MALDI-TOF MS-based identification and antibiotics profiling of Salmonella species isolated from retail chilled chicken in Saudi Arabia

2.1 Sample collection

A total of 112 chilled whole chicken carcass were procured from five local poultry companies at their retail outlets in Jeddah, Saudi Arabia. Each sample of a chicken carcass was put in a sterile plastic bag that was marked with the source and the date of collection. Collected samples were delivered in iceboxes immediately to the Microbiology Laboratory at the Department of Biological Sciences, King Abdulaziz University. After that, the samples were stored at 4 °C for future analysis.

2.2 Sample preparation and enrichment

A 25 g meat sample from each chicken carcass was put in a sterile stomacher bag in accordance with ISO 6579:2002 regulations. Thereafter, 225 ml of 2% buffered peptone water (Difco, Becton & Dickinson, MD, USA) was added to form a 1:10 dilution. The sample was then homogenised for 3 min at 2,000 rpm using a Stomacher 400 homogenizer (Seward Medical, England, UK). Following that, 10 ml of Rappaport-Vassiliadis-soya broth (RVS; Oxoid Ltd., UK, code: CM0866) was added to 1 ml of the pre-enriched sample, which was then incubated for 24 h at 41.5 °C. Thereafter, a 0.1 ml aliquot of the pre-enriched sample was added to 10 ml of Muller-Kauffmann Tetrathionate-Novobiocin Broth (MKTTn; Oxoid Ltd., UK, code: CM1048) and incubated for 24 h at 37 °C.

2.3 Isolation and characterization of Salmonella

10 µL aliquots of each prepared enriched sample were streaked onto Xylose Lysine Deoxycholate Agar (XLD; Oxoid Ltd., UK) and Brilliant Green Agar (BGA; Oxoid, Ltd., UK) plates and incubated for 24 h at 37 °C. On BGA plates, salmonella colonies showed up as pinkish-white or red colonies with a red halo, and pink-red colonies with black centres on XLD plates. Individual representative colonies were picked up and sub-cultured until similar colonies were gained. From each plate, presumptive Salmonella colonies were chosen, and inoculated on nutrient agar, and cultivated for 24 h at 37 °C overnight. Gram's stain was used to evaluate the staining characteristics of the isolates and primary biochemical tests were carried out to identify the isolates at the genus level. Thereafter, each Salmonella isolate was preserved in 50% glycerol at 80 °C for further examination (El-Tayeb et al., 2017).

2.4 MALDI-TOF biotyper identification

Using matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF MS; Bruker company Run identifier: 210221–1204-1011016777), presumptive Salmonella isolates were further identified at the species level (Dieckmann and Malorny, 2011). In this assay, individual presumptive Salmonella colonies were spread onto stainless steel MALDI plate, having Biotyper matrix solution. A pulsed laser then irradiates the loaded plate, causing desorption and extirpation of the sample and matrix material. In the hot column of the extracted gases, the molecules of the analyte are ionized to become deprotonated or protonated of ablated gases, and then they can be accelerated into the mass spectrometer for analysis (Dieckmann et al., 2008). The MALDI Biotyper CA System software was used to process the spectral data using the default settings. The smoothing, normalization, threshold exclusion, and peak selection were performed by the software, forming a list of a spectrum's most important peaks. The reference peak lists in the MALDI Biotyper database were compared to the peak lists produced from the MALDI-TOF mass spectra.

The final results were articulated as arithmetical score values between 0 and 3.00. An organism with a higher log (score) value has a higher similarity to an organism in the reference FDA-cleared database. The log (score) ≥ 2.00 is considered to be an excellent probability for the identification of a specific test organism at the species level (Singhal et al., 2015).

2.5 Test for antibiotic sensitivity

The test for the antibiotic sensitivity of the Salmonella isolates to conventional antibiotics was performed using an automated MicroScan WalkAway plus System with Gram-negative bacteria cards (Server version: 4.1.70 (PYTH) 48 2016–10-26_15-05–35). The interpretation of the results was as intermediate, resistant, or susceptible according to the breakpoints for each antibiotic.

2.6 The assay of the extended-spectrum beta-lactamases (ESBLs) production

A combined disc test was performed to investigate the ESBL‐producing species for isolates that displayed a zone of inhibition of ≤ 22 mm, ≤ 25 mm, and ≤ 21 mm for ceftazidime, ceftriaxone, and aztreonam, respectively (Korzeniewska and Harnisz, 2013). The test was conducted as per CLSI guidelines (Wayane, 2016). The antimicrobials used were ceftazidime (30 μg), ceftazidime/clavulanic acid (30/10 μg), cefotaxime (30 μg), cefotaxime/clavulanic acid (30/10 μg), aztreonam (30 μg) and aztreonam/clavulanic acid (30/10 μg). CLSI criteria were used to interpret the results (Wayane 2016). A 5 mm increase in the zone of inhibition for combined drugs to ceftazidime, cefotaxime, or aztreonam was an indicator of ESBL‐producing species (Wayane 2016; Korzeniewska and Harnisz, 2013).

2.7 Computation of multiple antibiotic resistance (MAR) index

The multiple antibiotic resistance index (MARI) was calculated by dividing the number of antibiotics to which the isolate was resistant by the total number of antibiotics to which the isolate had been exposed (Apun et al., 2008). MARI ≥ 0.4 is associated with human fecal sources of contamination. MARI greater than 0.2 implies the origin of the isolates is most likely from areas where antibiotics are frequently used, while MARI ≤ 0.2 implies the origin of the bacteria is from areas wherever antibiotics are less frequently consumed (Thenmozhi et al., 2014).

2.8 Data management and statistical analysis

MS Excel was used for the recording of data and designing the graphs. The organized data was subsequently examined using IBM SPSS version 25.0. By dividing the number of positive samples by the total number of samples analyzed, the prevalence was estimated. To calculate the percentage of susceptible (S), intermediate (I), or resistant (R) strains, frequency and percentile descriptive statistics were utilized. A p-value of < 0.05 was regarded as a value of statistically significant.

3.1 Assessment of Salmonella prevalence in retail chicken

Out of the collected 112 chicken meat samples, only 35 samples (31.3%) were positive for Salmonella based on the conventional identification via biochemical features. The Biotyper MALDI-TOF MS technology was used to further identify the isolates at the species level (Table 1). Out of the 35 Salmonella isolates submitted for MALDI-TOF MS, thirty four isolates had score values ≥ 2.0 and one isolate (sample no. 16) had 1.94 score. According to the MALDI-TOF-MS identification test, the 35 isolates were identified as Salmonella spp., S. Enteritidis or S. Typhimurium at high confidence levels (a log (score) value between 2.00 and 3.00), while one isolate was characterized as Salmonella sp. at a low confidence level (log (score) < 2.00).

The prevalence of Salmonella isolates varied across the five different poultry companies. The highest obtained Salmonella isolates were to company number 5 (n = 15, 42.9%) and the prevalence of each isolate was found to be 5.7%, 20%, and 17.1%, for Salmonella spp., S. Enteritidis, and S. Typhimurium, respectively (P < 0.05) (Table 2). On the contrary, out of all samples collected from company number 2, only 1 (2.9%) sample was Salmonella species, which is the lowest among all (Table 2).

3.2 Antimicrobial sensitivity test

The previous 35 Salmonella isolates were evaluated for antibiotic susceptibility against a cohort of 18 different antibiotics from eight distinct classes (Table 3). Levofloxacin, ciprofloxacin, and all tested carbapenems (meropenem, imipenem, and ertapenem) were effective against every isolate. The maximum percentages of resistance (65.7%) was found for cefotaxime, and ampicillin and followed by ceftazidime (62.9%). Sixteen (45.7%) isolates were resistant to clavulanic Acid-Amoxicillin 4 (11.4%) isolates were resistant to ampicillin-subaclam, indicating that they were possible ESBL producers.

3.3 Assessment of resistance profile of the isolated Salmonella species

Among the 35 Salmonella isolates subjected for sensitivity test, 23 (65.7%) isolates have shown resistance for three or more than three antibiotics belonging to different categories. Among these, 1 (2.9%), 3 (8.6%), 10 (28.6%), 6 (17.1%), and 3 (8.6%) isolates were resistant for three, four, five, eight and nine antibiotics, respectively (Table 4). In this regard, three isolates have shown resistance for nine antibiotics which is the highest pattern reported in this study. Of all the tested antibiotics, none of the isolates have shown resistance to carbapenems. The antibiotic resistance pattern indicated that some of the isolates showed similar resistance patterns as indicated in Table 4. Out of the tested Salmonella spp., eight species showed similar resistance patterns for five antibiotics (AMOX, AMP, CTX, CTZ, and MXF) (MARI = 0.28) and only one species displayed resistance to three antibiotics (MARI = 0.16). Similarly, three species displayed identical resistance patterns for four antibiotics (AMOX, AMP, CTX, and CTZ) (MARI = 0.22) and two isolates exhibited the same patterns for eight (MARI = 0.44) and nine (MARI = 0.5) antibiotics as presented in Table 4.

Among the eight classes of antibiotics tested, the highest number of resistances were developed to cephalosporins (n = 69) including cefuroxime (n = 9), ceftazidime (n = 21), cefotaxime (n = 23), cefepime (n = 7), and cefazolin (n = 9). In contrast, the lowest resistance was encountered for glycylcycline class of antibiotic (n = 1) (Fig. 1). All in all, 20 isolates were resistant to β-lactamase inhibitor combinations, 11 to folate pathway inhibitors, 23 to penicillin, 12 to fluoroquinolones.

Close

Fig. 1

The number of isolates showed resistance to five classes of antibiotics.

Export to PPT

3.4 ESBLs production assay

Nine (25.7%) and eight (22.9%) isolates were found to be ESBL producers for ceftazidime, cefoxaxime and aztreonam, respectively. All these ESBL producers showed resistance to fourth-generation cephalosporin (cefepime) (Table 5). However, these isolates were susceptible to combinations of β-lactam/βlactamase inhibitors (amoxicillin-clavulanic acid and ampicillin-sulbactam).