Complete Genome Sequence of the Proteolytic Clostridium botulinum Type A5 (B3′) Strain H04402 065

Abstract

H04402 065 is one of a very small group of strains of proteolytic Clostridium botulinum that form type A5 neurotoxin. Here, we report the complete 3.9-Mb genome sequence and annotation of strain H04402 065, which was isolated from a botulism patient in the United Kingdom in 2004.

TEXT

Proteolytic Clostridium botulinum neurotoxin causes food-borne, infant, and wound botulism. Three types of neurotoxin (A, B, and F) are formed by this organism, but type A presents the greatest bioterrorism threat (10). Five subtypes are known (A1 to A5), but to date, only genomes of strains forming subtypes A1 to A4 are published. Strains of proteolytic C. botulinum isolated from 4 of 40 wound botulism cases from the United Kingdom in 2004 (1) were closely related by whole-genome analysis, and each carried an identical subtype A5 gene with a standard ha neurotoxin gene cluster, plus a truncated type B3 neurotoxin gene (4, 5). The same arrangement was seen in a strain of proteolytic C. botulinum from a Californian infant botulism case (7). Here, we report the fully assembled complete genome sequence and annotation of C. botulinum type A5 (B3′) strain H04402 065.

Genomic DNA was sequenced using Roche 454 and Illumina GA2 platforms. The former generated a 61.85-Mb sequence (16× coverage) with contigs assembled using Newbler. Illumina sequencing generated 4,709 Mb (1,068× coverage) with paired-end lane-generated contigs assembled with ABySS (13). Close proteolytic C. botulinum genome synteny (11) enabled contig mapping to strains Kyoto (NC_012563) and Langeland (NC_009699). Misassemblies were recognized by dot matrix comparison (DNAMAN version 5.1.5; Lynnon Corporation). Gaps were closed by sequencing PCR products from the same DNA. Roche 454 reads, sensitive to homopolymeric tracts, introduced nearly 1,500 sequence ambiguities. These were corrected by comparison with Illumina contigs, which are unaffected by these tracts. Protein-coding regions were predicted using Glimmer (6) and GeneMark (3) with manual curation using Artemis (12). Automatic annotation using RAST, preserving gene calls (2), was complemented with manual annotation of interesting regions highlighted by comparisons with other C. botulinum genomes by using Artemis, including InterPro domains, TMHMM, and SigP analyses.

C. botulinum strain H04402 065 has a circular chromosomal genome of 3,919,740 bp with a 28.2% G+C content and no plasmids. Totals of 3,719 coding sequences, 72 tRNA genes, and 9 complete rRNA loci were identified. The coding density was 0.94 genes/kb, with an average gene length of 854 bp. Double reciprocal orthologue plots identified strains Kyoto (type A2) and CDC657 (type Ba4) as close relatives, but H04402 065 shows synteny with all proteolytic C. botulinum genomes.

Chromosomal neurotoxin gene clusters are found at one of three sites (9), and that of strain H04402 065 resides in the oppA-brnQ operon, as with some other type A and B strains (9). No other neurotoxin cluster genes were found. Strain H04402 065 contains only three complete spore germinant receptor operons, as with other type A strains (11). Comparative genomics of proteolytic C. botulinum showed that the flagellar glycosylation island (FGI) is a genetically heterogeneous part of the genome (5). Interestingly, that of strain H04402 065 includes five genes with 90% identity to Clostridium tetani aminophosphonate metabolic pathway genes CTC1698, CTC1699, CTC1700, CTC1704, and CTC1705 (8).

Nucleotide sequence accession number.

The complete genome sequence of strain H04402 065 has been deposited in EMBL/GenBank under accession number FR773526.