Multidrug-resistance and biofilm formation profiles of Salmonella spp. isolated from raw chicken meat

Available data showed that the recovery of multidrug-resistant and biofilm-producing Salmonella spp. from chicken meat is still scanty in Nigeria. Consequently, this study aimed to characterize the probable multidrugresistant and biofilm producing Salmonella spp. prevalent in chicken meat vended in southern Nigerian markets. About 240 randomly sampled chilled raw chicken meats were collected from open markets in Delta, Edo, Ekiti and Ondo States, and then were analysed for detecting the presence of Salmonella spp.; using rinse centrifugation-plating technique, serological examination and 16S rRNA gene sequencing. The confirmed Salmonella isolates were tested for multidrug-resistance and biofilm formation using Kirby Bauer disc diffusion test and tissue culture plates, respectively. Out of 229 presumptively examined Salmonella isolates, 52 isolates were confirmed as Salmonella spp., while 46 isolates were recorded as multidrug-resistant. The main serotypes recovered were; S. enterica subsp. enterica serovar Typhimurium (35/52; 67.31%), and S. enterica subsp. enterica serovar Enteritidis (17/52; 32.69%). Biofilm characterization of the recovered Salmonella isolates were; strong (OD > 0.240), 6 (11.5 %); moderate (0.120-0.240), 13 (25.0 %); weak (OD < 0.120), 19 (36.5 %), and nonbiofilm producers (OD < 0.120), 14 (26.9 %). This study showed that multidrug-resistant and biofilm-producing Salmonella spp. were prevalent in raw chicken meat; vended within southern Nigerian open markets. Thus, there is an urgent need for relevant regulatory agencies to enforce consumer's safety.


Introduction
Globally, chicken meat is recognized as a good source of animal protein, with low levels of cholesterol, and thus widely preferred over other types of meat (Ogu, et al., 2017a;2017b). Contamination of chicken meat occurs during slaughtering; feather plucking, evisceration, Novel Research in Microbiology Journal, 2021 washing, use of unhygienic processing and storage equipment (Ogu, et al., 2017b). Consumption of such poorly processed and undercooked meat could lead to food-borne illnesses (Antunes et al., 2016;Ogu and Akkinibosun, 2019). Several studies conducted by Eng et al., (2015); Antunes et al., (2016); Mouttotou, et al., (2017) have described chicken products as principal reservoirs of Salmonella spp., which are principle foodborne pathogens.
Salmonella spp. contain more than 2500 known serotypes; with S. enterica serovars Enteritidis and Typhimurium types (the non-typhoidal Salmonella group) being the most frequently implicated serovars in salmonellosis outbreaks (Eng et al., 2015;Heredia and García, 2018). Gastroenteritis is the most clinical feature of Salmonellosis, which is usually self-limiting; but might be invasive and severe, particularly in children, elderly and immunecompromised patients (WHO. 2014;Eng et al., 2015). Treatment with antibiotics is vital for management of severe or invasive human Salmonellosis. Recently, effective treatments of salmonellal infections have been negatively affected by the rising cases of multidrug resistant strains (Okeke et al., 2007;Browne, et al., 2018;Xu et al., 2020). The occurrence of resistance in Salmonella isolates is detrimental to both the veterinary and public health areas, due to the extensive usage of antimicrobials (Eng et al., 2015).
In addition to the worldwide concern of antimicrobial resistance, biofilm formation is a rising challenge. According to several studies of Flemming et al., (2016);Ziech et al., (2016);Farahani, et al., (2018); Abebe, (2020), biofilm is a matrix of extracellular biomolecules, which bind communities of microbes together on the surfaces of living and inanimate substrates, in order to survive unfavorable, hostile or stressed conditions, caused by the use of antibiotics, sanitizers and\or detergents. Clinically, biofilm formation is important because they confer the pathogens extra-tolerance and resistance to antibiotics and host immune system, thereby impeding their ease of eradication, as reported recently by Li and Lee, (2017);Oxaran et al., (2018); Tasneem et al., (2018); Sharma et al., (2019). Biofilms formation has been also reported in poultry farms and meat processing environments (Schonewille et al., 2012;Wang et al., 2013). Currently, different serotypes of S. enterica recovered from clinical and poultry samples have been identified to produce biofilms (Ćwiek et al., 2020).
In spite of the relatively high cost of the local chicken meat and the ban on importation of frozen chicken into Nigeria, the majority of Nigerian citizens still patronize chicken meat sold in open markets without recourse to their wholeness and\or processing conditions (Ogu et al., 2017b). This issue is of a public health relevance considering the possibility of exposure to the resistant strains of Salmonella, as well as biofilm producers.
Therefore, the objectives of this study were to assess the antibiotic resistance profiles and biofilm production potentials of several Salmonella spp.; isolated from raw chicken meat vended within southern Nigerian open markets.

Study area
The studied areas included four major open markets in Southern Nigeria, as presented in Fig. (1

Collection of samples
The sampling regime was carried out between October, 2017 and September, 2020. A total of 240 chicken meat samples were collected from open markets located in Delta, Edo, Ondo and Ekiti State, southern Nigeria, using simple random sampling methods. Portions of the sampled chilled raw chicken Novel Research in Microbiology Journal, 2021 meat used for bacteriological analysis included both the skin and its muscle tissues. The samples were placed in sterile stomacher bags and sealed appropriately. All samples were conveyed to the Microbiological laboratory after collection in black polyethylene bags placed within ice packs (WHO. 2010).

Isolation and phenotypic characterization of Salmonella spp.
Isolation and enumeration of Salmonella spp. from the raw chicken meat were carried out using the meat rinse centrifugation-plating technique, as previously described by Line et al., (2001); Rodrigo et al., (2006);Cox et al., (2014). About 25 portion of each chicken meat sample was cut into small pieces using sterile forceps/scissors, and then placed in a sterile glass container containing 150 ml of sterile 0.1 % buffered peptone water (Becton and Dickinson, USA). The chicken meat was massaged and rotated individually in the sterile container for at least 2 min., to rinse the meat into the peptone water. After that, 25 ml of the rinse was collected in a sterile flask and then centrifuged at 4470 g for 20 min., followed by the removal of 1 ml of the sediment that was used to make serial dilutions up to 10-6. An aliquot of 10 µl of each dilution was spread on the surface of Xylose Lysine Deoxycholate (XLD) agar plates supplemented with Novobiocin (15 mg/ l), using a sterile glass spreader. The inoculated plates were then incubated at 37 o C for Novel Research in Microbiology Journal, 2021 48 h. After incubation, colonies on the Petri plates were counted. The colony counts were used to deduce the presumptive Salmonella counts (PSC); expressed as presumptive Salmonella colony-forming units per ml (cfu/ ml) of the rinse.
The phenotypic techniques employed for the Genus-level identification of the presumptive Salmonella colonies were performed using the standard methods, according to Krieg and Holt, (1984 (Krieg and Holt, 1984;Cheesbrough, 2000).

Serological characterization
Confirmed Salmonella isolates that were identified by the phenotypic tests were subjected to serological examination. The antigenic formula of a pure Salmonella culture was identified using a slide agglutination test, as described previously by Cheesbrough, (2000). One drop of the different Salmonella O and H antisera (Oxoid, UK) was mixed individually with a saline emulsion of the tested pure Salmonella culture on a glass slide for 1 min., followed by observation of agglutination formation under indirect lighting over a dark background. The patterns of agglutination reactions were used to identify Salmonella serotype, by referring to the Kauffmann-White reference scheme (Kauffmann, 1974;Guibourdenche et al., 2010).

Molecular characterization of Salmonella isolates
Species-level identification of the Salmonella isolates was carried out using 16S rRNA gene analysis; followed by polymerase chain reaction (PCR), according to Lane, (1999). Ultrapure DNA templates were extracted from the tested Salmonella isolates using the Zymo-Spin column, as prescribed by the manufacturer (Zymo Research Corporation, Irvine, CA, USA). Universal 16S rRNA bacterial primers mainly; 27F 5'AGAGTTTGATCCTGGCTCAG';1492R 5'GGTTACCTTGTTACGACTT3'; 1466 base pair (Lane, 1999), often employed for bacterial taxonomy were used to determine the presence of 16S rRNA gene in the Salmonella isolates. S. enterica subsp. enterica serovar Typhimurium ATCC 14028 was used as a positive control strain for the PCR. DNA sequencing of the PCR products was performed using the Dideoxy chain termination method, in reference to Sanger et al., (1977). The PCR products were cleaned up with ExoSAP-IT (ThermoFisher Scientific, Waltham, MA), and then subjected to cycle sequencing with the Big Dye Terminator version 3.1 (Applied Biosystems) using the standard cycling conditions, followed by quality checking and proofreading with Sequencher version 4.10.1 (Gene Codes Corporation, Ann Arbor, MI). Comparison of the experimentally derived nucleotide sequences (query sequences) against the reference sequence database (rRNA_typestrains/prokaryotic_16S_ribosomal_RNA) was carried out with BLASTN 2.8.0+ program (National Centre for Biotechnology Information [NCBI]), to confirm the species level of the Salmonella isolates.

Estimation of the multiple antibiotic resistance indices (MAR)
The multiple antibiotic resistance indices (MAR) of the Salmonella isolates were determined according to the method prescribed by Krumperman, (1983). Using the following equation, obtaining a MAR value greater than 0.2 indicated a high-risk source of acquiring multidrug-resistant Salmonella from the tested samples

MAR = ∑
Where; MAR is the mean multiple antibiotic resistance indices, while AR is the antibiotic resistance scores of each Salmonella isolate, which is defined as the sum of antibiotic classes to which a particular Salmonella isolate exhibited resistance. A is the total number of antibiotic classes tested, and B is the total count of Salmonella isolates examined.

Detection of biofilm production
In vitro characterization of biofilm production by the Salmonella isolates was carried out using the Tissue culture plates (TCP) methods, described recently by Beshiru et al., (2018);Orjih et al., (2021) with slight modification. The Salmonella isolates were cultured individually overnight in Trypton soy broth (Merck, Darmstadt, Germany) at 37 o C before adjusting the cell suspension to cell density similar to 0.5 McFarland standards. An aliquot of 200 μl of each standardized isolate cell's was dispensed into each well of 96 well tissue culture plate. Sterile broth medium was used as negative control, while S. enterica subsp. enterica serovar Typhimurium ATCC 14028; included in separate well, served as a positive control strain. The tissue culture plate was incubated aerobically at 37 °C for 24 h. Thereafter, the wells were carefully tapped to remove the contents before washing thrice with 250 μl of sterile phosphate buffer saline (pH 7.2), to completely release the free-hanging bacterial cells. Sodium acetate (2 %) was used to fix the biofilms produced by the adherent cells before applying 0.2 % crystal violet stain. Excess stain was carefully rinsed off using deionized water, and then the dye incorporated by the adherent cells was solubilized by adding 200 µl of 33 % glacial acetic acid (Merck, Darmstadt, Germany) (Orjih et al., 2021). A Microplate reader (Molecular Devices San Jose, CA) set at wave length 570 nm was employed to read the optical density (OD) of each well. This was carried out in triplicates. Absorbance was determined by subtracting the OD 570 of the control from that of the test assays OD 570 , with a mean value determined for each isolate. Results of biofilm formation obtained were rated as OD 570 values; (< 0.120) as non-biofilm producers, (0.120-0.240) as moderate biofilm producers, and (> 0.240) as strong biofilm producers (Orjih et al., 2021).

Statistical analysis
Descriptive statistics of Salmonella counts and prevalence datasets were performed using NCSS version 12 data analysis software. Biofilm characterization was assayed using descriptive statistics and expressed as OD 570 values. Moreover, NCSS ver. 12 data analysis through the Shapiro-Wilk normality test and Fisher (F) one-way ANOVA test were used for normally distributed datasets. The test of hypothesis was considered statistically significant if the achieved level of significance (p) was less than 0.05.

Isolation and prevalence of Salmonella spp.
The prevalence and counts of presumptive Salmonella spp. recovered from the raw chicken meat samples are presented in Table (1). Overall, the prevalence of presumptive Salmonella is 22.71 %, while the mean counts are estimated as 6.19 ± 6.42 log 10 . Shapiro-Wilk test showed that the datasets of counts of presumptive Salmonella are normally distributed (p= 0.79; α= 0.05). Based on results of the normality test, the parametric Fisher one-way analysis of variance (ANOVA) tests within each of the datasets indicated no significant difference (p= 0.72) for presumptive Salmonella. The ANOVA test between the Salmonella datasets also indicated no significant difference (p = 0.54). The counts reported in this study are higher than those reported by previous work of Briehta-Harhay et al. (2008), who worked with chicken meat collected from the United States.

Phenotypic and molecular identification of the Salmonella isolates
Results of phenotypic identification of all the recovered Salmonella isolates showed that they were; Gram-negative, motile, short rods, catalase (+), urease (-), coagulase (-), citrate (-), indole (-), red slope/yellow butt reaction in triple sugar iron (alkaline slope/acid butt), and produced enough H 2 S. These characteristics agreed with the expected standard results of Genus Salmonella (Cheesbrough, 2000).
The 16S rRNA gene sequencing analysis showed that S. enterica is the main species detected currently in the chicken meat samples, as indicated in Fig. (2, 3). PCR amplifications yielded products of 1466 bp for the selected isolates. This is the anticipated base pair (bp) size of the samples recorded positive for Novel Research in Microbiology Journal, 2021 Salmonella, according to the Genus-specific PCR reaction applied in this study, in reference to Lane, (1999). Out of the 229 presumptive Salmonella isolates examined, 52 isolates are confirmed as Salmonella spp. Similarly, previous studies conducted by Akbar and Kumar, (2013); Pedro et al., (2016); Ugwu et al., (2019) also detected S. enterica in the raw chicken meat samples that they examined. However, Cretu et al., (2009) reported that Salmonella was absent in poultry products collected from Sweden. This was largely attributed to the strict compliance of chicken meat processors and poultry breeders to the regulatory programs set by the Swedish government.  The antibiotic resistance patterns of the Salmonella spp. recovered from the chicken meat samples are presented in Table (2). Of the 52 Salmonella isolates tested, 46 Salmonella isolates are recorded to be multidrug-resistant. The multidrugresistant Salmonella isolates are most prevalent in raw chicken meat samples vended in Edo State, and are least prevalent in Ekiti State. Overall, according to the recorded diameters of inhibition zones, the Salmonella isolates are most resistant to Ampicillin (96.15%), but are more sensitive to Gentamycin (40.39%). Amongst the tested Salmonella isolates, the MAR ranged from 0.69 to 0.87. Generally, MAR is estimated at 0.83. These MAR values detected in the raw chicken meat samples exceeded the recommended limit of 0.2. This indicates that raw chicken meat samples collected from Southern Nigeria are potential sources of multidrug-resistant Salmonella spp., with a probable significant health risk. Similarly, Antunes et al., , Ugwu et al., (2019);Parvin et al., (2020) have also detected multidrug-resistant Salmonella in chicken meat samples. Thus, poultry products are currently identified as a public health concern. In accordance, previous study of Antunes et al., (2016) reported that the association of multidrug-resistant Salmonella with chicken meat is extremely disturbing, due to the probable resistance of Salmonella spp. to multiple arrays of antibiotics that are clinically relevant.

Biofilm formation
Biofilm characterization of the 52 strains revealed that 73.1 % of the isolates produced biofilm detected using the tissue culture plate techniques. The production rate is classified: as strong biofilm producers (OD > 0.240), 6 (11.5 %); moderate (0.120-0.240), 13 (25.0 %), weak and non-biofilm producers (OD < 0.120), 19 (36.5 %) and 14 (26.9 %); respectively, as demonstrated in Table (3). Similarly, previous studies showed varying rates of biofilm productions; 98.3 % (Oliveira et al., 2014), 14.8 % (Laviniki et al., 2015), 100 % (Ziech et al., 2016) and 72.7 % (Sereno et al., 2017), were reported from poultry products, pig feed mills and processing materials in Brazil, respectively. Moreover, Portuguese S. enterica serovars yielded 100 % biofilms formation (Seixas et al., 2014). The Novel Research in Microbiology Journal, 2021 differences in biofilm production by Salmonella species was recently attributed to a number of factors, mainly; species diversity, differences in incubation temperature, growth medium and dynamism of the environment (Bashiru et al., 2018). Findings from this study supported a previous report that biofilm production in poultry products is on the rise and this represent a threat to the public health (Burmølle et al., 2010;Tasneem et al., 2018), due to the clinical implications of the biofilm-mediated antimicrobial resistance.

Conclusion
This study revealed that multidrug-resistant Salmonella spp. were prevalent in raw chicken meat vended within southern Nigerian open markets. Extensive exploitation of antibiotics in poultry farms could probably be the key factor for this. Moreover, this study characterized remarkably high percentages of biofilm production by the multidrug-resistant-Salmonella strains, thus posing a serious public health concern of the probable transmission of multidrug-resistant biofilm-producing Salmonella strains to the consumers of chicken meat. Accordingly, there is an urgent need for vigorous enlightenment campaigns and sanctions against antibiotics abuse in the poultry sector, coupled with appropriate monitoring and implementations of requisite food safety policy.