Quinolone and Tetracycline-Resistant Biofilm-Forming Escherichia coli Isolates from Slovak Broiler Chicken Farms and Chicken Meat

Abstract

Escherichia coli isolates from intensive poultry production are associated with antimicrobial resistance and worldwide health problems. The aim of the study was to detect and evaluate the phenotypic and genotypic antimicrobial resistance, biofilm formation, phylogenetic typing, and virulence factors in E. coli isolates from the rectal swabs of chickens from two farms and swabs of chicken meat purchased from Slovakian food markets. Interpretative readings of minimal inhibitory concentration (MIC) revealed dominant resistance to ampicillin (>50%) in both groups. We also detected higher resistance to ciprofloxacin (45%), tetracycline, ampicillin + sulbactam, and trimethoprim + sulfonamide (each >30%). Here, 28.57% of the strains studied were multidrug-resistant (MDR). The formation of weak biofilms was confirmed in 8.8% of E. coli, while one of the strains obtained from chicken cloacal swabs was classified as a strong biofilm producer. The most frequently confirmed phylogenetic groups in E. coli were B₁ and A₁ in all groups. PCR detection revealed the presence of genes encoding tetracycline resistance (tetAB) and plasmid-mediated quinolone resistance (qnrABS), and Int1 (52.9%), Tn3 (76.5%), kpsMT II (8.8%), fimA (97.1%), cvaC (38.2%), and iutA (76.5%) genes in the strains studied. Our results demonstrate that chickens and chicken meat were the source of antibiotic-resistant, biofilm-forming, and virulent E. coli, representing a potential risk from the point of view of the One Health concept.

1. Introduction

Demand for meat is increasing yearly, driven mainly by population growth, changes in consumption trends, and economic development. By 2050, meat production is expected to increase by 76% globally and 37% in developing countries. Rising levels of livestock production are also associated with the increased application of antibiotics in animals to recover and maintain their health and productivity [1].

Antibiotics (ATB) belong to a large group of antimicrobial substances, and are intended for the treatment or prevention of various bacterial infectious diseases. ATB have also been widely used in agricultural and food production, animal husbandry, and fisheries for decades [2]. Food-producing animals are considered potential reservoirs of pathogenic bacteria, such as Escherichia coli, Salmonella spp., Shigella spp., Listeria monocytogenes, Campylobacter spp., Yersinia enterocolitica, Vibrio spp., Enterococcus spp., and others that can enter the human population and lead to a variety of diseases [3]. Livestock continues to account for 50–80% of the world’s ATB production, which includes many ATB that are also medically important in human medicine. The agriculture and food sectors thus play a significant role in the emergence and spread of antimicrobial resistance [4].

Most E. coli strains are harmless commensals that are part of the intestinal microflora of mammals and birds, and rarely cause disease [5]. Due to the high plasticity of the genome, commensal strains can acquire virulence factors and genes encoding resistance to antimicrobial substances through horizontal transfer or mutation, and act as their reservoir. E. coli is among the bacteria that can act as both a donor and an acceptor of antimicrobial-resistant genes [6]. Resistant genes are encoded either chromosomally or transmitted via mobile genetic elements. Pathogenic E. coli strains are the causative agents of intestinal (InPEC) or extraintestinal (ExPEC) diseases [5].

E. coli, as a ubiquitous bacterium, can be found in various elements of the environment (such as water, soil, wastewater, plants, animals, and food), in which several strains survive for a long time, reproduce, and further spread in the environment [7]. E. coli is among the coliform bacteria, and its presence in drinking water, food, or the environment indicates fecal contamination, non-compliance with hygienic conditions, and the presence of other fecal microorganisms. Furthermore, E. coli is also considered an indicator bacterium in ensuring the safety and hygiene of food production [8]. The manner in which animals are handled before, during, and after slaughter makes meat a potential source of pathogenic bacteria. The meat may be contaminated during skin removal or evisceration during the processing of carcasses. The food-producing environment, instruments, handling by workers, improper storage and transport, and failure to follow sanitary principles during marketing also contribute to the cross-contamination of meat [3,9]. The consumption of raw or undercooked meat is also a threat [10].

The resistance of bacteria to antimicrobial substances (including quinolones and tetracyclines) has been increasing at a tremendous rate in recent years, meaning it is now considered by many to be one of the greatest threats of the 21st century [11]. There are several mechanisms mediating quinolone resistance, as follows: (1) Mutations in the subunits of DNA gyrase and topoisomerase IV enzymes, which cause a decrease in the affinity of ATB to the enzyme–DNA complex [9]. (2) Plasmid gene-mediated resistance, where qnr limits the binding affinity of enzymes and limits the number of available target sites on the chromosome, aac(6′)-Ib-cr modifies or inactivates quinolones, and oqxAB, qepA1, and qepA2 genes encode efflux pumps that reduce the intracellular concentration of ATB [12]. (3) Chromosome-mediated resistance ensures changes in the permeability of porins in combination with the excessive expression of efflux pumps [13].

Resistance to tetracyclines is the result of mutation as well as acquired mobile genetic elements. The three mechanisms counteracting tetracycline resistance include ribosomal protection, the enzymatic inactivation of the most important ones, and the efflux system [14]. In the case of E. coli bacteria, nine genes (tetA, tetB, tetC, tetD, tetE, tetG, tetJ, tetL, and tetY) encoding efflux pumps that actively expel antimicrobials from cells have been identified [15], in addition to three genes (tetM, tetO and tetW) encoding proteins providing ribosomal protection to drug delivery from the ribosome or its modification, and one gene (tetX) encoding an oxidoreductase that inactivates tetracyclines [16].

Another important and increasingly frequent problem is the adaptation of pathogens to environmental conditions in food processing establishments and the resulting biofilm formation. A biofilm is a collection of one or more species of microorganisms whose cells attach to surfaces or other cells and embed in an extracellular polymeric matrix that provides protection due to decreased permeability. It has been shown that bacterial antimicrobial resistance can increase by several hundred times in a biofilm. Industrial processes are significantly affected by the formation of E. coli biofilms, which can have a negative impact on food safety and lead to financial losses [17].

The aim of the present study was to detect the presence of and isolate E. coli from cloacal swabs of broiler chickens and chicken thigh muscle meat purchased from a commercial network. The key aim of the study was to obtain and compare results on their phenotype and genotype resistance to antimicrobial substances, and to evaluate biofilm formation. In genotype analysis, we mainly focused on the identification of quinolone resistance genes (qnrA, qnrB, and qnrS, which encode protective proteins) and tetracycline resistance genes (tetA and tetB, which encode efflux pumps). With this study, we want to provide an up-to-date view of the occurrence of antimicrobial-resistant E. coli during meat processing in the Slovak Republic. Overall, testing E. coli from chicken cloacal swabs and chicken thigh muscle meat also provides a comprehensive view of contamination risks, and helps ensure the safety of poultry products.

2. Materials and Methods

2.1. Samples, Bacterial and DNA Isolation

During a six-month period, a total of 105 chicken samples, consisting of 65 feces samples and 40 meat samples, were analyzed. The sampling and purchase of samples in this work were carried out in the eastern part of the Slovak Republic. A sterile cotton swab with the transport Amies medium Copan 108C (Copan Italia; Brescia, Italy) was used to collect cloacal swab samples from two broiler chicken farms. From a commercial network, we purchased chilled chicken thigh muscle meats. A stock suspension and further tenfold dilutions were prepared from the obtained samples according to ISO Standard 6887-1 [18] and then cultured on Endo agar (HiMedia Laboratories; Mumbai, India). Swabs from cloacal and meat samples were also investigated according to ISO Standards 6887-2 [19] and 16649-2 [20] using TBX agar (Oxoid; Hampshire, UK). The inoculated plates were incubated under aerobic conditions for 24 h at two temperatures: 37 °C (a temperature commonly used for the detection of coliform bacteria) and 44 °C (a more selective temperature for thermotolerant coliform bacteria). Typical colonies were then transferred to the surface of Nutrient agar (HiMedia Laboratories; Mumbai, India) and incubated for 24 h at 37 °C. From growth cultures, we performed the extraction of DNA and species identification.

DNA isolation was carried out by the modified freezing–boiling method (this study). Briefly, into 200 µL of ultrapure, sterile, and deionized water were added 2–3 solitary colonies of pure 24 h bacterial culture from the surface of the Nutrient agar. The tubes’ contents were then homogenized until a milky suspension formed. Afterward, the prepared tubes were placed in a freezer at −22 °C for 20 min. The prepared mixtures were boiled for 10 min at 95 °C, quickly cooled on ice, and then centrifuged at 13.500× g for 10 min. From the obtained supernatants, 100 µL was taken for further investigation. The obtained DNA was subjected to qualitative and quantitative analysis using a NanoDrop One spectrophotometer (Thermo Fisher Scientific; WI, USA). Each extracted DNA sample was diluted to 10 ng and used in all PCR reactions.

2.2. MALDI-TOF and PCR Identification

The species identification of E. coli strains was performed by MALDI-TOF mass spectrometry using an Ultraflex III instrument (Bruker Daltonics, Billerica, MA, USA). According to the manufacturer’s instructions [21], the supernatant was gained by extraction with ethanol and formic acid. Consequently, 1.0 µL of the supernatant was applied onto a MALDI plate. After drying, it was overlaid with a saturated solution of 4-hydroxy-α-cyanocinnamic acid in 50% acetonitrile and 2.5% trifluoroacetic acid in a volume of 1.0 µL. The results were processed using Flex Analysis software version 3.0 and assessed using BioTyper software version 1.1 (Bruker Daltonics; CA, USA).

The resulting score had a range of 1.90 to 2.41. A value score of 2.30 or higher indicated a species identification; a score of 2.0 to 2.29 indicated genus identification and probable species identification; 1.70 to 1.99 indicated probable genus identification; and lower than 1.69 indicated non-reliable identification.

Using primers that amplified the E. coli 16S rRNA gene, a species-specific PCR reaction was also used to identify the isolates of E. coli [22]. The total volume of the reaction mixture was 20 μL. The mixture consisted of one microliter of template DNA, ten microliters of either the commercially available DreamTaq Green PCR Master Mix (2X) (Thermo Scientific; Vilnius, Lithuania) or Hot FIREPol^® MasterMix (Amplia Ltd.; Bratislava, Slovak Republic), one microliter of each forward and reverse primer, in addition to nuclease-free water. Amplified DNA products were separated electrophoretically (Thermo Fisher Scientific; CA, USA and Major Science; OH, USA) on a 2% agarose gel with the addition of the fluorescent dye GoodViewTM (Amplia s.r.o.; Bratislava, Slovak Republic). A UV transilluminator was used to visualize and record DNA fragments (Major Science; CA, USA).

2.3. Antimicrobial Susceptibility Detection

Antimicrobial susceptibility was determined by means of minimum inhibitory concentration (MIC) testing using the modified microdilution method according to Gattringer et al. [23] using a commercial diagnostic kit (Bel-Miditech; Bratislava, Slovak Republic). Antimicrobial substances were tested in the following order: ampicillin (AMP); ampicillin + sulbactam (SAM); piperacillin + tazobactam (TZP); cefuroxime (CXM); cefotaxime (CTX); ceftazidime (CAZ); cefoperazone + sulbactam (SPZ); cefepime (FEP); ertapenem (ETP); meropenem (MEM); gentamicin (GEN); tobramycin (TOB); amikacin (AMI); tigecycline (TGC); ciprofloxacin (CIP); tetracycline (TET); colistin (COL); and trimethoprim + sulfonamide (COT). The Miditech software (Bel- MIDITECH s.r.o., Bratislava, SVK; cat. n. 002002) automatically generated the percentage of resistance to tested antimicrobial agents and also a probable mechanism of resistance in the investigated E. coli isolates. The EUCAST (version 14.0) [24] clinical breakpoints were used to interpret the results of the MIC values of each antimicrobial substance. In E. coli isolates, MIC xG represents the geometric mean MIC values of an antimicrobial agent (mg/L) described in our previous study [25].

2.4. Detection of Biofilm Formation

A modified colorimetric method, according to O’Toole et al. [26], was utilized to assess the biofilm formation of E. coli strains. A pure 24 h bacterial culture in 2 mL of sterile saline was used to prepare a suspension with a turbidity of 1.0 on the McFarland turbidity scale. Into each well of a 96-well microtiter plate were added 100 µL of modified Brain Heart Infusion broth (HiMedia Laboratories; Mumbai, India) and an equal volume of the prepared suspension. As positive and negative controls, the biofilm-forming Staphylococcus aureus CCM 4223 and the non-biofilm-forming Staphylococcus epidermidis CCM 4418 (both from the Czech Collection of Microorganisms) were chosen.

After a 24 h incubation period at 37 °C, the planktonic cell medium was removed, and distilled water was used to wash the wells. The biofilms were stained using a 0.1% crystal violet solution (Merck; Darmstadt, Germany). The dye, after incubation for 30 min at room temperature, was carefully removed from the wells. The wells were repeatedly washed with distilled water and then allowed to dry at room temperature. The residual dye was eliminated using 30% acetic acid. Finally, the optical density (OD) was determined using a Synergy reader (Merck; Darmstadt, Germany) at a wavelength of 550 nm.

For the interpretation of the results obtained, the strains were classified as non-biofilm producers (OD ≤ 0.150), weak biofilm producers (0.151 ≤ OD ≤ 0.300), moderate biofilm producers (0.301 ≤ OD ≤ 0.600) or strong biofilm producers (OD ≥ 0.601) according to the criteria described by Stepanović et al. [27] with modifications according to Ballén et al. [28].

2.5. Phylogenetic Typing of Strains

Triplex PCR was used to identify phylogenetic groups of E. coli isolates following the methodology of Doumith et al. [29]. Based on the presence or absence of the chuA and yjaA genes and the TspE4.C2 DNA fragment, the isolates were categorized into groups and subgroups, as reported by Escobar-Páramo et al. [30]. As reported by Clermont et al. [31], commensal E. coli strains belong to phylogenetic groups A and B₁. Pathogens are typically classified into phylogenetic group B₂, and somewhat less so into group D. The steps outlined in Section 2.2 were followed for the composition of the reaction mixture, the PCR reaction duration, and the visualization of the DNA fragments.

2.6. Detection of Genes Encoding Resistance to Antimicrobial Substances and Virulence Factors

Using simplex and multiplex PCR, genes encoding resistance to antimicrobial agents and virulence factors were found. The individual reaction PCR protocols can be summed up as follows. Following an initial denaturation at 94 °C/95 °C for 1 to 12 min, there were 25–34 cycles with the following parameters: denaturation for 30–60 s at 94 °C/95 °C, annealing for 30–60 s at 52–65 °C (based on primer sequence), and extension for 25 s–3 min at 68 °C/72 °C (dependent on the length of the amplified section). The last phase was to add a final extension step for two to ten minutes at 72 °C. The steps outlined in Section 2.2 were followed for the composition of the reaction mixture, the PCR reaction duration, and the visualization of the DNA fragments.

Table 1. List of primers used for detection of E. coli, phylogenetic analysis, and detection of genes of antimicrobial resistance and virulence factors.

Gene	Primer Sequences (5′–3′)	Annealing Temperature	Product Size	References
16S rRNA	GACCTCGGTTTAGTTCACAGA CACACGCTGACGCTGACCA	55 °C	585 bp	[22]
chuA	ATGATCATCGCGGCGTGCTG AAACGCGCTCGCGCCTAAT	65 °C	281 bp	[29]
yjaA	TGTTCGCGATCTTGAAAGCAAACGT ACCTGTGACAAACCGCCCTCA	65 °C	216 bp	[29]
TspE4.C2	GCGGGTGAGACAGAAACGCG TTGTCGTGAGTTGCGAACCCG	65 °C	152 bp	[29]
qnrA	ATTTCTCACGCCAGGATTTG GATCGGCAAAGGTTAGGTCA	57 °C	516 bp	[32]
qnrB	GATCGTGAAAGCCAGAAAGG ACGATGCCTGGTAGTTGTCC	57 °C	469 bp	[32]
qnrS	ACGACATTCGTCAACTGCAA TAAATTGGCACCCTGTAGGC	57 °C	417 bp	[32]
tetA	GGCCTCAATTTCCTGACG AAGCAGGATGTAGCCTGTGC	54 °C	372 bp	[33]
tetB	GAGACGCAATCGAATTCGG TTTAGTGGCTATTCTTCCTGCC	54 °C	228 bp	[33]
Int1	GGGTCAAGGATCTGGATTTCG ACATGCGTGTAAATCATCGTCG	62 °C	483 bp	[34]
Int2	CACGGATATGCGACAAAAAGGT GTAGCAAACGAGTGACGAAATG	62 °C	788 bp	[34]
Tn3	CACGAATGAGGGCCGACAGGA ACCCACTCGTGCACCCAACTG	62 °C	500 bp	[35]
kpsMT II	GCGCATTTGCTGATACTGTTG CATCCAGACGATAAGCATGAGCA	63 °C	272 bp	[36]
fimA	GGCGAATTCTGTTCTGTCGGCTCTGTC TTGGAATTCAACCTTGAAGGTCGCATC	52 °C	510 bp	[37]
papC	GACGGCTGTACTGCAGGGTGTGGCG ATATCCTTTCTGCAGGGATGCAATA	65 °C	328 bp	[38]
iutA	GGCTGGACATCATGGGAACTGG CGTCGGGAACGGGTAGAATCG	63 °C	300 bp	[39]
cvaC	CACACACAAACGGGAGCTGTT CTTCCCGCAGCATAGTTCCAT	63 °C	680 bp	[36]

Description: 16S rRNA = E. coli identification; chuA, yjaA, TspE4.C2 = phylogenetic grouping; qnrA, qnrB, qnrS = quinolone resistance; tetA, tetB = resistance to tetracycline; Int1, Int2 = integron; Tn3 = transposon; kpsMT II = capsule; fimA = type 1 fimbriae; papC = P fimbriae; iutA = receptors for aerobactin; cvaC = colicine V.

contains a list of all primers used in each PCR reaction.

The presence of genes encoding quinolone resistance—qnrA, qnrB and qnrS—and tetracycline resistance—tetA and tetB—was investigated. Furthermore, the prevalence of genes encoding mobile genetic elements (MGEs)—integrons (Int1 and Int2) and transposons (Tn3)—was assessed. These MGEs have the ability to capture and express genes in gene cassettes, thereby significantly increasing antimicrobial resistance in Gram-negative bacteria [40].

Moreover, we evaluated the presence of genes encoding the following virulence factors: capsule (kpsMT II), type 1 fimbriae (fimA), P-fimbriae (papC), receptors for aerobactin (iutA), and colicine V (cvaC).

3. Results

Overall, one hundred and five E. coli strains were isolated from chicken cloacal swabs (n = 65) and thigh muscle meat from a commercial network (n = 40) after microbiological analysis and species identification by MALDI-TOF MS and by PCR, with positive detection of the 16S rRNA gene.

3.1. Antimicrobial Susceptibility Profile

The antimicrobial testing results show visible differences between the groups under investigation. The percentages of antimicrobial resistance shown by all 105 E. coli strains isolated from samples of cloacal swabs from broiler chickens and samples of chicken thigh muscle meat purchased from a commercial network are compared in Figure 1. The E. coli strains also exhibited diverse probable antimicrobial resistance mechanisms that are responsible for the resistance phenotype exhibited (Figure 2).

E. coli isolates obtained from chicken cloacal swabs frequently showed resistance to AMP, with a rate exceeding 60%. Additionally, over 32% of the tested isolates displayed resistance to TET and COT. Resistance to SAM, CIP, and CAZ was observed in 3% of isolates examined. We also observed an increased level of intermediate susceptibility to CIP (nearly 22%) and COT (>7%). Compared to the EUCAST clinical breakpoint [24] for AMP (MIC > 8 mg/L), our MIC level was higher (MIC > 18 mg/L). A penicillinase with low expression was revealed as the most common mechanism of resistance by interpretive reading. To a lesser extent, incomplete resistance to quinolones was also found. Penicillinase with a high expression of genes and multiresistance mechanisms were present in two and one isolates.

Resistant E. coli isolated from thigh muscle samples of chickens showed the highest resistance to AMP (55%), CIP (45%), and SAM, TET, and COT (30%). Our MIC levels for AMP (MIC > 18 mg/L) and CIP (MIC > 1 mg/L) were higher compared with the EUCAST clinical breakpoint [24] for AMP (MIC > 8 mg/L) and CIP (MIC > 0.5 mg/L). An elevated level of intermediate susceptibility to CIP was discovered in more than 38% of the isolates. The most prevalent resistance mechanism was incomplete resistance to quinolones, followed by penicillinase with high expression and penicillinase with low expression.

According to Alkofide et al. [41], resistance to three or more antimicrobial agents from different groups is known as multi-drug resistance (MDR). Of all the strains studied, we confirmed MDR in 28.57% of E. coli isolates.

Overall, the strains located on TZP, CXM, CTX, SPZ, FEP, ETP, MEM, GEN, TOB, AMI, TGC and COL had 100% sensitivity.

3.2. Evaluation of Biofilm Formation

An important virulence characteristic of E. coli is its ability to form biofilms. A biofilm formation test was conducted on 34 E. coli strains (23 strains obtained from chicken cloacal swab samples and 11 strains from chicken thigh muscle meat). The strains were chosen based on their resistance to at least one antimicrobial, as indicated by the MIC results, or the system automatically generated a likely resistance mechanism for them.

Our results show that 88.23% of the E. coli strains tested were unable to form biofilms. The formation of weak biofilms was confirmed in three isolates that were obtained from chicken cloacal swabs (n = 2) and chicken thigh muscle meat (n = 1). Moderate biofilm formation was not demonstrated by any E. coli strain. In addition, one of the strains obtained from chicken cloacal swabs was categorized as a strong biofilm producer.

3.3. Phylogenetic Analysis

Table 2. Classification of isolates into phylogenetic groups and subgroups.

	Phylogenetic Groups and Subgroups
Number (%) of Samples	A0	A1	B1	B22	B23	D1	D2
Chicken cloacal swabs (n = 23)	0 (0)	6 (26.1%)	6 (26.1%)	2 (8.7%)	1 (4.3%)	1 (4.3%)	7 (30.5%)
Chicken thigh muscle meat (n = 11)	1 (9.1%)	3 (27.3%)	4 (36.4%)	0 (0)	0 (0)	1 (9.1%)	2 (18.2%)

summarizes the percentage representations of phylogenetic groups and subgroups in the selected E. coli isolates.

The phylogenetic typing of the obtained E. coli isolates shows that they were most often categorized into groups A₁ (26.1 and 27.3%), B₁ (26.1 and 36.4%), and D₂ (30.5 and 18.2%), followed by phylogenetic subgroups D₁, A₀, B2₂, and B2₃.

MDR E. coli obtained from chicken cloacal swabs and chicken thigh muscle meat was categorized into phylogenetic groups and subgroups, specifically D₂, B₁, A₁, D₁, and A₀, listed in decreasing order.

3.4. Genotype Resistance to Antimicrobial Agents

The resistance of E. coli strains to quinolones and tetracyclines is increasing globally every year. Therefore, in our study, we focused on monitoring genetic determinants encoding resistance to quinolones (qnrA, qnrB, and qnrS) and tetracyclines (tetA and tetB). We confirmed the presence or absence of these genes by the PCR method. A genotypic analysis was conducted on 34 E. coli strains—23 strains obtained from chicken cloacal swab samples and 11 strains from chicken thigh muscle meat.

Our results reveal that 26.5% of E. coli isolates were qnrA-positive (14.7% = cloacal swab samples and 11.8% = chicken thigh muscle meat); one isolate was qnrB-positive; and 14.7% were qnrS-positive (only from cloacal swab samples). The simultaneous prevalence of qnrA and qnrB genes was detected in one isolate from chicken thigh muscle meat. More than 52% of the isolates studied were negative for qnr genes.

The tetA gene was present in 70.6% of the E. coli isolates, and it was identified in 50.0% of the chicken swab samples and 20.6% of the chicken thigh muscle meat. One sample from each of the two groups showed the presence of the tetB gene. We observed the simultaneous presence of both genes in 8.8% of the examined isolates. The absence of genes was detected in more than 14% of E. coli isolates.

3.5. Presence of Mobile Genetic Elements and Virulence Factors

We confirmed that mobile genetic elements mediated antimicrobial resistance in the selected E. coli isolates. Integron cassette class 1 (Int1) was detected in 52.9% of E. coli strains (41.2% = cloacal swabs, 11.7% = chicken thigh muscle meat). We did not detect the presence of an integron class 2 (Int2) cassette in any isolate examined. Additionally, we noted that 76.5% of the isolates under investigation had transposon Tn3 genes (61.8% = cloacal swabs, 14.7% = chicken thigh muscle meat). The prevalence of both genes Int1 and Tn3 simultaneously was found in 52.9% of E. coli isolates (41.2% = cloacal swabs, 11.7% = chicken thigh muscle meat).

Table 3. Distribution of resistance genes, virulence factors, and biofilm formation in commensals and pathogens in selected E. coli strains.

Genetic Determinant	Cloacal Swabs (n = 23)		Chicken Meat (n = 11)
	Commensals	Pathogens	Commensals	Pathogens
Isolates with Virulence Genes	11 (n = 12)	11 (n = 11)	8 (n = 8)	3 (n = 3)
qnrA	4	1	4	–
qnrB	1	–	–	–
qnrS	2	3	–	–
qnrAB	–	–	–	1
tetA	9	8	6	2
tetB	1	2	–	1
tetAB	–	2	–	1
Int1	5	9	3	1
Int2	–	–	–	–
Tn3	10	11	4	1
kpsMT II	1	1	–	1
fimA	11	11	8	3
papC	–	–	–	–
iutA	7	11	6	2
cvaC	3	4	4	2
biofilm formation	2	1	–	1

Description: n = number of isolates. Abbreviations of genetic determinants are mentioned in the “Materials and Methods” section.

shows the distribution of resistance genes, virulence factors, and biofilm formation in commensals and pathogens in selected E. coli strains.

Here, 97.1% of the selected 34 E. coli strains were found to be potentially pathogenic, meaning they carried one or more of the virulence genes. We found that the most prevalent genes among the examined isolates were fimA (97.1%), followed by iutA (76.5%), cvaC (38.2%), and kpsMT II (8.8%). The presence of the papC gene in E. coli isolates was not detected.

The antimicrobial resistance and pathogenicity profiles of the selected E. coli strains are summarized in

Table 4. Comprehensive evaluation of E. coli strains: resistance, biofilm activity, phylogenetic typing, mobilome, and virulence factors.

No. of Strain	Phenotypic Resistance	Resistance Mechanism	Biofilm Formation	Phylogenetic Typing	Genotypic Resistance	Mobile Genetic Elements	Virulence Factors
E. coli from chicken cloacal swabs
1C	AMP	–	Non-biofilm	A1	tetA	Tn3	fimA, iutA
2C	–	Quinolone Incompl. Resistance	Weak	B1	qnrA	–	fimA, cvaC, iutA
3C	AMP	Quinolone Incompl. Resistance	Non-biofilm	A1	tetA	Tn3	fimA, iutA
4C	AMP	–	Non-biofilm	B23	tetA, tetB	Int1, Tn3	fimA, iutA
5C	AMP	–	Non-biofilm	A1	tetA	Tn3	fimA, iutA
6C	AMP, TET, COT	Penicillinase:low	Non-biofilm	D2	tetA	Int1, Tn3	fimA, cvaC, iutA
7C	AMP	–	Non-biofilm	D2	qnrA	Tn3	fimA, cvaC, iutA
8C	AMP	–	Non-biofilm	A1	tetA	Tn3	fimA, iutA
9C	AMP, TET	Penicillinase:low	Non-biofilm	D2	tetA	Int1, Tn3	fimA, iutA
10C	AMP, COT	Penicillinase:low	Non-biofilm	B1	tetB, qnrA	Int1, Tn3	fimA, cvaC, iutA
11C	AMP	Penicillinase:low	Non-biofilm	B1	tetA, qnrA	–	fimA
12C	AMP, TET, COT	–	Weak	D2	tetA, qnrS	Int1, Tn3	fimA, cvaC, iutA
13C	AMP, TET, COT	Penicillinase:low	Non-biofilm	D2	tetA	Int1, Tn3	fimA, iutA
14C	AMP, TET, COT	–	Non-biofilm	B1	tetA, qnrS	Int1, Tn3	fimA
15C	AMP, TET	Penicillinase:low	Non-biofilm	D2	tetA, qnrS	Int1, Tn3	fimA, cvaC, iutA
16C	AMP, TET, COT	Penicillinase:low; Quinolone Incompl. Resistance	Non-biofilm	A1	tetA, qnrB	Int1, Tn3	–
17C	AMP, TET, COT	Penicillinase:low	Non-biofilm	D2	tetA	Int1, Tn3	fimA, iutA
18C	COT	Quinolone Incompl. Resistance	Non-biofilm	B22	tetA, tetB, qnrS	Int1, Tn3	kpsMT II, fimA, iutA
19C	AMP	Quinolone Incompl. Resistance	Non-biofilm	B22	tetA	Tn3	fimA, iutA
20C	AMP, SAM, TET, COT	Quinolone Incompl. Resistance	Non-biofilm	B1	tetA, qnrS	Int1, Tn3	fimA
21C	AMP, TET	Penicillinase:low	Strong	A1	tetA	Tn3	kpsMT II, fimA
22C	AMP, CAZ, CIP, TET, COT	Multiresistance!	Non-biofilm	D1	tetA	Int1, Tn3	fimA, iutA
23C	AMP, COT	Penicillinase:high!	Non-biofilm	B1	qnrA	Int1, Tn3	fimA, cvaC, iutA
E. coli from chicken meat samples
1M	AMP, SAM	Quinolone Incompl. Resistance	Non-biofilm	A1	tetA	–	fimA, cvaC, iutA
2M	–	Quinolone Incompl. Resistance	Non-biofilm	A1	tetA	–	fimA, cvaC, iutA
3M	–	Quinolone Incompl. Resistance	Non-biofilm	A1	tetA	–	fimA, cvaC, iutA
4M	AMP, SAM, CIP, TET, COT	–	Non-biofilm	B1	tetA, qnrA	Int1, Tn3	fimA, iutA
5M	AMP, CIP, TET, COT	–	Non-biofilm	A0	tetA	Int1, Tn3	fimA
6M	AMP, SAM, CIP	–	Non-biofilm	B1	qnrA	Tn3	fimA
7M	AMP, CIP	Penicillinase:low	Weak	D2	tetA, qnrA, qnrB	–	kpsMT II, fimA, cvaC, iutA
8M	AMP, CIP, TET, COT	Penicillinase:high!	Non-biofilm	D2	tetB	Int1, Tn3	fimA, iutA
9M	–	Quinolone Incompl. Resistance	Non-biofilm	D1	tetA, tetB	–	fimA, cvaC
10M	–	Quinolone Incompl. Resistance	Non-biofilm	B1	qnrA	–	fimA, cvaC, iutA
11M	AMP, SAM, CIP, TET, COT	–	Non-biofilm	B1	tetA, qnrA	Int1, Tn3	fimA, iutA

Description: 1C–23C = E. coli from chicken cloacal swabs; 1M–11M = E. coli from chicken meat samples. Abbreviations of antimicrobial substances and genetic determinants are mentioned in the “Materials and Methods” section.

, based on the results of the analyses performed.

4. Discussion

Global public health issues include the safety of manufactured food, the transmission of microorganisms, the incidence of foodborne illness, and, in recent years, antimicrobial resistance too. Administering antimicrobial substances to food-producing animals can improve health and increase productivity during intensive animal production. At the same time, it can contribute to the emergence and spread of resistance to these medications [42]. Since January 2022, the European Union has strengthened the rules for fighting against antimicrobial resistance. The new legislation significantly restricts the prophylaxis and metaphylaxis use of antimicrobials. The antimicrobial treatment of animals will be administered only in necessary cases [43].

Poultry meat plays a significant role in people’s nutrition, and its production and consumption are increasing rapidly around the globe [1]. In the present study, we compared the antimicrobial resistances of E. coli isolates from broiler chicken cloacal swabs and purchased chicken thigh muscle meat. We found differences in the percentages of resistance, the occurrence of antimicrobial-resistance genes, biofilm formation, and phylogenetic groups of E. coli strains.

By antimicrobial susceptibility testing, we found that E. coli isolates from cloacal swabs and meat samples have dominant resistance to ampicillin. Similar to the other studies, most avian E. coli isolates were here found to be highly resistant to ampicillin [44,45]. Ampicillin (AMP) belongs to the semisynthetic β-lactams and is widely used to treat E. coli infections in humans and animals, where it acts on the active replication stage of the bacteria and inhibits the synthesis of the bacterial cell wall. In recent years, the rate of resistance to ampicillin has increased [46].

We also confirmed increased resistance to tetracycline, trimethoprim–sulfonamide, ciprofloxacin, and ampicillin–sulbactam. Based on our results, we can conclude that the presence of E. coli resistant to several antimicrobials in chicken swab samples and chicken meat has several important implications for health in our country. The presence of resistant E. coli in chickens can significantly affect animal health, production losses and the safety of the food produced, as well as increasing veterinary costs. Resistant strains of E. coli can spread between chicken populations and other animals and humans, raising public health concerns. Resistant E. coli can spread through environmental routes, such as manure, which affects ecosystems and can contribute to the spread of resistance. The consumption of contaminated chicken can lead to infections resistant to commonly prescribed antimicrobials, complicating treatment and increasing health risks and healthcare costs.

The research conducted by Parvin et al. [47] and Alam et al. [48] supports our findings, as they reported a high resistance rate (60.7–89.5%) among E. coli strains from chicken samples to trimethoprim + sulfonamide, tetracycline, ampicillin, and ciprofloxacin. Many isolates studied were also resistant to tetracycline, doxycycline, nalidixic acid, amoxicillin–clavulanate, ciprofloxacin, trimethoprim + sulfonamide, chloramphenicol, gentamicin, azithromycin and erythromycin [44,45]. Caruso et al. [49] reported that β-lactams, tetracyclines, sulfonamides, and quinolones are the most often utilized antimicrobials in global broiler chicken production.

According to Pormohammad et al. [50], multidrug-resistant E. coli bacteria are more frequent in animal samples than in human isolates. Poultry farms and their waste, which includes manure and wastewater, are a common source of antimicrobial-resistant and multidrug-resistant bacteria, as well as antimicrobial residues. These elements create a constant source of selection pressure that drives the evolution of resistance in bacterial populations [48].

Here, 60.95% of the one hundred and five E. coli isolates tested were found to be resistant to at least one antimicrobial agent in our experiment. Multidrug-resistant strains together represented 28.57%, while the most common MDR profile was resistant to three different antimicrobials: AMP, CIP and TET. Studies by other authors have shown significantly higher rates of MDR E. coli in chicken samples in Sri Lanka (82.6%) [51], Egypt (97.6%) [45], and Bangladesh (100%) [47]. These results are alarming because MDR bacteria are the cause of difficult-to-treat infectious diseases in both human and veterinary medicine [48].

Bacterial biofilm formation poses a difficulty in many areas, e.g., food processing facilities, food-producing animals, and veterinary practice. A significant problem in the food chain is the creation of biofilms by MDR bacteria [52]. Biofilms consist of microbial communities that irreversibly adhere to surfaces and are encased in an extracellular matrix. The matrix provides protection for microorganisms against unfavorable environmental conditions, and promotes the sharing of antimicrobial-resistant genes via horizontal gene transfer [53].

According to Rodrigues et al. [54], most E. coli strains of avian origin form biofilms, which increases the persistence of these pathogens on chicken farms. In our study, the majority of E. coli isolates did not form a biofilm, but we confirmed weak and strong biofilm formation. Our results indicate that the risk to the consumer associated with the presence of potentially highly pathogenic strains of E. coli is relatively low. Similarly, in the study by Barilli et al. [55], the majority of E. coli strains obtained from Italian retail chicken meat showed weak biofilm activity, while the E. coli isolated from beef samples exhibited strong biofilm formation. In research conducted in China, Wang et al. [56] reported that 81.6% of E. coli isolated from poultry meat formed biofilms, with 25.4% being strong producers. Menck-Costa et al. [57] reported that the biofilm formation of E. coli acquired from chicken samples was categorized as moderate in 55.0%, weak in 22.9%, strong in 12.8%, and very strong in 2.8% of the cases.

E. coli strains can be classified into four phylogenetic groups, according to research conducted by Clermont et al. [31]. Pathogenic strains belong to groups B₂ and D, whereas the commensal microbiota consists of groups A and B₁. E. coli isolates from broiler chicken cloacal swabs and meat samples were analyzed and classified into all four phylogenetic groups. A and B₁ were the groups that dominated, followed by D and B₂. E. coli commensal strains frequently exhibit multidrug resistance. In the context of One Health, their share increases annually, and adds to the growing antimicrobial resistance [58].

Several studies confirm that food-producing animals, including poultry, are reservoirs of E. coli and Klebsiella spp., which are producers of ESBLs. Nowadays, not only ESBL genes, but also other antimicrobial resistance genes, mainly qnr and tet genes, are increasingly being detected [48,59].

The qnrABS genes encode proteins that protect E. coli DNA gyrase from inhibition by quinolone, even at low drug concentrations [59]. Our research has revealed that most strains with qnr genes present did not have phenotypic resistance to ciprofloxacin. In E. coli isolates, the qnrB gene was present in only one isolate, whereas the qnrA (26.5%) and qnrS (14.7%) genes were more frequently present. In one isolate, both qnrA and qnrB genes were present. Our findings are consistent with those of Kurnia et al. [60], who detected qnrA (61.6%), qnrB (7.7%), and qnrS (23.0%) genes in APEC isolates. However, Ahmed et al. [61] found aac(6′)-Ib-cr, qnrA, and qnrB in addition to a 15.1% abundance of qnrS. Similar to this, in avian pathogenic E. coli isolates, Seo and Lee [62] discovered all four genes: qnrS (9.4%), qnrA (6.6%), qnrB (3.8%), and aac(6′)-Ib-cr (1.9%). Rodríguez-Martínez et al. [63] reported that qnrABS-positive plasmids often contain genes that also encode resistance to β-lactams, tetracyclines, sulfonamides, trimethoprim, aminoglycosides, chloramphenicol, and rifampicin.

E. coli’s resistance to tetracycline involves an efflux mechanism, which is controlled by as many as nine genes—tetA, tetB, tetC, tetD, tetE, tetG, tetJ, tetL, and tetY. [15,60]. Our research focused on the detection of tetA and tetB genes, which are among the most commonly found tet genes in Enterobacteriaceae bacteria. Alam et al. [48] found that tetA was the most abundant gene in E. coli obtained from chicken samples, similar to our study. Van et al. [64] also reported that the tetA gene was the most prevalent tetracycline resistance gene (71.1% of isolates), followed by tetB (18.4%), in poultry in Vietnam. The most prevalent tetB gene was found in avian pathogenic E. coli isolated from broiler, layer, and breeder chickens in Nepal [65]. The present study confirms the increasing prevalence of tet and qnr genetic determinants among E. coli isolates of avian origin. According to Al-Bahry et al. [66] and Richer et al. [67], their widespread distribution among bacterial populations is due to the association of tet or qnr genes with mobile genetic elements.

Mobile genetic elements (MGEs) are often involved in horizontal gene transfer (HGT) processes in bacteria, and facilitate their rapid evolution and adaptation. Kocúreková et al. [68] found an increased abundance of MGEs (Int1 and Tn3) in broiler chicken samples, which is compatible with our findings. We detected the class 1 integron cassette (Int1) in more than half of the E. coli strains. The majority of the isolates studied had Tn3 transposon genes. The presence of both Int1 and Tn3 genes simultaneously was found in 52.9% of E. coli isolates.

E. coli strains of avian origin may contain various virulence factors such as adhesins, invasins, toxins, iron uptake systems, and protectins that highlight their zoonotic potential [69]. In our study, we also confirmed the presence of four of the five tested genes associated with virulence in isolates of avian E. coli, namely, fimA (97.1%), iutA (76.5%), cvaC (38.2%), and kpsMT II (8.8%). The presence of the papC gene in E. coli isolates was not detected. The virulence genes fim (type 1 fimbriae) and pap (P-fimbriae) are colonization factors in extra-intestinal infections, and they contribute to biofilm formation. The iutA gene encodes an aerobactin receptor, which captures iron to support bacterial growth and development in iron-deficient conditions. CvaC represents a colonization-facilitating factor, and kpsMT II is a protective factor against phagocytosis, as well as being a spreading factor [70]. One of the pathogenic E. coli strains (D₂) originating from chicken thigh muscle contained kpsMT II, fimA, cvaC, and iutA genes, as well as tetA, qnrA, and qnrB genes, had no MGEs present, and had weak biofilm formation. While it did not produce a biofilm, one of the pathogenic E. coli strains (B2₂) obtained from chicken cloacal swab samples did contain the genes kpsMT II, fimA, and iutA in addition to tetA, tetB, qnrS, integron 1, and transposon 3. The kpsMT II, fimA, tetA, and Tn3 genes were also found in the E. coli commensal strain (A₁), which was isolated from chicken cloacal swab samples and had a high biofilm formation potential. Kocúreková et al. [68] discovered six genes associated with virulence factors in E. coli isolates from broiler chicken gut microbiota. The genes are kpsMT II (30.4%), iss (29.6%), cvaC (21.7%), tsh (9.6%), papC (4.3%), and ibeA (3.5%). In a study conducted by Savin et al. [69], they also identified genes encoding virulence factors in E. coli isolated from poultry and swine slaughterhouses in Germany. They found that almost all isolates carried fimH (encoding fimbriae type 1), which is also in agreement with our results. A high percentage of the isolates studied had astA, iutA, iroN, and fyuA genes. In contrast to us, they also detected the presence of papCEFG genes. One isolate from a pig abattoir had the hlyD and cnf1 genes present. The prevalence of kpsMT II genes was higher in isolates from poultry abattoirs compared to isolates from pig abattoirs (22.9% versus 5.6%).

Several researchers suggest that the presence of E. coli with genes encoding virulence factors that facilitate the development of human ExPEC infections in foods of animal origin, especially poultry, is now an important food safety issue that requires continuous monitoring [71].

5. Conclusions

Antimicrobial resistance arises and spreads in large part due to conditions on farms and in slaughterhouses. According to our findings, chickens raised for food production can be considered reservoirs of potential pathogenic E. coli strains. The presence of E. coli in meat and meat products is not only a sign of improper food handling or hygiene procedures, but these bacteria also carry virulent genes that have the potential to cause a variety of intestinal and extra-intestinal diseases in humans. Although processing meat can lessen the risk of antimicrobial-resistant bacteria (such as E. coli), cross-contamination in the food environment may still pose a threat to consumers. Furthermore, the capacity to create a biofilm may increase the possibility of antimicrobial-resistance transmission through dietary consumption.

Finally, by limiting the use of antimicrobial agents in animal husbandry and adhering to good manufacturing and hygiene practices during the production, transportation, and preparation of food, humans themselves can have an impact on the spread of antibiotic resistance in the future.