Isolation, antimicrobial resistance and virulence characterization of Salmonella spp. from fresh foods in retail markets in Hangzhou, China

Abstract

Salmonella can cause severe foodborne diseases. This study investigated the prevalence of Salmonella spp. in fresh foods in Hangzhou market and their harborage of antibiotic resistance and virulence genes, antibiotic susceptibility, and pathogenicity. A total of 500 samples (pork, n = 140; chicken, n = 128; vegetable, n = 232) were collected over a one-year period. Salmonella was found in 4.2% (21) of samples with the detection rate in pork, chicken and vegetables as 4.3% (6), 6.3% (8), and 3% (7), respectively. One Salmonella strain was recovered from each positive sample. The isolates were identified as six serotypes, of which S. Enteritidis (n = 7) and S. Typhimurium (n = 6) were the most predominant serotypes. The majority of isolates showed resistance to tetracycline (85.7%) and/or ciprofloxacin (71.4%). Tetracycline resistance genes showed the highest prevalence (90.5%). The occurrence of resistance genes for β-lactams (blaTEM-1, 66.7%; and blaSHV, 9.5%) and aminoglycosides (aadA1, 47.6%; Aac(3)-Ia, 19%) was higher than sulfonamides (sul1, 42.9%) and quinolones (parC, 38.1%). The virulence gene fimA was detected in 57.1% of isolates. Gene co-occurrence analysis implied that resistance genes were associated with virulence genes. Furthermore, selected S. Typhimurium isolates (n = 4) carrying different resistance and virulence genes up-regulated the secretions of cytokines IL-6 and IL-8 by Caco-2 cells in different degrees, suggesting that virulence genes may play a role in inflammatory transcription. In in vivo virulence test, microbiological counts in mouse feces and tissues showed that all included S. Typhimurium were able to infect mice, with one strain showing significantly higher virulence than others. In conclusion, this study indicates Salmonella contamination in fresh foods in Hangzhou market poses a risk to public health and it should be closely monitored to prevent and control foodborne diseases.

Fig 1. Serotype distribution of the Salmonella isolates.

a) Fresh foods. b) Seasons.

Fig 2. The detection rate of resistance genes.

Error bars correspond to the standard deviation of the means and letters indicate statistically significant differences between groups.

Fig 3. Co-occurrence gene network of ARG subtypes and VGs in MDR isolates collected.

9 different colors represent 5 kinds of antibiotics and 4 virulence genes. Nodes belonging to the same kinds of VGs or resistant to the same class of antibiotics are presented in the same color. A connection represents a significant (P<0.05) correlation. The size of each node is proportional to the number of connections.

Fig 4. Percentage of invasion of Caco-2 cells by Salmonella strain.

Error bars correspond to the standard deviation of the means and letters indicate statistically significant differences between groups.

Fig 5. Expression of IL-6 and IL-8 mRNA was determined by real-time qPCR and normalized to GAPDH using the 2^-ΔΔCt method.

Bars represent the mean of triplicates. Error bars correspond to the standard deviation of the means and letters indicate statistically significant differences between groups.

Fig 6. Body weight observation of mice infected with Salmonella strains.

Data represent the mean of 6 mice per group. The means of weight in each group were statistically different (P<0.05).

Fig 7. Microbiological analysis of feces and tissues.

a) Colony counts in feces of infected mice (log₁₀CFU/g). The mean number of feces in each group was statistically different (P< 0.05). b) Colony counts in tissues of infected mice (log₁₀CFU/g). Means in each tissue were statistically different (P< 0.05).