The Escherichia coli Sequence Type 131 Harboring Extended-Spectrum Beta-Lactamases and Carbapenemases Genes from Poultry Birds

Introduction

The extended-spectrum beta-lactamases (ESBLs) as well as carbapenemases are increasingly being spreading to members of the Enterobacteriaceae family, which results in limiting the therapeutic options for the management of infections caused by these organisms.^1–3 Escherichia coli strains that harbor these enzymes have been reported to colonize healthy human beings as well as animals such as poultry birds.⁴ It is widely accepted that the increased and indiscriminate use of antimicrobial drugs for humans and animals is the main driving force for the increasing colonization and infections caused by ESBL producing E. coli strains.^5–7

The gut colonization by the ESBL producing E. coli (ESBL-Ec) and carbapenemase-producing E. coli (CPEC), is an important feature for the spread of these multidrug resistant (MDR) bacteria among livestock especially poultry.^4,8 Studies have widely reported the presence of ESBL-producing E. coli (ESBL-EC) strains in the food-producing animals, especially poultry, which poses a serious threat to public health.^9,10 ESBL producing MDR strains have also been reported from European countries such as Netherlands and Germany from poultry birds despite the controlled use of antibiotics in poultry farming.^11,12 The ESBL-Ec in poultry farms has also been reported from Asian countries, such as China, Japan, Thailand, and Vietnam, although, the prevalence of ESBL-Ec strains varied with the sampling strategy and the study catchment area.^13–16 Moreover, the clonal dissemination of ESBL-Ec strains among the poultry and surrounding environments, as well as poultry-human transmission, has also been reported.^16,17 Despite the carbapenems are not approved for use in livestock according to the Animal Medicinal Drug Use Clarification Act (1994) in the United States and similar guidelines in many other countries. However, studies have demonstrated the spread of E. coli harboring these resistance determinists among pets or livestock and to exposed humans.^18,19

In Pakistan, the few published studies have confirmed the occurrence of ESBL-Ec isolates from poultry.^20,21 Most studies are limited to a single large poultry farm and no data is available regarding the sequence types of ESBLs and carbapenemases producing E. coli. The extent of the spread of ESBLs and carbapenemase among poultry birds can be helpful to determine the implication of these potential reservoirs as a source of transmission of such MDR strains to humans. Therefore, it is important to screen the poultry for the colonization of ESBLs and carbapenemase-producing strains in poultry birds. Therefore, the present study was designed to explore the occurrence of the sequence types ESBLs and carbapenemase-producing E. coli from the poultry farms of the capital region of Pakistan.

Materials and Methods

Ethical Approval

The research proposal was approved by the Ethical Review Committee (ERC), Government College University, Faisalabad, Pakistan, with reference number GCUF/ERC/17/16. Furthermore, the prior consent to collect poultry samples was taken from farm owners individually.

Samples and Bacterial Strains

The current study was conducted in Islamabad, the capital territory of Pakistan, between February 2018 to June 2019 at 25 poultry farms in and around Islamabad, Pakistan. The poultry farms were having around 500–1500 chickens on each farm and infrastructure was comparable in all the poultry farms. The 250 chicken ceca samples (10 samples from each farm) were collected and transported to the laboratory within 2 hours.

Screening for ESBL and Carbapenemases Producing E. coli

The cecal material (1 gram) was suspended in 9 mL sterile saline (0.9%) and mixed by gentle vortex for 30–45 seconds. A small amount (100μL) of this cecal saline suspension was inoculated on the commercial HiCrome™ ESBL Agar Base (HiMedia, India) and MacConkey agar comprising cefotaxime (4 μg/mL) for screening the ESBL producing E. coli. For the screening of carbapenem-resistant E. coli, the MacConkey agar supplemented with imipenem (1 μg/mL) was used. The plates were aerobically incubated at 37°C for 24 hours.

The isolated colonies were sub-cultured on nutrient agar for phenotypic identification, preservation, and subsequent antimicrobial susceptibility testing. The species-level identification was done by standard microbiological methods and further confirmed by the API 20E kit (BioMérieux, France). All isolates including the ESBL-Ec and CPEC were preserved in brain heart infusion (BHI) broth and glycerol in cryovials and stored at −20°C.

Phenotypic Detection of ESBL and Carbapenemase Production

For the detection of ESBL production, the double-disc synergy test was performed for all isolates using the combination disc test (CDT) according to the CLSI guidelines. Briefly, the test discs containing cephalosporin a (cefotaxime and ceftazidime) alone and the combination of cephalosporins in combination with clavulanic acid were applied on the surface of Muller Hinton agar (Oxoid, UK). The ESBL production was considered positive in case, the zone of inhibition for the cephalosporin-clavulanic acid disc is ≥ 5 mm larger than the zone of the cephalosporin disc alone. The isolates were further screened for the carbapenemase production by the modified Hodge test as described previously.²²

Antibiotic Susceptibility Testing

The antibiotic susceptibility testing of the ESBL positive isolates was performed by the Kirby Bauer disc diffusion method as per CLSI guidelines and interpreted accordingly.²³ The following antimicrobial agents were used; cefuroxime, ceftazidime, cefotaxime, cefixime, ceftriaxone, cefepime, ampicillin-clavulanic acid, and piperacillin-tazobactam, imipenem and meropenem, aztreonam, ciprofloxacin, amikacin and gentamicin, and trimethoprim-sulfamethoxazole. The E. coli (ATCC^® 25922) and K. pneumoniae (ATCC^® 700603) were used for quality control. The antimicrobial disks used in the study were obtained from Oxoid (Oxoid, UK).

The minimal inhibitory concentrations (MICs) of cefuroxime, cefotaxime, cefepime as well as imipenem, and colistin were determined by the broth microdilution method according to the Clinical and Laboratory Standards Institute (CLSI, 2018) guidelines. Pseudomonas aeruginosa (ATCC^® 27853) and E. coli (ATCC^® 25922) and were used for quality control.

Molecular Characterization of β-Lactamase Encoding Genes

All the 156 isolates were also screened for the ESBL encoding genes including bla_CTX-M, bla_TEM, and bla_SHV genes. The isolates positive for bla_CTX-M were further screened for the bla_CTX-M-1, bla_CTX-M-2, bla_CTX-M-8, bla_CTX-M-9, and bla_CTX-M-25. The presence of class B beta-lactamases such as bla_IMP, bla_VIM, and bla_NDM was also detected by PCR assays for all the 156 isolates. The primers (Macrogen, South Korea) were used and the expected size of PCR products for these enzymes are shown in Table 1. For the DNA extraction, a commercially available DNA extraction kit FavorPrep™ (FAVORGEN Biotech Corporation, Taiwan) was used. The PCR reactions (30 μL) was used using 15 μL 2X DreamTaq Green master mix (Thermo Fisher Scientific, Massachusetts, USA), 1 μL of each primer (forward and reverse) were used and 1 μL of the template DNA was used. The PCR water was used to make a total volume of 30 μL. The agarose gel (1.5%) was used to visualize the amplicons after electrophoresis at 100V for 30–45 minutes. All the amplified samples were sent for Sanger sequencing from Macrogen, Inc (South Korea) The obtained sequences were aligned and compared with the sequences in the GenBank database using the NCBI BLAST tool.

Table 1 Primers and Annealing Temperature for the Amplification of ESBLs and Carbapenemases Genes

Multi-Locus Sequence Typing (MLST)

For the MLST, seven housekeeping genes (adk, fumC, gyrB, icd, mdh, purA, and recA) were amplified and sequenced according to the conditions provided by the EnteroBase Database.²⁴ The obtained PCR products were sent for sequencing to Macrogen (South Korea). The raw sequences were edited using the ChromasPro software (Technelysium, Australia). The sequences were aligned using the ClustalW algorithm (MEGA7 software). The allele number was assigned to each gene locus and the sequence types (STs) were determined according to the allelic profiles for each of the isolates.

Results

Prevalence of ESBL and Carbapenemases Producing E. coli

A total of 156 (62.4%) strains were positive for ESBL production. Among these 156 strains, 119 strains were positive for ESBLs only whereas a total of 37 (14.8%) strains were found positive for carbapenemases production in addition to ESBL production.

Antibiotic Resistance Profile

The data obtained from the antibiotic sensitivity testing obtained by disc diffusion tests have been shown in Table 2. The ESBL-Ec isolates showed the highest resistance to sulfamethoxazole-trimethoprim (83%) followed by ciprofloxacin (74.4%) and aminoglycosides ie gentamicin (45%) and amikacin (44%). 37 (23.7%) isolates were found resistant to carbapenems (imipenem and meropenem). All (n=156) ESBL-Ec were found susceptible to colistin.

Table 2 Overall Distribution of MICs of Various β-Lactams Agents Against Escherichia coli (n=156) Strains

Diversity of Beta-Lactamases Genes

All one hundred and fifty-six isolates were screened for the ESBL gene and all were found positive for the bla_CTX-M gene. Overall, the bla_CTX-M-1 was found in 131 isolates followed by bla_CTX-M-9 which was found in 27 isolates. The bla_CTX-M-2 and bla_CTX-M-9 were found from 19 isolates. The bla_TEM was the second most common ESBL gene which was found in 71 isolates, whereas the bla_SHV was found in 28 isolates. Many of the strains were found to co-harbor multiple ESBL genes. Among the carbapenemases genes, the bla_VIM was the most common gene and was found in all the 37 carbapenem-resistant isolates whereas the bla_NDM gene was found in 31/37 (83.8%) isolates and bla_IMP were found in 16/37 (43.2%) isolates.

Sequence Types of ESBL-Ec

The analysis of MLST results has shown that the 48 E. coli strains were classified into 17 different sequence types (STs) including the 10 already reported STs and seven new STs. The ST131 (22/48, 46%) was the most common ST, followed by ST8051 (10/48, 21%). The ST8900 and ST2847 accounted for 4%, whereas the STs including ST8420, ST8431, ST2741, ST3499, and ST6293 accounted for 2% (1/48) (Table 3). The ST131 was found to harbor the ESBL (bla_CTX-M-1, bla_CTX-M-2, bla_CTX-M-8, bla_CTX-M-9 bla_TEM, bla_SHV) and carbapenemases (bla_VIM and bla_NDM) genes. The ST8051 was harboring the bla_CTX-M-1, bla_SHV, bla_IMP, and bla_VIM gene combinations.

Table 3 Distribution of Various Genetic Determinants Among Different Escherichia coli Sequence Types (STs) Isolated from Poultry in Islamabad, Pakistan

Discussion

Antimicrobial resistance is a serious global issue; however, in developing countries such as Pakistan, the indiscriminate use of antimicrobials especially in veterinary settings has worsened the situation.^25–27 The emergence of ESBLs and carbapenemases among the Enterobacteriaceae family in the food-producing animals can impact the consumer’s health as there is growing evidence supporting the hypothesis that the animals can be a source of antimicrobial resistance and can be involved in the spread of such resistant bacterial strains. Recent studies have highlighted the similarity between the E. coli strains isolated from the poultry and humans in terms of antimicrobial resistance determinants, which suggests the transmission of these strains from poultry to humans.^28,29

In this study, 62.4% (156/250) of the chicken cecal samples were found to harbor the ESBL producing E. coli. This is comparable to previous studies showing a varying level of ESBL phenotypes in poultry birds. In a study from Punjab, India a total of 1,556 Escherichia coli isolates were obtained from 18 poultry farms and have reported the prevalence of ESBL-producing E. coli strains as 87% in broilers while 42% in layers.³⁰ The incidence of ESBL-Ec in poultry as reported in the previous studies was relatively less, ie, 30% in a study from Bangladesh³¹ and 10.7% in a report in France.³²

In our study, the ESBL-Ec strains were resistant to ciprofloxacin (74.4%), sulfamethoxazole-trimethoprim (83%), gentamicin (45%), and amikacin (44%). The susceptibility profiles of the E. coli isolates from the poultry are quite variable. A recent study from Ethiopia has shown that the E. coli isolates were 100% susceptible to fluoroquinolones and sulfamethoxazole-trimethoprim, whereas the susceptibility to gentamicin, streptomycin, and kanamycin was 93%, 85%, and 75% respectively.³³ However, a few studies have reported highly resistant strains of E. coli which were resistant to 100 resistant to nalidixic acid while the resistance to kanamycin and streptomycin was 77% and 67% respectively.³⁴ These variations may be attributed to the differences in the use of antimicrobial agents in different regions in humans and poultry or due to the presence of diverse clones of E. coli.

The 37 isolates were found resistant to carbapenems (imipenem and meropenem). A study from China has recently isolated a bla_NDM-5 producing E. coli strain ST156 from a poultry farm. Which showed resistance to twenty-seven antimicrobial agents including carbapenems.³⁵ In a study comprising of 220 samples of chicken litter, chicken feed, intestinal content, and water, the Escherichia species were the most prevalent (39%) bacterial species. A total of 42 carbapenem-resistant isolates were found.³⁶ In a systematic review published in 2018, the prevalence of CRE was reported from the livestock. The prevalence was low in the European countries including the UK, Netherlands, and Switzerland, ie, <1%. However, the prevalence was higher in the Asian countries including China, India, Lebanon, and Algeria.¹⁹

In the present study, we have found that ST131 was the most prevalent sequence type (ST) among the E. coli strains subjected to MLST ie 46% (22/48). The previous studies have shown that the ST131 E. coli have been frequently isolated in the clinical settings and frequently harboring the bla_CTX-M gene with multidrug-resistant phenotypes.³⁷ This ST is also identified in various animal reservoirs, which shows that this clone has not only adapted to humans but other reservoirs as well, including animals.^37,38 It is also suggested that due to the acquisition of multiple virulence factors, the ST131 is also involved in the increase of E. coli carrying the bla_CTX-M genes responsible for ESBL phenotypes.³⁷

The studies that have indicated the presence of ST131 in the food animals are quite limited. In a study from Italy, only one E. coli ST131 isolates that was not producing ESBL was identified among a total of 101 E. coli isolates in healthy turkeys and chickens.³⁹ One E. coli ST131 isolates harboring bla_CTXM-1 was found among 22 ESBL producers in a total collection of 1378 E. coli strains from different animals mainly poultry, pigs, and cattle.⁴⁰ A single E. coli ST131 isolate was found in a pig with a gastrointestinal infection. A study reported the ESBL producing E. coli ST131 carrying bla_CTX-M-9 from the poultry birds in different countries from the samples collected at different periods.⁴¹ In contrast to these findings, many studies indicated the absence of E. coli ST131 from broilers, turkey sheep, cattle, and pigs.^42–45

The ST8900 has recently been reported among the avian origin colistin-resistant E. coli in Shandong Province, China harboring the mcr-1 gene in addition to bla_CTXM and bla_TEM genes.⁴⁶ The E. coli ST2847 have also been reported from Faisalabad, Pakistan from healthy broilers and was found resistant to colistin and positive for mcr-1 gene.⁴⁷ In contrast to these studies, both STs (ST8900 and ST2847) in our study were susceptible to colistin although positive for ESBL phenotypes and harboring the bla_CTXM and bla_TEM genes as found in the previous reports.

Conclusion

The molecular characterization and MLST of ESBL and carbapenemases producing Escherichia coli have helped to establish the genetic determinants among various sequence types for the first time in poultry settings in Pakistan. The disparities in the distribution of genetic determinists among various strains as seen in our study compared to the other reports may be attributed to the variable practices of poultry farming and the use of antimicrobial agents. It is important to find the presence of ST131 from poultry birds as this ST was previously reported from the clinical settings in most of the studies. The further characterization of ESBL producing E. coli from poultry, human as well as environmental sources in Pakistan is essential to understand the transmission dynamics of resistance determinants. The control strategies are necessary to prevent the dissemination of ESBL-Ec and CRE in animals, humans, and foods.