Genetic Analysis of Non-Hydrogen Sulfide-Producing Salmonella enterica Serovar Typhimurium and S. enterica Serovar Infantis Isolates in Japan

TEXT

Salmonella enterica subsp. enterica serovar Typhimurium and S. enterica serovar Infantis of the genus Salmonella cause acute enterocolitis. Generally, H₂S production is considered to be indicative of the presence of pathogenic food-borne Salmonella serovars and is an important index for screening and confirmation of Salmonella using various agars, including DHL (deoxycholate hydrogen sulfide lactose), SS (Salmonella-Shigella), and TSI (triple sugar iron) agars.

We isolated non-H₂S-producing S. Typhimurium and S. Infantis strains from retail chicken meats in Gunma prefecture, Japan, in 2010. Forty-seven Salmonella spp. isolates were obtained from 95 poultry meat samples (49.5%) between June and December 2010. Four (8.5%) were non-H₂S-producing Salmonella (S. Typhimurium isolates GST-106, GST-108, and GST-204; S. Infantis isolate GSI-9) isolates. H₂S production and lysine decarboxylase testing of Salmonella on TSI and LIM (lysine indole motility) agar was subsequently conducted (see Fig. S1 in the supplemental material). The other isolates found to be H₂S producing were as follows: S. enterica serovar Enteritidis (1 strain), S. enterica serovar Yovokome (1 isolate), S. enterica serovar Livingstone (1 isolate), S. enterica serovar Schwarzengrund (3 isolates), S. enterica serovar Manhattan (3 isolates), S. Infantis (38 isolates), and S. Typhimurium (2 isolates).

Pulsed-field gel electrophoresis (PFGE) analysis (1) suggested that the profile of the S. Infantis and S. Typhimurium strains were similar but not identical among these serovars (see Fig. S2 in the supplemental material). Multilocus sequence typing (MLST) for S. enterica (http://mlst.ucc.ie/mlst/dbs/Senterica/) suggested that the S. Typhimurium strains belonged to ST328, while the S. Infantis strains belonged to ST32 (see Table S1 in the supplemental material).

To elucidate the genetic alteration responsible for H₂S production, Illumina GAIIx 81-mer paired-end short read sequencing of S. Infantis and S. Typhimurium isolates was performed (see Table S1 in the supplemental material), followed by short-read mapping (2) against the corresponding reference genome sequences of S. Infantis SIN (downloaded from the Wellcome Trust Sanger Institute [ftp://ftp.sanger.ac.uk/pub/pathogens/Salmonella/SIN.dbs]) and S. Typhimurium T000240 (3), respectively. Whole-genome sequencing suggested that some nucleotide variations were present between H₂S-producing and non-H₂S-producing isolates (see Tables S2 and S3 in the supplemental material). Notably, the phsA gene, encoding the thiosulfate reductase precursor, was terminated in the coding sequence by an alteration to the stop codon (nonsense mutation) in both S. Infantis (GSI-9) and S. Typhimurium (GST-106, GST-108 and GST-204) (Fig. 1; also, see Tables S2 and S3 in the supplemental material). The phs genes are essential for the dissimilatory anaerobic reduction of thiosulfate to H₂S in S. Typhimurium (4, 5), suggesting that the disruption of the phsA gene could be involved in H₂S production. Intriguingly, a nonsense mutation in the phsA gene was also identified in low-H₂S-producing S. enterica serovar Paratyphi A (AKU_12601 and ATCC 9150) (6); thus, comparative genomic analysis also supports our findings.

• Open in new tab
• Download powerpoint

Fig 1

Schematic representation of the nonsense mutation in the phsA gene encoding the thiosulfate reductase subunit. Nucleotide variations and corresponding amino acid mutations are shown below the gene organization schemes.

H₂S is highly toxic to mammalian cells, including those of humans, and the cecal mucosa detoxifies H₂S by converting it to thiosulfate (S₂O₃⁻) (7). Luminal inflammation generates reactive oxygen species, leading to the conversion of thiosulfate to tetrathionate (S₄O₆²⁻). S. enterica utilizes tetrathionate as an electron acceptor in anaerobic respiration using TtrABC and TtrRS located on a SPI2 pathogenicity island (8). It is possible that non-H₂S-producing phenotypes generate more thiosulfate, leading to enough substrate for TtrABCRS anaerobic respiration in Salmonella spp. Indeed, recent studies have suggested that TtrABCRS anaerobic respiration based on tetrathionate contributes to competitiveness with other microbiota in intestinal inflammation (9, 10, 11), implying that reduced H₂S production might lead to increased TtrABCRS anaerobic respiration that promotes growth and colonization in the gut.

Furthermore, the non-H₂S-producing S. Infantis and S. Typhimurium isolates carried CMY-2 β-lactamase (Table 1) and demonstrated reduced susceptibility to cefazolin (data not shown). Aminoglycoside and trimethoprim resistance genes were also identified (Table 1). A plasmid carrying the bla_CMY-2 gene was detected by S1 nuclease–PFGE and Southern blot analysis (12), indicating that it appears to have been an independent acquisition by two different serovars (Fig. 2). The CMY enzyme is the enzyme most frequently detected in E. coli, Klebsiella pneumoniae, and Salmonella spp. (13), and poultry meat is most frequently contaminated with CMY-2-producing Salmonella (14). Indeed, CMY-2-producing S. Infantis and S. Typhimurium isolates have been found in retail chicken and cattle in Japan (15, 16). Because of the clinical importance of third- and fourth-generation cephalosporin in human and veterinary medicine, it is of further concern that insufficient testing might lead to antimicrobial-resistant Salmonella isolates being missed.

• Open in new tab
• Download powerpoint

Fig 2

PFGE-Southern blot hybridization for bla_CMY-2 detection. (A) S1 nuclease-PFGE analysis of the Salmonella isolates. (B) Southern blot obtained using a bla_CMY-2-specific DNA probe. M, lambda ladder marker for PFGE.

In conclusion, we isolated non-H₂S-producing S. enterica isolates. Whole-genome analysis suggested that this lack of H₂S production may be associated with a phsA gene mutation and that these strains may be disseminated in poultry. Although we did not examine the pathogenicity of these strains in experimental animals, this defect may be associated with energy production and gut colonization. In addition, such strains may slip unnoticed past screening methods, resulting in incorrect identification of S. enterica, including antimicrobial resistant isolates. Comprehensive testing will be required to detect unique isolates and to map the overall epidemiology of S. enterica.