Complete Genome Sequence of the Arcobacter canalis Type Strain LMG 29148.

Arcobacter canalis was originally recovered from shellfish and from a sewage-contaminated canal. Arcobacter canalis is closely related to the marine bacterium Arcobacter marinus This study describes the complete whole-genome sequence of the A. canalis type strain LMG 29148 (=F138-33T; =CECT 8984T), which was recovered from oysters.

ANNOUNCEMENT

Arcobacter canalis was originally recovered from shellfish and sewage-contaminated canal water in Catalonia, Spain (1). Based on phylogenetic and genomic analyses (1), A. canalis was determined to be highly related to Arcobacter marinus (2). The original description of A. canalis also identified several phenotypic discriminatory markers to distinguish it from A. marinus and ostensibly from other Arcobacter spp. (1). These included, for example, nitrate reduction, catalase activity, and growth on medium containing 2% NaCl. As part of a project to obtain complete genomes for all Arcobacter type strains, we sequenced an A. canalis type strain. In this study, we report the first closed genome sequence of the A. canalis type strain LMG 29148 (=F138-33T; =CECT 8984T), which was recovered from oysters.

A. canalis strain LMG 29148T was obtained from the BCCM/LMG culture collection and grown, both initially and in one subculture, aerobically at 30°C for 48 h on brain heart infusion agar (Thermo Fisher Scientific, Waltham, MA) amended with 5% horse blood and 2% (wt/vol) NaCl. Approximately 5 μl of cells, representing multiple individual colonies, were removed from the plate using a sterile inoculating loop. Genomic DNA was prepared from these cells using the Promega Wizard genomic DNA purification kit (Madison, WI); a single preparation of genomic DNA was used to construct the Illumina and PacBio libraries. The Illumina MiSeq library was constructed using the Illumina Nextera DNA Flex kit, and the 20-kb PacBio SMRTbell library was prepared using the SMRTbell template prep kit 1.0, following the manufacturer’s instructions. Illumina sequencing was performed on a MiSeq instrument at 8.0 pM, with dual-index paired-end reads, using the MiSeq reagent kit v2 (300 cycle). PacBio sequencing was performed on an RS II sequencer, with reads assembled using the Hierarchical Genome Assembly Process (HGAP) v. 3.0 in the SMRT Analysis software v. 2.3.0. Default parameters were used for all software unless otherwise specified. Sequencing metrics are presented in Table 1. A single PacBio contig was obtained and processed using Geneious Prime v. 2019.1.3 (Biomatters Ltd., Auckland, New Zealand), as follows. The PacBio contig was circularized manually within Geneious, thus removing >99.9% of the bases with a Q score of <40 (generally those within ∼8 kb of the contig ends). Then, the Illumina MiSeq reads were quality trimmed and assembled onto the circularized PacBio contig. Using the “Find Variations/SNPs” module with a default minimum variation of 0.3, 25 variations (1-bp indels) were identified and corrected to the MiSeq consensus sequence. The final coverage was 357×.

TABLE 1

Sequencing metrics and genomic data for A. canalis strain LMG 29148T

Featurea	Value(s)b
Sequencing metrics
Illumina MiSeq platform
No. of reads	1,764,438
No. of bases	260,767,753
Avg length (bases)	148
Coverage (×)	92.2
PacBio platform
No. of reads	96,414
No. of bases	749,500,177
Avg length (bases)	7,773.8
Coverage (×)	264.9
Genomic data
Chromosome
Size (bp)	2,829,476
G+C content (%)	27.49
No. of CDSc	2,679
Assigned function (% CDS)	1,017 (38.0)
General function annotation (% CDS)	1,025 (38.3)
Domain/family annotation only (% CDS)	182 (6.8)
Hypothetical (% CDS)	455 (17.0)
No. of pseudogenes	23
Genomic islands/CRISPR
No. of genetic islands	5
No. of CDS in genetic islands	175
CRISPR/Cas locus type	I-B
Gene content/pathways
Signal transduction
Che proteins	che(A)2(B)2CD(R)2V(W)2(Y)4
No. of methyl-accepting chemotaxis proteins	33
No. of response regulators	57
No. of histidine kinases	67
No. of response regulator/histidine kinase fusions	4
No. of diguanylate cyclases	23
No. of diguanylate phosphodiesterases (HD-GYP, EAL)	9 [1]
No. of diguanylate cyclase/phosphodiesterases	11
No. of other signal transduction genes	12 [1]
Motility
Flagellin genes	fla1–fla7
Restriction/modification
No. of type I systems (hsd)	1
No. of type II systems	0
No. of type III systems	2
Transcription/translation
No. of transcriptional regulatory proteins	60 [1]
Non-ECF σ factors	σ70
No. of ECF σ factors	0
No. of tRNAs	63
No. of ribosomal loci	6
Nitrate/nitrite reduction	napABDFGH, nirA, nrfAH
Nitrogen fixation (nif)	No
Osmoprotection	(BCCT)5,d ectABC
Pyruvate → acetyl-CoA
Pyruvate dehydrogenase (E1/E2/E3)	Yes
Pyruvate/ferredoxin oxidoreductase	por
Urease	No
Vitamin B12 biosynthesis	Yes

CDS, coding sequences; ECF, extracytoplasmic function; CoA, coenzyme A.

Numbers in square brackets indicate pseudogenes or fragments.

Numbers do not include pseudogenes.

BCCT, betaine/carnitine/choline transporter.

Genomic data for strain LMG 29148T are presented in Table 1. The genome size is 2,829,476 bp, with a G+C content of 27.5%. Protein-, rRNA-, and tRNA-encoding genes were identified using GeneMark, RNAmmer, and ARAGORN (3–5), respectively, and annotated as described previously (6). The data presented here support the taxonomic relatedness of A. canalis and A. marinus; the genomes of these species are highly syntenic, with an average nucleotide identity of 95.2%. However, we identified errors in the original phenotypic discriminatory markers for both species (1, 2). Although A. canalis was originally described as being unable to reduce nitrate (1), the A. canalis genome contains a full set of nitrate reductase genes (Table 1); controlled tests performed in our laboratory (7) demonstrate that strain LMG 29148T is able to reduce nitrate, thus confirming the genomic data. Similarly, we have demonstrated that A. marinus strain JCM 15502T is catalase positive, contradicting previous descriptions (1, 2), and its genome encodes the catalase gene. Additionally, we routinely grow A. canalis on medium without amended NaCl (0.5% [wt/vol final]), thus disputing the original description (1). These discrepancies suggest that a complete reassessment of the phenotypic markers used to discriminate A. canalis and A. marinus should be performed.

Data availability.

The complete genome sequence of A. canalis strain LMG 29148T has been deposited in GenBank under the accession number CP042812. All MiSeq and PacBio sequencing reads have been deposited in the NCBI Sequence Read Archive (SRA) under accession number SRP217059.