Prevalence and Characterization of Extended-Spectrum β-Lactamase-Producing Antibiotic-Resistant Escherichia coli and Klebsiella pneumoniae in Ready-to-Eat Street Foods

As the global urban populations increase with rapid migration from rural areas, ready-to-eat (RTE) street foods are posing food safety challenges where street foods are prepared with less structured food safety guidelines in small and roadside outlets. The increased presence of extended-spectrum-β-lactamase (ESBL) producing bacteria in street foods is a significant risk for human health because of its epidemiological significance. Escherichia coli and Klebsiella pneumoniae have become important and dangerous foodborne pathogens globally for their relevance to antibiotic resistance. The present study was undertaken to evaluate the potential burden of antibiotic-resistant E. coli and K. pneumoniae contaminating RTE street foods and to assess the microbiological quality of foods in a typical emerging and growing urban suburb of India where RTE street foods are rapidly establishing with public health implications. A total of 100 RTE food samples were collected of which, 22.88% were E. coli and 27.12% K. pneumoniae. The prevalence of ESBL-producing E. coli and K. pneumoniae was 25.42%, isolated mostly from chutneys, salads, paani puri, and chicken. Antimicrobial resistance was observed towards cefepime (72.9%), imipenem (55.9%), cefotaxime (52.5%), and meropenem (16.9%) with 86.44% of the isolates with MAR index above 0.22. Among β-lactamase encoding genes, blaTEM (40.68%) was the most prevalent followed by blaCTX (32.20%) and blaSHV (10.17%). blaNDM gene was detected in 20.34% of the isolates. This study indicated that contaminated RTE street foods present health risks to consumers and there is a high potential of transferring multi-drug-resistant bacteria from foods to humans and from person to person as pathogens or as commensal residents of the human gut leading to challenges for subsequent therapeutic treatments.


Introduction
Ready-to-eat (RTE) street foods have food safety challenges globally and are defined as foods for immediate consumption or subsequent use without further processing or preparation and are sold as common street foods in small roadside outlets [1]. RTE street foods could be consumed raw or cooked, hot or cold. Fruits (cut fruits or fruit mixtures) bought directly from street vendors, or local street markets could also be considered RTE if eaten immediately, i.e., without necessarily having to wash, peel, or cut before consumption [2]. As the world urbanizes more rapidly, there is a higher dependency on street foods and specifically RTE foods among the urban population due to time and cost pressures [3]. The consumption of street food increases the potential risk of foodborne illnesses such as diarrhea or traveler's diarrhea [4]. The entry points of foodborne pathogens are from readily contaminated foods from different sources which lack structured hygiene practices during processing, preparation, and storage [5,6].
The World Health Organization (WHO) estimates that foodborne and waterborne diarrheal diseases kill around 2.2 million people annually. According to the Centre for Disease Control and Prevention (CDC), millions of illnesses occur throughout the world each year due to foodborne pathogens [7]. Further, with the increased use of antibiotics to treat foodborne bacterial illness it increases antimicrobial resistance and subsequent ineffectiveness of antibiotics [8]. Therefore, this has resulted in increased infections caused by these Multidrug-resistant (MDR) organisms and is a serious problem encountered throughout the world as they are associated with reduced therapeutic options, increased mortality, and a lengthy hospital stay [9,10]. Escherichia coli, for example, have become a dangerous foodborne pathogen globally responsible for gastroenteritis epidemics in North America, Europe, Asia, and Africa and many strains have been frequently implicated in undercooked foods, contaminated ground beef, raw milk, unpasteurized cider and apple juice, bean sprouts, or fresh leafy vegetables such as lettuce and spinach [11] and have a potential for the emergence of antibiotic resistance. Similarly, resistance in K. pneumoniae has been observed in various community sources including raw vegetables and RTE foods over the years and there are published reports of K. pneumoniae having developed an acquired resistance to carbapenems, the last-line drugs [12].
Among challenges of antibiotic resistance, the increased presence of ESBL-producing foodborne bacteria in street foods is a significant risk for human health because of its epidemiological importance [13]. Hence, there is a need to screen these ready-to-eat street foods to reduce their dissemination into the environment, including food and water. The present study was undertaken to evaluate the potential burden of ESBL-producing antibiotic-resistant Escherichia coli and Klebsiella pneumoniae contaminating the RTE street foods and to assess the microbiological quality of foods in a typical emerging and growing urban suburb of India where RTE street foods are being rapidly establishing with public health implications.

Sample Collection
A total of 100 RTE street food samples were purchased and collected between June and October 2019 (location: Deralakatte suburb of Mangalore city in India) to analyze for ESBL-producing antibiotic-resistant E. coli and K. pneumoniae. The samples consisted of vegetarian and non-vegetarian salads, different sandwiches, a variety of chutneys, sauces and dressings, instant noodles and pasta, vegetarian and non-vegetarian fritters, paani puri water and potato mix (a local Indian street food), mayonnaise, samosas and cutlets, chaats like bhelpuri (mixture of puffed rice with sweet and sour chutney) and potato chaat (mixture of potatoes with green and red chutney, curd, cut vegetables and some dry masala powder), patties with the filling of eggs/ vegetables /chicken, chicken sausages and salami, wet and dry pickles, fruits and vegetable juices, cakes and muffins, and different types of cheese. The food samples mentioned above were collected in a sterile plastic container from a range of street food options such as the local food stalls, canteens, roadside vendors, and small outlets and transported to the laboratory at 4 • C.

Isolation and Identification of Organisms
Approximately 25 g from each sample was mixed and macerated with 3 mL of brain heart infusion broth (BHI) (HiMedia, Mumbai, India) in a sterile mortar and pestle. Then, the samples were homogenized by mixing into 225 mL of buffered peptone water (HiMedia, Mumbai, India) followed by incubation for 24 h at 37 • C. The enrichment was streaked onto MacConkey agar (MA) and Nutrient agar (NA) (HiMedia, Mumbai, India) [14]. Plates were incubated at 37 • C for 24-48 h. Up to three suspected E. coli and K. pneumoniae colonies from each plate were subcultured onto MA with cefotaxime (1 µg/mL) for the detection of potential ESBL producers followed by incubation at 37 • C for 24 h. Bacterial isolates were subjected to a diverse array of standard biochemical tests to identify E. coli and K. pneumoniae [15].

Antimicrobial Susceptibility Testing
The antibiotic susceptibility testing was done and interpreted using standard Kirby-Bauer disk diffusion technique on Muller-Hinton agar ( [16] guidelines using ATCC (USA) standard culture of Escherichia coli 25922 and Klebsiella pneumoniae 76003 as controls. In case of tigecycline, susceptibility was determined according to USA-FDA breakpoints [17]. The multiple antibiotic resistance (MAR) index was determined as the ratio of the total antibiotics used to the number of antibiotics to which the bacterial isolate was resistant [18].

Phenotypic Detection of ESBL
The detection of ESBL was done using double disk synergy test (DDST) and combined disc diffusion test (CDDT) [16]. For DDST, the isolates were swabbed onto MHA and tested for antibiotic resistance to amoxyclav (20 µg/10 µg), ceftazidime (30 µg/mL), and cefotaxime (30 µg/mL). Upon incubation at 37 • C for 18-24 h, ESBL production was detected by the formation of zone of inhibition around the cephalosporins that increases towards the amoxyclav resulting in synergy formation. For CDDT, isolates were lawn cultured on MHA and tested against ceftazidime (30 µg) and cefotaxime (30 µg) with and without clavulanic acid. Plates were incubated at 37 • C for 24 h. A zone difference of ≥5 mm between the disk of cephalosporin and cephalosporin/clavulanate combination, confirmed for ESBL production. The isolates were also confirmed using triple ESBL detection strip Ezy MIC TM (MIX + /MIX) (HiMedia, Mumbai, India) as per the manufacturer's instructions.

Detection for Metallo-β-Lactamases (MBL) and Minimum Inhibitory Concentration (MIC)
Screening for metallo-β-lactamase was tested for isolates showing resistance to any of the carbapenems (imipenem/meropenem) by DDST [19] and confirmed by using meropenem (MRP) and MRP with ethylene-diamine-tetra acetic acid (EDTA) Ezy MIC TM Strip (HiMedia, Mumbai, India) method as per the manufacturer's instructions. K. pneumoniae ATCC 2146 strain was used as control. The results were interpreted as MBL positive when the ratio of the value obtained for MRP to the value of MRP + EDTA was >8 or if the zone was observed on the side coated with MRP + EDTA and no zone observed on the opposite side coated with meropenem.
The MIC against meropenem for all MBL positive isolates was determined using Ezy MIC TM strips (HiMedia, Mumbai, India) as per the manufacturer's instructions and the results were read as per CLSI breakpoints [16]. Bacterial cultures were grown for 6-7 h in 5 mL of Mueller-Hinton Broth (MHB) (HiMedia, India), after which they were lawn cultured onto MHA plates. The E-strips were placed on the center of the dried plates aseptically using an applicator followed by incubation at 37 • C for 24 h. The MIC of the isolate was read as the value where the ellipse intersects the MIC scale on the strip.

DNA Extraction
Bacterial DNA was extracted using the Cetyl trimethyl ammonium bromide (CTAB) method [20]. The bacterial culture grown overnight at 37 • C in LB broth was centrifuged at 11,200× g for 5 min. The bacterial pellet obtained was resuspended in 567 µL of 1× TE buffer (10 mM Tris-HCL, 1 mM EDTA; pH: 8.0), 30 µL of 10% SDS and 3 µL of 20 mg mL −1 proteinase K. The mixture was incubated at 37 • C for 1 h following which, 100 µL of 5 M NaCl and 80 µL of CTAB/ NaCl solution was added, mixed thoroughly, and incubated for 10 min at 65 • C in a water bath. An equal volume of chloroform/ isoamyl alcohol (24:1) mixture was added, vortexed, and centrifuged at 11,200× g for 5 min. The aqueous phase was transferred to a fresh tube and extracted by adding an equal volume of phenol/chloroform/isoamyl alcohol (25:24:1) mixture and by centrifuging at 11,200× g for 10 min. The supernatant was transferred to a fresh tube, and DNA was precipitated by adding 0.6 volumes of isopropanol followed by centrifugation at 11,200× g for 5 min. DNA pellet was washed with 1 mL of 70% alcohol by centrifugation at 11,200× g for 5 min and was vacuum dried. Finally, the DNA was dissolved in 100 µL of 1× TE buffer and stored at −20 • C for further use. The purity and DNA concentration was estimated using a Nanodrop spectrophotometer (ND-1000, V3.3.0, Thermo Fisher Scientific, Waltham, MA, USA).

Detection of Genes Encoding β-Lactamases
Polymerase chain reaction (PCR) assay was carried out to detect three β-lactamase genes bla CTX , bla SHV , and bla TEM and metallo-β-lactamase gene bla NDM in a 30 µL reaction mixture containing 22.2 µL of sterile distilled water, 0.2 µL Taq DNA polymerase, 3 µL of Taq buffer with MgCl2 (TaKaRa, Bangalore, India), 0.6 µL dNTP, 1 µL of 10 pM primer (each forward and reverse) and 2 µL of template DNA. The primer sequences were procured from Juniper Life Sciences, Bangalore, India and the details about their cycling conditions and the references are given in (Table 1). The amplification was carried out in an Eppendorf Mastercycler nexus GX2 (Thermo Fisher Scientific, Waltham, MA, USA) with the optimized PCR program. The amplified products were resolved in 1.5% agarose gel using a 100 base pair DNA ladder (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA) as a size marker. The gels were visualized in a gel documentation system (Bio-Rad, Hercules, CA, USA).

Statistical Analysis
The data were analyzed using Statistical package for social sciences software version 16 (SPSS Inc., Chicago, IL, USA). Descriptive statistics were performed. Antibiotic resistance data were treated as a binary variable (1= susceptible; 2 = resistant). The collected information was summarized using frequency and percentage for qualitative data. Since the sample size was small and the expected frequencies were more than 20% for a two-bytwo table, Fisher's exact test was used to compare the difference in antibiotic sensitivity for E.coli and K. pneumoniae. A significant difference was observed if the p-value was <0.05.

Multidrug Resistance in Escherichia coli and Klebsiella pneumoniae Isolates
The multiple antibiotic resistance (MAR) index of 86.44% of the isolates was above 0.22 which revealed that most isolates were multi-drug resistant, and no isolate was found to be susceptible to all 23 antibiotics tested (Table 4).

Detection of MBL Producers and MIC Value by E-Test
All isolates showing resistance to one or more carbapenem drugs were tested for phenotypic detection of MBL. A total of 12 isolates were found to be positive for MBL by MRP-EDTA Ezy MIC TM Strip test. Out of 12 MBL producing isolates, eight were from K. pneumoniae and four were from E. coli. Of 12 phenotypically positive MBL isolates, 6 (50%) isolates showed MIC value of >32 µg/mL, 4 (33.3%) isolates showed MIC value ranging from 2-8 µg/mL and 2 (16.7%) from 3-16 µg/mL.

Characterization of β-Lactamase Genes
The molecular detection of β-lactamase genes in this study showed a high prevalence of gene bla TEM , followed by bla CTX and bla SHV . There were 24 (40.68%) isolates positive for gene bla TEM , 19 (32.20%) for gene bla CTX and 6 (10.17%) isolates were positive for bla SHV gene. All three genes were positive in only 2 (3.39%) isolates, and both were K. pneumoniae. None of the E. coli isolates were positive for all the genes tested. It was noted that 8 (13.56%) isolates showed the presence of both bla CTX and bla TEM , 2 (3.39%) isolates were positive for both bla CTX and bla SHV , and 5 (8.47%) showed the presence of bla SHV and bla TEM genes together. In 12 (20.34%) of the isolates, bla NDM gene was detected (Figures S1-S4).

Discussion
The present study detected a high prevalence of Klebsiella pneumoniae (27.12%) and Escherichia coli (22.88%) in 100 different common ready-to-eat street food samples. Our results correlate with other studies for the prevalence of E. coli, where 13.43% E. coli were isolated from raw salad samples and 43.3% of K. pneumoniae isolated from different fresh vegetables [24]. However, there are limited studies from India currently reporting the prevalence of ESBL-producing K. pneumoniae from RTE foods and minimal data is available from other countries. The reason seems to be mainly because of different approaches to prevalence estimation. Since there is substantial data available to focus on E. coli and other major foodborne pathogens from various sources, this may have resulted in underestimating K. pneumoniae as a potential organism prevalent in RTE street foods and ESBL producers. Our study reports that 18.51% of ESBL producers were E. coli and 34.37% were K. pneumoniae. Contrary to most of the studies screening for ESBL producers, our study reports higher prevalence of ESBL in K. pneumoniae [14]. In agreement with our results, a study from South Korea has noted 15.8% of ESBL positive E. coli and 84.2% of ESBL-producing K. pneumoniae isolated from sprouts [25]. The primary sources of ESBL producers in this study were different chutneys, chaats, paani puri water, vegetarian and chicken salads, and chicken salami and sausages. This can be due to unhygienic practices starting from processing, packaging, storing and handling of food items and a high potential for impact of impaired and contaminated water quality. In addition, the presence of E. coli is an indication of fecal contamination of food [26] which also suggests unhygienic practices of the food handlers by using contaminated water for cooking and storing food in contaminated vessels and containers.
The present study reports a high prevalence of multi-drug resistance where 51 isolates fell in the category of MAR index >0.22 out of total 59 Escherichia coli and Klebsiella pneumoniae isolates. Each isolate was resistant to at least three antibiotics. This suggests an extensive and inappropriate use of antibiotics and chances of contamination from high-risk sources [17] in the food service industry. Many studies from India have reported extensive use of antibiotics in poultry [27] and agriculture [28,29] which provides insights into the prevalence of the high antibiotic-resistant bacterial pathogens from RTE street foods. The resistance towards cefepime (72.9%), imipenem (55.9%), cefotaxime (52.5%), and meropenem (16.9%) observed from the current study should be viewed seriously as these antibiotics are classified under the category of lifesaving drugs used in treating serious infections [30,31].
Resistance to beta-lactams highlights the need for continuous monitoring of antibioticresistant patterns of E. coli and K. pneumoniae. Carbapenems are the choice of drug for ESBL-producing organisms [32], but resistance to them leaves treatment solutions with no evident alternative for treatment. A recent report from the Infectious Diseases Society of America (IDSA) listed ESBL-producing Escherichia coli among the most important six drug-resistant bacteria to which new therapies are urgently needed [33]. Therefore, this study suggests appropriate measures to be taken for controlling the dissemination of resistant genes containing pathogens by avoiding indiscriminate use of antibiotics without prescriptions for treatments and improving sanitation and hygiene standards for RTE street foods with food handling procedures and practices for food safety. This study has certain limitations. First, a relatively small sample size, and secondly, a limited geographical area selected for sample collection accounting for it, and hence, the results may not be generalized to other sites. Future studies with a more significant number of samples from a wide selected geographical area will provide more comprehensive information on contamination of RTE street foods by multi-drug-resistant bacteria.

Conclusions
This study indicates that contaminated RTE street foods with multi-drug-resistant bacteria collected from street food serving cafeterias and local small-scale food vendors present health risks to consumers. Also, there is a high chance that the multi-drug-resistant genes from one bacteria could horizontally transfer to another among the Enterobacteriaceae family with the help of plasmids and thereby transferring antimicrobial resistance leading to difficulties in selecting and using appropriate therapeutic treatments. Hence appropriate measures should be taken for controlling the dissemination of these resistant genes by planning and following proper antibiotic stewardship regimes in the community. In addition, knowledge about proper sanitation, use of clean water for drinking and cooking, cleaning of utensils and containers for storing food items is essential, along with spreading awareness among the local street food vendors and handlers towards hygienic practices. This study's insights and future extensions also call for harmonizing food safety practices and training for street food establishments on a continuous basis with oversight from local municipalities regulating these food service enterprises.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.