Detection of Extended Spectrum ß-Lactamase-Producing Escherichia coli with Biofilm Formation from Chicken Meat in Istanbul

Antimicrobial resistance is one of the major public health problems worldwide. This study aimed to detect the presence of extended-spectrum β-lactamase-(ESBL-)producing Escherichia (E.) coli in chicken meat in Istanbul, Türkiye. Raw chicken meat samples (n = 208) were collected from different sale points and analyzed for ESBL-producing E. coli. In total, 101 (48.5%) isolates were confirmed as E. coli by PCR, of which 80/101 (79.2%) demonstrated multiple antibiotic resistance. Resistance against amoxicillin-clavulanic acid was most frequent (87.1%). Eighteen isolates (17.8%) demonstrated phenotypical ESBL resistance, as assessed by the double disc synergy test (DDST). Isolates were tested for the presence of β-lactamase genes and mobilized colistin-resistant genes. The blaTEM group was most frequently detected (97.02%), followed by blaCTX m (45.5%), blaSHV (9.9%), and blaOXA-2 (0.9%). However, mcr genes and blaNDM, blaKPC, blaVIM, and blaOXA-48 genes were not found in any isolate. E. coli strains were tested for biofilm formation in six different media [Nutrient broth, LB broth, Tryptone Soya broth (TSB), TSB containing 1% sucrose, TSB containing 0.6% yeast extract, and BHI]. Biofilm formation by E. coli isolates (44/101, 43.5%) was highest in TSB with 1% sucrose. It is worth noting that all biofilm-producing isolates were found to harbor the blaTEM-1 gene, which can indicate a high level of antibiotic resistance. This is the first report about ESBL-producing E. coli in poultry meat, the exposure of consumers in Istanbul metropolitan areas, and the ability of E. coli from this region to produce biofilms.


Introduction
Antimicrobial resistance (AMR) is a major public health concern worldwide, leading to the clinical failure of antimicrobial therapy.Poultry production worldwide uses substantial amounts of antibiotics, and there are concerns about high AMR levels among bacteria isolated from poultry samples.The frequency of AMR and genes causative for AMR in Escherichia coli (E.coli) isolates are reported from many different countries [1,2], including Türkiye [3,4].A one-health approach is required to control the emergence and effects of antibiotic resistance [5].In this context, multiple drug-resistant (MDR) E. coli in chicken meat are of concern, with implications for human consumers, the health of animals, and the environment [6].The rapid emergence of multidrug-resistant E. coli strains has resulted in human morbidity and even fatalities [7].
Beta-lactam antibiotics represent one of the major classes of antimicrobials.Emerging antibiotic resistance has compromised their antibacterial efficacy [8].In some bacteria, particularly Gram-negative bacteria, beta-lactamases have evolved, i.e., enzymes that can break down beta-lactam antibiotics.The expression of beta-lactamases is one of the most studied and widespread mechanisms of antimicrobial resistance [8].ESBL producers have been identified mainly in the taxonomic order Enterobacterales.These bacteria can harbor multiple determinants of antibiotic resistance, making it more difficult to treat infections caused by these pathogens [9].Enterobacterales producing ESBL are thought to have colonized more than 1.5 billion individuals around the world, essentially in devastated countries and also in industrialized countries.Furthermore, ESBL producers have complex epidemiology, most prominently E. coli and Klebsiella pneumoniae, whose reservoirs include the environment (soil and water), wildlife, livestock, food, and pets [10].
E. coli that produces ESBLs has been identified as a major multi-resistant pathogen associated with serious hospital-and community-acquired infections worldwide, particularly where sanitation and hygiene practices are poor or lacking [11].The European Food Safety Authority (EFSA) has identified ESBL/AmpC-producing E. coli as one of the main priority hazards, especially in poultry.In various studies, ESBL-producing E. coli have been isolated in broiler farms and slaughterhouses, concluding that chicken meat is a potential source of infection for humans [12].The source and transmission methods of ESBL-producing E. coli strains have yet to be fully explored.However, over the last decade, more research on ESBL-producing E. coli in animals and animal feeds has been undertaken in Europe [13,14].
Biofilms generally comprise extracellular polymeric substances that enable bacteria to adhere to surfaces and communicate with each other [15].Biofilms are described as the most widespread and most successful life forms on Earth.Quorum sensing (QS) alters the gene expression of bacteria according to the size of the bacterial community formed by the biofilm.The production of antimicrobial proteins can be promoted by QS, which can lead to increased antimicrobial resistance.Moreover, the matrix formed by biofilms facilitates the transfer of antibiotic resistance genes [16].It has been suggested that betalactamases can be secreted from bacteria into the surrounding biofilm matrix and, thus, into the environment [17].

Sampling
In total, 208 raw chicken meats were collected from different sale points (market,

Isolation and Identification of E. coli by Conventional Methods
The isolation and identification of E. coli were conducted according to the ISO 16649-2 standard method [18].Several pieces were taken from each sample to give a weight of 10 g.To this 10 g, 90 mL of Buffered Peptone Water (Oxoid CM 0509, Basingstoke, UK) was added, and the suspension was mixed in a stomacher (Interscience, Saint Nom la Bretèche, France).Subsequently, Tryptone Bile X Glucuronide Agar (TBX; Oxoid CM 0945) was inoculated and incubated at 41 ± 1 • C for 24 h.Suspected (green) E. coli colonies on TBX agar were transferred onto Eosin Methylene Blue Agar (EMB; Oxoid 0069B) plates, which were incubated at 37 • C for 24 h for verification.
After the isolation, E. coli strains were plated for purity testing, and a single colony was streaked onto Tryptone Soya Agar (TSA; Oxoid CM 0131) from EMB and incubated at 37 • C for 24 h.Isolated strains were frozen in 20% glycerol stocks (Sigma G5516, Sigma Aldrich, Darmstadt, Germany) and stored at −20 • C for further analysis.E. coli strains were cultured on Tryptone Soya broth (TSB; Oxoid CM 0129) at 37 • C for 24 h.In total, 750 µL from this enrichment culture broth was transferred into Eppendorf tubes (2 mL) and centrifuged at 10,000 rpm for 5 min.Then, the bacterial pellet was resuspended in TE (10 mM Tris-HCl pH 8.0 and 1 mM EDTA) containing lysozyme (Sigma 7651) and incubated at 37 • C for 18 h.In the next stage, 250 µL of 10% SDS and 20 µL of 20 mg/mL proteinase K (Sigma P2308-100MG)/dH 2 O were added into the Eppendorf tube, and the tubes were incubated at 56 • C for 2 h.Then, 750 µL of phenol/chloroform/isoamyl alcohol (25:24:1) (Amresco K169, Solon, OH, USA) was added, and the tubes were centrifuged at 14,000 rpm for 15 min.The supernatant was transferred into another Eppendorf tube.First, 150 µL of 5 M NaCl and then 700 µL of 2-propanol (Merck 1096342511, Darmstadt, Germany) were added for precipitating DNA, followed by centrifugation at 14,000 rpm for 10 min.Then, the supernatant was removed, and the pellet was washed twice with 1 mL of 80% cold ethanol (Sigma 459844) and centrifuged at 14,000 rpm for 10 min [19].The pellet was resuspended in 75 µL of ultra-pure water.The acquired DNA was tested by BioTek Epoch2 (Agilent, Santa Clara, CA, USA) for its quality and stored at −20 • C.
Petri dishes were evaluated after 18 ± 2 h of incubation at 35 ± 2 • C, and E. coli strains were established to be sensitive or resistant following the EUCAST [22] and the Clinical and Laboratory Standards Institute (CLSI) [23] guidelines that defined the zone diameter breakpoints for each antimicrobial agent tested.For tetracyclines, breakpoints were provided only by CLSI [23] but not by EUCAST [22].

Detection of ESBLs Using Double Disc Synergy Test
For the phenotypic confirmation of ESBL in Enterobacterales, the double disc synergy test was used.For this test, paired discs of CAZ (30 µg) and CTX (30 µg) were used, and each was positioned at distances of 20 mm (center to center) from the AMC disc (AMC, 20 + 10 µg) [24].

Genotypic Determination of Antibiotic Resistance Genes in E. coli Strains Determination of ESBL Genes in E. coli Strains
A PCR assay was conducted to determine whether the isolates (101 E. coli) harbored bla SHV , bla TEM , bla CTX-M , and bla OXA .The PCR mix was as follows: 2.5 µL of DNA samples, a 10× KCL buffer at 2.5 µL, a dNTP mix (dATP, dCTP, dGTP, and dTTP) at 2.5 µL, MgCl 2 at 1.5 µL, each primer at 0.5 µL, Taq DNA polymerase (Thermo Fisher EP0404; Thermo Fisher Scientific, Waltham, MA, USA) at 0.4 µL and dH 2 O at 12 µL, to give a final volume of 25 µL.A multiplex PCR to detect ESBL's genes was applied, and initial denaturation at 95 • C for 15 min was followed by 30 cycles of 94 • C for 30 s, 62 • C for 90 s, and 72 • C for 60 s, with a final extension at 72 • C for 10 min in the thermal cycler (Veriti; Applied Biosystems, Waltham, MA, USA).The amplified PCR products were subjected to electrophoresis at a 1.5% agarose gel with the addition of 5 µL of safe view (ABM, Richmond, BC, Canada) (Table 1).
Monoplex PCR was applied to detect carbapenem resistance genes, with the following conditions: initial denaturation at 94 • C for 5 min, followed by 30 cycles of 94 • C for 30 s, 30 s at the specific melting temperature given in Table 2, then 72 • C for 60 s, and a final extension at 72 • C for 10 min in the thermal cycler.The amplified PCR products were subjected to electrophoresis with 1.5% agarose gel and an addition of 5 µL of safe view (ABM, Richmond, BC, Canada) [29] (Table 2).

Target Gene
Primer Sequence (5 Detection of mcr Genes in E. coli Strains The PCR assay was conducted to determine whether the isolates (101 E. coli) contained mcr-1, mcr-2, mcr-3, mcr-4, and mcr-5 genes.Conditions of multiplex PCR to detect mcr-1, mcr-2, mcr-3, mcr-4, and mcr-5 genes were as follows: initial denaturation at 95 • C for 15 min, followed by 30 cycles of 94 • C for 30 s, 58 • C for 90 s, and 72 • C for 60 s, and a final extension at 72 • C for 10 min in the thermal cycler (Table 3).The PCR assay was conducted to determine whether the isolates harbored mcr-6, mcr-7, and mcr-8.The monoplex PCR conditions to detect mcr-6, mcr-7, and mcr-8 genes conditions were as follows: initial denaturation at 95 • C for 15 min, followed by 30 cycles of 94 • C for 30 s, 90 s at the specific melting temperature given in Table 4, and 72 • C for 60 s; the final extension was conducted at 72 • C for 10 min in the thermal cycler (Table 4).The amplified PCR products were subjected to electrophoresis using 1.5% agarose gel with an addition of 5 µL of safe view (ABM, Canada) for mcr-6 [30], mcr-7 [31] and mcr-8 genes [32].
E. coli isolates were incubated overnight (12-18 h) on TSB Agar at 37 • C.Then, the overnight cultures were adjusted to a density of 0.5 of McFarland and pipetted into microplates (3599 Corning Costar; Corning, NY, USA) with different media, followed by incubation at 37 • C for 24 h.After incubation, absorption was measured using a microplate reader (BioTek Epoch; Agilent, Santa Clara, USA) at 600 nm.The microplates were washed three times for biofilm detection as follows: 0.9% NaCl, methanol, crystal violet (Merck 1159400, Darmstadt, Germany), and 33% acetic acid (Merck 159166, Darmstadt, Germany) to remove the excess stain.After drying, the microplates were incubated with 5 mL of 96% ethanol for 15 min.The optical density (OD) at 595 nm was measured with a microplate reader (Bio Tek Epoch; Agilent, Santa Clara, USA) [34].
Chicken meat is widely processed and served in fast-food establishments and restaurants [35].Therefore, poultry meat comprises about two-thirds of the total meat production in the world.Poultry meat production in Türkiye amounted to 2,245,770 tons in 2021, making Türkiye the world's seventh largest chicken meat exporter in 2021 [36].
Baran et al. [3] obtained E. coli isolates from 105 (70%) of 150 chicken thighs sampled in Erzurum, Türkiye.In the present study, a lower percentage (48.5%) of E. coli was isolated when only chicken thighs were considered.In contrast, Guven and Kizil [4] reported a prevalence of E. coli of merely 7% (7/100) in chicken breast samples originating from Ankara, Türkiye.These results were lower than our findings (27.7%).The differences may be attributable, in part, to the continuous temperature control of refrigerators or ambient temperature in supermarkets [35].
None of the isolates was resistant to TZP.Resistance to CTX, MEM, CIP, AMP, and SXT was higher in E. coli from chicken meat samples taken from the European side of Istanbul.In comparison, resistance to CN, FOX, CAZ, CXM, ATM, F300, and C was higher in E. coli from chicken meat samples taken from the Asian side of Istanbul.This difference in antibiotic resistance was the most pronounced regarding resistance against MEM (Table 6).
In a previous study [4], E. coli isolated from chicken meat in Ankara, Türkiye, were tested for phenotypic AMR by the disc diffusion test.All isolates were found to be resistant against TE.For AMP and CIP, AMR was found in 85.7% of the isolates; for C, SXT, CAZ, CTX, MEM, and CN, the corresponding figures were 71.4%, 42.8%, 28.5%, 28.5%, 28.5%, and 14.2%.A study conducted in Hatay, Türkiye, demonstrated similar results, with 91.4% of E. coli isolated from chicken meat showing resistance against TE, 75.6% against CXM, and 42.8% against SXT [37].Our results indicate a similarly high frequency of AMR against AMP, TE, and SXT resistance, which is comparable to the findings of Güven and Kizil [4] and Önen et al. [37].However, Inat et al. [38] reported lower frequencies of AMR in E. coli isolates originating from Samsun in Türkiye.Merely 15% of the E. coli strains were resistant to AMP, 11.3% to CN, 10% to C, and 8.8% to IMP.Interestingly, meropenem and/or imipenem resistance was detected in E. coli strains in recent studies [4,38], which has not been reported before [3,37].[22] does not provide a breakpoint value for this antibiotic.
Among the 18 E. coli strains in our study that tested positive for ESBL by DDST, 17 (94.4%)exhibited MDR (Table 7).In detail, 17 strains contained bla TEM , 12 strains contained bla CTX-M , 1 strain contained bla SHV , and 1 strain had bla OXA .Other workers found only 4 of 18 ESBL-producing isolates using the DDST test [38].This difference can be explained by the fact that bla TEM was detected in 97% of the E. coli strains in our study.In our study, 78% of the E. coli strains were AMP-resistant since they contained the bla TEM gene, which is one of the main genes responsible for AMP resistance.The high frequency of the bla TEM gene, which is one of the ESBL genes, is contrasted with the low percentage of phenotypic ESBL (17.8%).Similarly, phenotypic ESBL production was less frequent than the presence of bla CTX-M (45.5%), which is the most important ESBL gene, indicating that the relationship between genotypic and phenotypic beta-lactamase production needs to be investigated.Similar to our study, Inat et al. [38] reported that the presence of genotypic ESBL genes was more frequent than phenotypic ESBL production.It is important to note that both studies investigated phenotypic ESBL production by using the double disc synergy test.tourist destinations can be critical.Istanbul is a metropolis and the biggest city in terms of population and economy in Türkiye.Istanbul hosted almost 17.5 million tourists and became the most visited tourist destination place in the world in 2023 [42].In this context, Istanbul is an import and export center and transport corridor between Europe, Asia, and the Middle East.
In Germany, 185 ESBL-producing E. coli were found in 175 of 399 chicken meat samples [43].ESBL genes were identified by multiplex PCR: bla TEM-52 (n = 16), bla CTX-M-1 (n = 77), and bla SHV-12 (n = 82).The percentage of bla CTX-M-1 was similar to that in our study.Such a high incidence of bla CTX-M-1 in Berlin, the most populated city in Germany, and Istanbul, the largest city in Türkiye, indicate that this gene threatens large populations in different geographical areas.
In Türkiye, 152 E. coli strains were analyzed by combining disk diffusion tests, and 28 (18.4%)strains were found to produce ESBL [41].In another study in Türkiye, 100 chicken meat samples were collected, and 214 E. coli isolates were cultivated from 72 samples positive for E. coli [44].Five strains (2.3%) were phenotypically ESBL, whereas PCR analyses detected bla TEM , bla SHV , and bla CTX-M in 8, 3, and 7 E. coli strains, respectively.In the present study, the frequency of ESBL-producing E. coli from chicken meat was significantly higher than that reported by Çil et al. [41] and Bilge et al. [44].In particular, there was a high increase in the frequency of bla TEM but no significant difference in the rate of phenotypic ESBL-producing E. coli.Many ESBL-related genes have been discovered in recent years.This study contributes to a better understanding of the ESBL-producing E. coli epidemiology in Türkiye and provides important data for future studies.
Our study showed no evidence of E. coli strains carrying mcr and carbapenem resistance genes.However, similar studies are reporting on these extremely important antimicrobials.Randall et al. [45] did not detect colistin-and carbapenem-resistant E. coli isolates in 622 chicken meat samples; all ESBL-producing E. coli isolates were sensitive to meropenem and colistin.The striking difference to our study is that bla CTX-M-1 was the dominant gene in the isolates tested by Randall et al. [45], whereas the bla TEM-1 gene was commonly observed in our study.We detected meropenem-resistant E. coli isolates, especially in samples from the European side of Istanbul, whereas bla NDM , bla KPC , bla VIM , and bla OXA-48 genes related to this resistance were not detected.It should be noted that carbapenem resistance genes were not detected by PCR either in our study or in that of Randall et al. (2021).However, Randall et al. [45] used CHROMagar KPC in their carbapenem resistance study, which may account for some differences.

Distribution of MDR in E. coli Strains
The number of MDR bacteria is increasing, and the foodborne transfer of antimicrobial resistance is an important issue.Bacteria harboring resistance genes may originate from (food-producing) animals or be present in foods due to cross-contamination, thus threatening public health.Increased numbers of infections with bacteria and antibiotic-resistant genes complicate treatment [46,47].
Baran et al. [3] detected MDR in 99 (94.29%)E. coli strains in Türkiye.All E. coli isolates were sensitive to meropenem.Conversely, the rate of MDR (79.2%) was lower than our study.However, the higher rate of meropenem resistance (34%) is of concern.This difference may have been caused by the fact that our study was conducted in Istanbul, where chicken meat from many production points in Türkiye is available, whereas Baran et al. [3] tested samples in Erzurum province, which mostly reflects the production in eastern Anatolia.
No isolate showed resistance against TZP.In a study conducted in Brazil, 200 swabs were taken from broilers, and 13 ESBL-producing E. coli strains were found as an MDR [48].In addition, Ferreira et al. [48] identified 16 ESBL-producing Enterobacteria isolates.Among them, 13 isolates were E. coli.The frequency of resistance to CN, C, and SXT was similar to that reported in our study, whereas resistance against CIP was much more frequent (15/16, 93.7%).Considering the contribution of chicken meat to human diets and exposure to raw chicken meat during food processing, MDR in E. coli isolated from chicken is a serious health concern.

Biofilm Formation of E. coli and ESBL-Producing E. coli Strains
The results of the biofilm assay indicate that E. coli strains can form more biofilm in TSB media containing 1% (w/v) sucrose (n = 44) than in other media.Moreover, the highest number of biofilm-producing E. coli strains was isolated in the wing part of a chicken meat sample from the European side with a value of 3223 at 595 nm in TSB containing a 0.6% (w/v) yeast extract (Table 8).There are 7 E. coli strains in TSB as follows: 6 E. coli strains in TSB containing 1% (w/v) sucrose, 5 E. coli strains in TSB containing 0.6% (w/v) yeast extract, 3 E. coli strains in BHI, and 1 E. coli strain in NB-produced biofilms.Genotypic ESBL-producing E. coli strains demonstrated their ability to produce biofilms with 38 strains in TSB, 50 strains in TSB containing 1% (w/v) sucrose, 34 strains in TSB containing 0.6 % (w/v) yeast extract, 18 strains in BHI, 5 strain in NB and 4 strains in LB.The formation of biofilms by bacterial strains is a concern in both food crops and food processing facilities.The production of biofilms by MDR bacteria is a major concern in the food chain.Limited data on biofilm-producing E. coli strains in chicken meat samples are available worldwide.In Brazil, 150 samples were collected from the largest chicken meat exporter, and 88 E. coli strains were found [40].Among these strains, 84 (56%) could produce biofilms, as assessed by the microplate method.In addition, 17.04% of the E. coli strains were capable of ESBL production.The authors [40] used only TSB as a medium to study biofilm formation, whereas six different media were used in our study.The present study obtained results similar to those of Crecencio et al. [40], with bla TEM-1 (73.3%) being the most common gene found in their analysis of ESBL-producing E. coli.The significant difference between the two studies was the rate of bla SHV-1 .They reported the presence of bla SHV-1 in 46.6% of the bacterial isolates in 2020, while the frequency observed in this study was 9.9%.Similar results in samples from two different continents demonstrate the importance of biofilm-forming E. coli isolates.These isolates, which are more resistant to environmental conditions, threaten global health.The high level of plasmid-origin gene transfection in both studies indicates that ESBL-producing bacteria will continue to increase.
butcher) from May to August 2021 in Istanbul, Türkiye.Half of the samples were collected from the European side of Istanbul [drumsticks (n = 14), breasts (n = 25), thighs (n = 25), and wings (n = 40)].The other 104 samples were sampled from the Asian side [drumsticks (n = 14), breasts (n = 34), thighs (n = 27), and wings (n = 29)].All samples were transported in thermal boxes at ≤+4 • C to the laboratory (Department of Food Hygiene and Technology, ˙Istanbul University-Cerrahpaşa) and were processed immediately upon arrival.

Table 1 .
Primers used for the detection of different β-lactamase genes by multiplex PCR.

Table 2 .
Primers for the detection of different carbapenem resistance genes by PCR.

Table 4 .
Primers used for the detection of different mobilized colistin resistance genes by monoplex PCR.

Table 5 .
Antibiotic susceptibility of E. coli strains.

Table 6 .
Distribution of antibiotic-resistant E. coli strains according to place of sampling (European or Asian part) in Istanbul.

Table 8 .
Biofilm-producing E. coli number and distribution.