Assessment of the Presence of Resistance Genes Detected from the Environment and Selected Food Products in Benin

Gram-negative bacilli can spread from the environment and through food products. This study aimed to characterize ESBL production and virulence genes from multidrug-resistant Gram-negative bacilli isolated from specimen collected from the environment, kitchen, and food products. A total of 130 samples were collected at local markets in seven different communities in Benin (Abomey-Calavi, Ouidah, Bohicon, Abomey, Parakou, Djougou, and Grand-Popo). Samples were cultured on McConkey and ChromID™ ESBL agar plates. The isolates were identified by the API 20E gallery. An antibiotic susceptibility test was carried out, and the detection of ESBL production and virulence-associated genes was carried out by Polymerase Chain Reaction (PCR). The data collected was coded and analyzed using GraphPad prism 7 software and Excel. The software R was used to calculate the correlation coefficient between the results of the detection of ESBL+ on agar and by the effect of the double synergy. The results showed that sixty-three (63) bacterial strains were isolated from the 130 samples, of which the dominant species was Chryseomonas luteola (10/63). The kitchen samples were the most contaminated with 36.50%. More than 40% of the isolates were resistant to at least three different classes of antibiotics. Also, blaSHV gene was detected in 33.33% (21/63) of the isolates and in all isolates of Pseudomonas aeruginosa (5/5%). 11.11% (7/63) of isolates were virulent with dominance of the fimH gene, especially with Escherichia coli (83.33%). The kitchen samples showed a high prevalence of ESBL-producing strains with fimH gene. This raises the problem of non-compliance with hygiene rules in community cooking and food handling.


Introduction
Antimicrobial resistance has become a major threat to global public health in the 21 st century [1]. It is estimated that, by 2050, 10 million deaths per year and up to $100 trillion of global economic loss will occur if the problem of antimicrobial resistance is not immediately resolved [2]. In recent years, in the Enterobacteriaceae family in particular, there has been increasing evidence of a high prevalence of multiresistant Gram-negative bacteria. e use of antibiotics in human medicine, veterinary medicine, and agriculture has been the main selective pressure for the emergence of drugresistant bacteria [3]. e emergence and rapid spread of extended spectrum β-lactamase-(ESBL-) producing bacteria poses a threat to the population, not only in developing countries, but also in developed countries, as ESBLs may degrade not only penicillin, but also other classes of β-lactams. In addition, most ESBL-producing bacteria are multiresistant [4]. e increased presence of Enterobacteriaceae producing ESBL in non-clinical environments such as food, animal husbandry, and agriculture and in different aquatic environments has been shown in several studies [5][6][7]. In addition, multiresistant enterobacteria have been isolated from healthy individuals [8], suggesting that the community may serve as a reservoir. It is known that the intestinal tract is an excellent reservoir for these bacteria and that colonized people may be infected later. In fact, the presence of Escherichia coli producing CTX-M has been reported in healthy children [9], which shows the potential risk of spreading β-lactamasemediated resistance. e dissemination of β-lactamase-encoding genes in different geographical regions may partly be due to the fact that many resistance genes are carried on selftransmissible or mobile plasmids capable of spreading horizontally between and within species [3]. During gene transfer, resistance mechanism to other classes of antibiotics can easily be combined [10].
ere are different types of β-lactamase enzymes present in bacteria inhabiting diverse environments and exposed to different selection pressures [11]. e virulence factors of Enterobacteria have an important role in the development of infections [12]. e molecular characteristics and functions of these virulence factors have been determined [13]. e most likely theory is that strains of pathogenic enterobacteria evolved primarily from non-pathogenic strains by acquiring virulence genes (by horizontal transfer of plasmid, bacteriophage, transposon, and pathogenicity located at chromosomal loci), which provides increased adaptive capacity to new niches and allows pathogens to increase their ability to cause wide range of infections [14,15]. Moreover, evolution of virulence and resistance occurred at different time points. Virulence has evolved over millions of years, but resistance is a new phenomenon that has occurred over the past few decades (after introduction of antibiotics). As aquatic bacterial genetic reactors, aquatic environments promote the transmission of resistance genes virulence, and virulencerelated characteristics among bacterial strains [16].
In Benin, several studies have reported the increasing evolution of the prevalence of broad-spectrum β-lactamase-producing enterobacteria in clinical infections [17][18][19]. But few studies have reported broad-spectrum β-lactamase-producing bacteria isolated simultaneously from external environment, kitchen environment, and food products. e current study is aimed at providing scientific knowledge regarding the spread of broad-spectrum β-lactamase-producing enterobacteria in different environments in Benin, and also at characterizing the resistance genes and determining their virulence factors. Such data will be used to better understand clinically observed antibiotic resistance. e study will also serve as a basis for recommendations to combat antibiotic resistance.

Zone and Period of Study.
e study was carried out mainly in the South of Benin, and in some towns in Northern Benin. 2 . In Southern Benin, the average annual temperature is 28°C and the humidity varies between 69 and 97% [19]. e phytogeographic districts covered by this study are Coastal, Pobè, Ouémé valley, Plateau, Zou, Borgou, and Donga. e study was done in December 2020. e samples were collected from community kitchen environment (cloth, sink, sponge, soap, and bench), external environments (sediment/mud, sewage, feces of animal's cattle, pig, and chicken), and foods products (crabs, fresh shrimps, dried fishes, fermented fishes, tomatoes, and leafy vegetable) ( Figure 1).

Seeding
Sewage Samples. After homogenization, 1 ml of each sample was taken and added to 9 ml of sterile distilled water contained in a screw tube for the preparation of 10 −1 dilution. Successive decimal dilutions were made up to 10 −3 and 100 μl of each dilution was then inoculated on McConkey and ChromID ™ ESBL solid media (Biomerieux, France).
Solid Samples. Previously, all food samples were ground in a sterile mortar. 9 ml of sterile distilled water was added to 1 g of each solid sample (sediment/sludge, animal feces, and crushed food item) contained in a screw tube (stock solution). After homogenization through a vortex, 1 ml of the stock solution (10 −1 dilution) was added again to 9 ml of distilled water (10 −2 dilution). Successive decimal dilutions were made up to 10 −4 . e last three dilutions (10 −2 , 10 −3 , and 10 −4 ) were then seeded on McConkey solid media and ChromID ™ ESBL (Biomerieux, France).  e reference strain (E. coli ATCC 25923) and the strains to be tested were previously subcultured on the MH agar and incubated for 24 h at 37°C in order to provide young colonies. For each strain inoculum correspondent standard McFarland 0.5 was prepared. Using a sterile swab, the inoculum obtained was cultured by swabbing on dried MH II medium. e spreading was rotated three times in order to obtain a homogeneous distribution of the inoculum. e pre-impregnated antibiotic discs were then gently deposited on the surface of the agar with a disc dispenser. e agar plates were incubated for 24 h at 37°C.
A double disc diffusion method was used for the phenotypic confirmation of ESBL production according to the guidelines described by the Clinical Laboratory Standards Institute [21]. Klebsiella pneumonia (ATCC 700603) was used as the standard strain.

Molecular Detection of Virulence-Associated and ESBL
Genes.
e DNA of all isolates was extracted using the Qiagen red extraction kit. e isolates were tested for different virulent genes using standard PCR with six sets of specific primer pairs (Table 2). e reaction mixture consisted of 9.5 μL of water without DNA, 12.5 μL of the 2XPCR Master Mix reagent, 0.5 μL of each primer pair, and 2 μL of bacterial DNA. e virulence genes that were targeted for amplification by PCR were iss, fimH, Stx-1, Stx-2, biofilm, and PapC, whereas the ESBL genes targeted for amplification were blaTEM, blaSHV, blaCTX-M1, blaCTX-M2, blaCTX-M9, and blaCTX-M15. , bla CTX-M9 , and bla CTX-M15 genes was carried out using the method described in [22,23]. e amplification of fimH and iss genes was performed using the method described in [24]. Regarding PapC gene, its amplification was also carried out using the method previously described in [25]. e amplification conditions of the other genes (Stx-1, Stx-2, and Biofilm) involved initial denaturation at 94°C for 30 sec, followed by 30 cycles at 94°C for 30 secs of denaturation at 68°C for 1 min of hybridization, at 68°C elongation for 1 min. Final extension was carried out at 68°C for 5 min and maintained at 4°C. Amplification products were separated by 2% agarose gel electrophoresis with 5 μg/ml red gel and a 100 bp DNA ladder as a molecular weight marker. e migration was carried out at a scale of 90 V/cm for 20 min. Amplification bands were visualized and photographed under ultraviolet light (UV).

Statistical Analyses.
e data collected was coded and analyzed using GraphPad prism 7 software. Phenotypic identification methods were performed independently in triplicate. e software R was used to calculate the correlation coefficient between the results of the detection of ESBL + on agar and by the effect of the double synergy. e latter provided us with preliminary data on strains that produce β-lactamase. Of the 130 samples analyzed, 21 β-lactamase-producing strains were isolated. Furthermore, 47.61% β-lactamase-producing strains were isolated from community kitchen samples, 42.85% from external environmental samples, and 9.52% from food products.

PCR Detection of Virulence-Associated Genes of Isolates.
Six genes (iss, fimH, stx1, stx2, biofilm, and PapC) of virulence were investigated in the genome of the 63 identified isolates. Only the fimH gene was detected as indicated in Table 4. Figure 3(a) shows the agarose gel of fimH after PCR.

Prevalence of Pathogens in the Samples.
Of the 130 samples analyzed, we were able to isolate 63 bacterial strains, all Gram-negative bacilli. e kitchen environment samples contained 36.50% (23/63) of the isolated bacterial strains, followed by external environmental samples (33.33%) (21/63), whereas the samples from food products were 30.15% (19/63). In addition, 47.61% of the bacteria isolated from cooking samples are ESBL producers. ese results show that there is a high risk of contamination of food products in these kitchen environments if the appropriate practice of hygiene is not maintained. A report from a study on ESBL-producing Enterobacteriaceae isolated from food prepared in the kitchen of a hospital center [26] showed that 8.04% of foods from the hospital kitchen were contaminated with ESBL-producing strains. is low prevalence reflects the minimum respect of hygiene rules in the kitchen of the hospital. us, the risk of contamination was lower in this environment, but this should not be extrapolated to the community kitchen environment, where safety standards are probably less rigorously respected [27]. Of 48 external environmental samples (wastewater and sediment), 42.85%% of the strains produced ESBL. is proportion is relatively lower than that obtained by [28], where 52.69% of the ESBL-producing Enterobacteriaceae were reported. However, these results are relatively high than those obtained by [29], where 15.47% of the ESBLproducing bacteria were noted. ese different results show that there is a real problem of environmental contamination in Benin. Several studies have been conducted on the isolation of ESBL-producing E. coli from rivers. e human and animal ESBL-producing E. coli in various river waters may be related to the nature of their watersheds, the intensity of    Journal of Environmental and Public Health antibiotic use, and the relative degree of fecal contamination by humans and animals. Clonal groups similar to those found in human infections have also been isolated from contaminated rivers adjacent to large cities [30,31]. e prevalence of ESBL-producing bacteria was 13.75% from fecal animal samples. Also, this same proportion was obtained in piglet feces in India [32]. ese results reflect the risk of contamination of slaughter meat and food products of animal origin. Of the 43 food samples, there is a prevalence of 9.52% of ESBL-producing bacterial strains. is result is relatively small compared to the 20.75% of Gram-negative Bacillus (GNB) ESBL producers obtained by [33] from hospital samples in Benin.

Antimicrobial Resistance.
In this study, 63 bacterial strains from 21 species were isolated. e most identified bacterial species is C. luteola (10/63). Antibiotic susceptibility analysis showed high resistance of bacterial isolates to penicillin, cephalosporins, and aztreonam. is high resistance of the isolates coincides with the results obtained by [28] that was conducted on environmental samples and food in Spain. In addition, it was noted that the isolates showed significant resistance to β-lactamase inhibitors such as amoxicillin + clavulanic acid (70.39%). ese results coincide with those from [34], where resistance of strains to betalactamase inhibitors is noted. is suggests that there is an increasing need to encourage the use of alternative betalactamase inhibitors. For the multidrug resistance analysis of bacterial isolates, 10 (16.81%) of the isolates were resistant to at least three different families of antibiotics. ese results further emphasize the problem of multidrug-resistant bacteria in the environment which may threaten both human and animal health. e resistance of a high proportion of

ESBL and Virulence-Associated Genes Detection.
Phenotypic detection of bacterial ESBL-producing strains was carried out by the double synergy technique. Only 21/63 of isolates were ESBL-producing; a correlation (r = 1) was noted between the two phenotypic detection techniques of ESBL-producing isolates. is involves the culture of samples on ChromID ™ ESBL agar and the search for the double synergistic effect during the antibiotic susceptibility test. e presence of the bla SHV gene in the genome of 33.33% of isolates confirms this correlation. P. aeruginosa (100%) was the most predominant among these strains with ESBL genes. However, among the ESBL genes sought, only the bla SHV gene was found in the ESBL-positive isolates. e bla SHV carrying organisms have also been detected in a variety of environmental and food samples [36]. Despite the high cefotaxime resistance observed, the bacterial strains were negative to bla CTX-M , suggesting that other genes may be involved in the resistance mechanism. e search for virulence genes also showed the presence of a single virulence gene in the isolates; the fimH gene, also known as the fimbriae adhesion gene type 1 (fimH), recorded the prevalence of 11.11% in this study. E. coli (83.33%) was the most predominant bacterial among this group. It is believed that fimH contributes to the protection of E. coli from heterophilous hosts [37]. Its role in the virulence of Avian Pathogenic E. coli strains remains controversial and is the subject of conflicting reports [38]. Other studies found higher occurrences of the fimH gene, particularly those from [39], which found a prevalence of 98.1% among 524 Avian Pathogenic E. coli isolates. *

Conclusion
is study allowed us to characterize the multidrug resistance, virulence, and ESBL production of bacterial strains isolated from the external environment, food products, and community kitchen samples. Only 33.33% of the isolates were ESBL-positive and 11.11% harboured the virulence fimH gene. Only the bla SHV gene was present in the genome of isolated strains. e community kitchen samples showed a high prevalence of ESBL-producing bacteria and fimH gene. is raises the problem of non-compliance with hygiene rules in community kitchen.

Limitations
is study presents some limitations. We did not identify exact bla SHV alleles due to PCR limitation. Secondly, we could not confirm carbapenem resistance genotypically. In addition, we could not detect other β-lactamase genes and virulence factors that may be present. Future studies including whole genome sequencing could help to address these limitations.

Data Availability
All data generated or analyzed during this study are included in this published article.

Conflicts of Interest
e authors declare that they have no conflicts of interest.