nucleotide database of all other bacterial genomes in GenBank using BLAST (13) indicated that the genome shows the highest similarity (E value ≤E-5) to Comamonas testosteroni (99 out of 320 contigs). The second best match is Delftia acidovorans (47 out of 320 contigs). Several other Acidovorax and Alicycliphilus strains also have some similarity (12 to 14 out of 320 contigs). All these genera belong to the Comamonadaceae family.

Accession number(s).

This genome has been deposited in GenBank/ENA/DDBJ under accession number CP016603.