Xlt 4

A second 10 g aliquot from each package of meat was sterilely inoculated into 90 mL of buffered peptone water (BPW; Becton Dickinson) and incubated overnight at 37 °C. We transferred a 100 µL aliquot of the meat/BPW homogenate to 10 mL of Rappaport-Vassilidis R10 (RV; Becton Dickinson) broth that was then incubated overnight at 42 °C. RV broth was subsequently inoculated onto xylose-lysine-Tergitol 4 agar (XLT-4; Becton Dickinson) for the differentiation of Salmonella suspect isolates. Characteristic black colonies on XLT-4 agar were tested for lactose fermentation on MacConkey agar and agglutination with polyvalent antisera O. Lactose negative, agglutinating isolates were speciated using MALDI-TOF, and confirmed Salmonella underwent whole genome sequencing (MiSeq, Illumina, San Diego, CA). Reads were assembled using the SPades assembler version 3.9 and post-processed with MisMatch corrector available online from the Center of Genomic Epidemiology (CGE)^{52 (link)}. We confirmed Salmonella genus and species, identified Salmonella serotype, and determined acquired antimicrobial resistance genotype using the KmerFinder, SeqSero, and ResFinder online databases available at CGE^{53 (link)–55 (link)}.

Following the FDA BAM method for Salmonella detection (Andrews et al., 2016 ), chili powder and aged cilantro samples were pre-enriched in a non-selective broth for 24 h (24-h samples), and then aliquots were transferred to selective RV and TT broths for a secondary 24 h enrichment (48-h samples). Ice cream samples were enriched for 48 h following the FDA BAM method for Listeria but without acriflavin, cycloheximide, and nalidixic acid (Hitchins et al., 2016 ). For all foods, a 10 μL aliquot was streaked onto XLT4 (Becton, Dickinson and Company, Sparks, MD, United States) agar plates at two time points, 24 and 48 h, and plates were incubated for 24 h at 35°C. Suspect black colonies observed on XLT4 agar plates were confirmed as Salmonella using the VITEK 2 system (BioMérieux, France). Only samples with confirmed Salmonella were considered culture-positive. Microbiome cell pellets were also collected at each 0, 24, and 48-h time point, centrifuged at 7,100 × g for 30 min, and stored at -20°C until extracted from cell pellets. For chili powder, metagenomic samples were first vigorously mixed, passaged through Miracloth filter membrane (EMD Millipore, Billerica, MA, United States), and then sedimented at 200 × g for 3 min, to remove chili powder from bacterial cells, prior to harvesting the cell pellets by centrifugation, as described above.

After preparation, the samples were centrifuged at 200× g for 5 min at 4 °C. One loopful of the supernatant (0.01 mL) was taken and streaked onto to selective agar, namely MacConkey (pink/red colour; Becton Dickinson, Bergen, NJ, USA), XLT4 (black; Becton Dickinson, Bergen, NJ, USA), Baird–Parker (grey/black; Oxoid, Basingstoke, UK) and Kenner KF (purple; Becton Dickinson, Bergen, NJ, USA) agar to identify E. coli, Salmonella, S. aureus and Enterococcus, respectively. For quality control of each batch, all of the selective media were incubated at 38 °C for 24 h. As positive controls, the four bacteria were directly applied to the selective agar. The isolation and identification of pathogenic colonies were performed according to each manufacturer’s instruction manual.
Twenty isolated colonies from each selective agar were streaked onto the blood agar at 38 °C and incubated for 24 h. For Gram-positive bacteria, the incubation time was 24–36 h (FAO regional antimicrobial resistance monitoring and surveillance guidelines, 2019). The isolated colonies from the blood agar were taken for MALDI-TOF-MS analysis (Bruker, Germany) at the Bacteriology Laboratory of the Kimron Institute for confirmation of their identities [44 (link)].

Thirty grams of litter from each sample were weighed out and combined with 270 mL of buffered peptone water (BPW; BD, Franklin Lakes, NJ) in a sterile Whirl-Pak® bag (Fig 1); sample weights were not adjusted according to moisture content. Each bag was vigorously hand-shaken/massaged for one min. Bags were transferred to a shaking incubator set to 37°C and 50 rpm. After 24 h incubation, 1 mL was transferred from the Whirl-Pak® bag (without filter) into 9 mL of BPW, which was returned to the shaking incubator for an additional 20 h at 37°C and 50 RPM. After 20 h, 100 μL was transferred into tubes containing 10 mL Rappaport-Vassiliadis (RV; BD, Franklin Lakes, NJ) broth and 1000 μL was transferred into 10 mL Tetrathionate (TT; BD, Franklin Lakes, NJ) broth. Tubes were incubated at 42°C for 48 h (RV) or 24 h (TT). After incubation, 10 μL broth was streaked onto a Xylose Lysine Tergitol-4 (XLT-4; BD, Franklin Lakes, NJ) plate and incubated at 37°C for 24–48 h. Up to four presumptive positive colonies (black, or red with a black center) were re-streaked onto individual XLT-4 plates and incubated an additional 24–48 h prior to PCR confirmation.