Complete Genome Sequence of Pragia fontium 24613, an Environmental Bacterium from the Family Enterobacteriaceae

Abstract

The complete genome sequence of Pragia fontium 24613 was determined using PacBio RSII, Roche 454, and SOLiD sequencing. A total of 3,579 genes were predicted, including 3,338 protein-coding sequences and 146 pseudogenes. This is the first whole-genome sequence of a strain belonging to the environmental genera of the family Enterobacteriaceae.

GENOME ANNOUNCEMENT

Pragia fontium is a Gram-negative, mesophilic, rod-shaped, motile bacterium from the family Enterobacteriaceae. While one isolate originated from a stool sample from a healthy woman, all other isolates came from water wells or water pipes. The genus Pragia contains only one species, P. fontium, which was first described in Czechoslovakia in 1988 (1). Strain P. fontium 24613 was isolated from a water pipe in 1983 (1). P. fontium produces H₂S and oxidizes gluconate, which distinguishes this species from other enterobacterial H₂S producers. The genus Pragia is one of the few genera from the family Enterobacteriaceae that is isolated almost exclusively from environmental samples.

The total DNA genome of P. fontium 24613 was sequenced using PacBio RSII (GATC Biotech, Inc., Constance, Germany), Roche 454 (Roche Genome Sequencer FLX; Eurofins Genomics, Inc., Ebersburg, Germany), and SOLiD V3Plus (SeqOmics, Inc., Mórahalom, Hungary) platforms. PacBio single-molecule real-time (SMRT) Analysis version 2.3 was used for PacBio raw read treatment (covering ≈300 Mbp). HGAP software (2) was used for de novo genome assembly, with ≈30× coverage of self-corrected reads with length >4,746 bp. Contig accuracy was enhanced with the Quiver tool using the entire read set (70× coverage). The contigs were ordered according to an optical map (OpGen, Inc., Gaithersburg, MD, USA), which compared the AflII restriction pattern of the P. fontium chromosome to the in silico-obtained contigs. Contig overlaps were manually trimmed using the Geneious software (3), and the remaining gaps were filled with Sanger sequencing of the PCR products. Roche 454 reads (11× coverage) and SOLiD reads (50× coverage) were used for increasing accuracy and final corrections of the whole-genome sequence. Ori-Finder (4) was used for oriC detection, and the genome sequence was numbered from oriC. Annotation was performed with the NCBI Prokaryotic Genome Annotation Pipeline (http://www.ncbi.nlm.nih.gov/genome/annotation_prok/). Open reading frames were predicted using GeneMarkS+ and ProSplign. Ribosomal RNAs, tRNAs, and small noncoding RNAs (ncRNAs) were identified using BLASTn, tRNAscan-SE, and Cmsearch.

P. fontium 24613 was found to contain only chromosomal DNA; no plasmids were detected. The whole-genome sequence of P. fontium 24613 comprised 4,094,629 bases, and the G+C content was 45.4%. In total, 3,579 genes were predicted, including 3,338 protein-coding sequences (CDSs) and 146 pseudogenes. Twenty-two rRNA and 72 tRNA genes were also identified in the genome sequence. The genome size was only slightly smaller than the genome sizes of closely related strains of the genera Yersinia and Pectobacterium; the G+C content was quite similar as well. Several genes predicted in the Pragia genome are similar to genes detected in plant pathogens and growth-promoting rhizobacteria (5). The genome of the Pragia strain will be analyzed in detail and compared to other Enterobacteriaceae genomes.

Nucleotide sequence accession number.

The P. fontium 24613 whole-genome sequence has been deposited in GenBank under the accession no. CP010423.