Isolation of E. coli and antimicrobial susceptibility
A total of 2,291
E. coli were isolated from 790 fecal samples of pigs, of which 1113 and 1178 were from organized and unorganized farms, respectively (Table 1). All the isolates exhibited resistance to at least 3 antimicrobial classes. Amoxycillin (84.81%) showed highest level of resistance followed by cefalexin (77.17%), sulphafurazole (56.79%), piperacillin (46.40%), tetracycline (38.29%), cefexime (35.66%), co-trimoxazole (27.94%), trimethoprime (26.32%), ampicillin (26.09%), ceftazidime (22.87%), nalidixic acid (21.82%), aztreonam (18.37%), gentamicin (16.27%), streptomycin (9.12%), cefotaxime (8.25%), ceftriaxone (7.51%), ciprofloxacin (6.11%), and imipenem (0.22%) (Table 2). Of the 1113 isolates obtained from organized farms, imipenem (99.82%) was found to be the most sensitive antibiotic followed by amoxicillin (81.94%), sulphafurazole (48.70%), piperacillin (42.95%), tetracycline (39.64%), cefixime (36.03%), ceftazidime (27.94%), ampicillin (27.13%), nalidixic acid (22.46%), trimethoprim (21.38%), aztreonam (20.58%), co-trimoxazole (21.20%), gentamicin (14.93%), streptomycin (11.41%), ceftriaxone (8.18%), cefotaxime (6.92%), ciprofloxacin (4.85%) and imipenem (0.18%). Similarly, of the 1178 isolates obtained from unorganized farms, amoxicillin (87.52%) exhibited the highest level of resistance followed by cefalexin (71.05%), sulphafurazole (64.43%), piperacillin (49.66%), tetracycline (37.01%), cefexime (35.31%), co-trimoxazole (34.30%), trimethoprime (30.98%), ampicillin (25.09%), nalidixic acid (21.22%), ceftazidime (18.08%), gentamicin (17.57%), aztreonam (16.30%), cefotaxime (9.51%), ciprofloxacin (7.30%), streptomycin (6.96%), ceftriaxone (6.88%) and imipenem (0.25%).
Amoxicillin is a widely used antibiotic against
E. coli infection in man and animals
(Xia et al., 2012). However, because of indiscriminate use, misuse and abuse of such antibiotic, bacteria developed the highest level of resistance, which makes the drug a completely irrelevant option against
E. coli. As mentioned in Table 1, amoxicillin (84.81%) exhibited highest level of resistance, which is also in corroboration of the observation by
Nijsten et al., (1996), who reported prevalence of
E. coli isolates from pigs in the south of Netherlands with 92-100% resistance against amoxicillin.
The resistance pattern of
E. coli isolates against all the antimicrobial agents are variable within the isolates obtained from different farms of the different states of NER. In this study, the resistance pattern of
E. coli isolates against all the four third generation cephalosporins exhibited variable resistance (Table 1). National Antimicrobial Resistance and monitoring System (NARMS) report showed that resistance to ceftriaxone ranged from 6.3% to 13.5% among
E. coli isolated from chickens during 2000-2008. In contrast to our result,
Sasirekha et al., (2010) and
Singh and Goyal (2003) reported that 84%, 75% and 85%
E. coli were resistant to cefotaxime, ceftriaxone and ceftazidime, respectively. On the other hand,
Rosengren et al., (2008) reported no resistant isolates against ceftriaxone and less than 1% resistant to cefoxitin and ceftiofur.
Gebreyes et al., (2004) also mentioned that resistant strains are more commonly isolated from the pigs in the production system that used antimicrobials, including beta-lactams and drugs of the tetracycline class more often than from pigs in another type of large production system. With the availability of third and fourth generation of cephalosporins and without any restriction upon their use in veterinary and human health practice, a high level of resistance against them are increasingly evidenced (
Myeyer, 2010;
Moghnieh et al., 2019). Although, some antimicrobial agents are not in use for veterinary practices in India, but a high level of resistance against them is recorded in the fecal
E. coli of swine. The persistence of resistance might result from the process of co-selection by other antimicrobials being used in pigs
(Binch et al., 2008).
Interestingly, imipenem under-performed against
E. coli isolates as compared to high susceptibility in other studies
(Shams et al., 2018). Imipenem is not in use for veterinary practices in any of the North eastrern region (NER) states of India. Detection of low level of resistant
E. coli in pig intestine is an indication of environmental transfer of such organism from human to animals. The reverse activity of zoonotic transfer of such organisms from animals to man may worsen the situation of treatment option in human practice. Decreased susceptibility to imipenem is a matter of great concern for treating infections caused by gram negative bacteria and indicates the urgent need for improved infection control strategies.
Most of the ESBL producing organisms were found to be co-resistant to flouroquinolones, aminoglycosides and co-trimoxazole, which corroborates with the study done by
Jabeen et al., (2005) and
Denholm et al., (2009). Perez et al., (2007) also reported the ESBLs producing enteric bacteria resistant to other group of antibiotics including aminoglycosides, tetracycline, sulfonamides, trimethoprim and chloramphenicol. Development of co-resistance against other antibiotics along with β-lactam antibiotics by the ESBLs producing organisms generally appeared in the large plasmids, where most of the resistance genes may co-exist. Extended-spectrum cephalosporins are important therapeutic agents in veterinary and human medicine and are often used as first line agents for invasive Gram-negative infections. The prevalence of third generation cephalosporin resistance in
E. coli in pigs in this study suggests that these agents may be rapidly losing their efficacy as treatment option. Additionally, high rates of co-resistance to other classes of antimicrobial agents are further aggravating the situation. Over 10% of the isolates, which are resistant to third generation cephalosporins are also co-resistant to tetracycline, sulfonamides, trimethoprim, gentamycin, streptomycin, ciprofloxacin and aztreonam.
Confirmatory test for ESBLs
Of the 1,094 (47.75%)
E. coli isolates found to be positive for suspected ESBL by antimicrobial sensitivity test, 654 (28.55%) were confirmed as ESBL producers by DDST screening method, of which 121 (41.44%), 60 (22.30%), 44 (14.86%), 108 (34.39%), 87 (33.72%), 73 (29.80%), 55 (23.61%) and 106 (27.60%) were from Arunachal Pradesh, Assam, Manipur, Meghalaya, Mizoram, Nagaland, Sikkim and Tripura, respectively (Table 3). Several studies from India and abroad have also reported the prevalence of ESBLs producers varying from 6.6% to 91% from time to time
(Jain et al., 2003; Wattal et al., 2005; Bhattacherjee et al., 2008; Basavaraj et al., 2011). The wide variation in the prevalence of ESBLs producing organisms are probably due to the variation in the risk factors as well as the extent of antibiotics used. The relatively high rate of resistance to β-lactam antibiotics in primary screening test in comparison to the confirmatory DDST method may be because of different factors,
viz., decreased affinity of the target penicillin binding proteins (PBPs), decreased permeability of the drug into the target cell or the antibiotic is attributed only to the production of β-lactamase enzymes (
Sanders and Sanders, 1992;
Al-Chharak et al., 2011; Jacoby et al., 2011).
Genotypic detection of beta-lactamase gene blaCTX-M
Out of the 654 DDST positive
E. coli isolates, 65 (2.84%) isolates were found to be positive for
blaCTX-M gene. Amongst all the
E. coli isolates from entire 8 NER states, a total of 10 (3.42%), 4 (1.49%), 4 (1.35%), 5 (1.59%), 13 (5.04%), 7 (2.85%), 6 (2.57%) and 16 (4.16%) isolates were found positive for
blaCTX-M gene in Arunachal Pradesh, Assam, Manipur, Meghalaya, Mizoram, Nagaland, Sikkim and Tripura, respectively. Globally, the prevalence of
CTX-M producing
E. coli is varied between 0.8% - 3.4%
(Adeleke et al., 2012; Senthilkumar, 2012). In some Danish farms, the production of
CTX-M-1 β-lactamase in
E. coli strains recovered from feces of pigs has been associated with ceftiofur use
(Jørgensen
et_al2007). On the other hand, significant decreases in the carriage prevalence of
CTX-M-producing
E. coli and fecal counts of CTX-resistant coliforms were detected during the pig production cycle
(Hansen et al., 2013). The variation of results might be due to lower expression of
CTX-M gene.
blaCTX-M gene spread throughout the community, mostly through the transmission of plasmids, and some studies have also reported that animals may serve as a possible source for the dissemination of ESBL-encoding genes to humans
(Dutta et al., 2013).
Cloning and sequencing
Nucleotide blast search of sequence of
blaCTX-M (KF650780, KJ863555 and KJ863556) showed high degree of homology (99-100%) with the sequence reported for
blaCTX-M-15. A phylogenetic and bootstrap analysis for the
blaCTX-M gene showed minor nucleotide differences among the sequences.
In silico translation showed non-synonymous amino acid sequence. There were three amino acid differences within the sequence fragment. There was substitution of ‘G’ (glycine: non-polar hydrophobic aa) by ‘D’ (Aspartic acid: charged amino acid) in isolates PE:34/ii and PE:137/iv. Then ‘R’ (Arginine) is changed with ‘H’ (histidine), both positively charged hydrophilic amino acid in isolate No. PE:34/ii and PE:137/iv and substitution of ‘N’ (aspargine) by ‘H’ in in isolate No. PE:34/ii. Antibiotic profile of those isolates (PE:34/ii and PE:137/iv) revealed wider zone of inhibition against cefotaxime and ceftriazone, which indicated that our isolates were similar to
blaCTX-M-15 isolates reported from other sources and any differences in the activities towards various β-lactam antibiotics may be due to point mutations at selected places and the differences in the amino acid sequence leading to extended spectrum activity.
Horizontal gene transfer
The trans-conjugates were detected only by broth mating method, where the target gene was detected by PCR from the trans-conjugates but was absent in recipient strain. The frequency of transfer from donor to the recipient was between 3.6±2.07´10-8 to 4.4±2.88×10
-8 trans-conjugants per donor for
blaCTX-M gene. The pigs were tested prior to inoculation of the donor and/or recipient strains and no
E. coli and/or
Salmonella were isolated with the resistance characteristics of the donor, recipient or putative trans-conjugants strains. Both donor and recipient strains were detected within 24 hours from each pig after oral feeding. Recovered recipient organisms were tested for antibiotic susceptibility testing and produced similar pattern, which indicated the colonization of recipient strains in the pig intestinal tract. No pigs showed signs of clinical salmonellosis during the experiments.
Trans-conjugates were obtained from all the pigs and were resistant to cefotaxime and carrying the plasmids with target genes. Trans-conjugates were detected from the pigs with antibiotic treatment group from the 6
th day post inoculation, whereas, it could be detected from the 8
th day post inoculation from the group without any antibiotic pressure. The detection of trans-conjugates from both the groups could be done up to 10 days post inoculation. The target gene was detected from the trans-conjugates obtained from the plate, whereas it was absent in recipient strain, as detected by PCR assay. In case of
in vivo trans-conjugation study without any antibiotic selection pressure, all the isolates positive for
blaCTX-M gene could transfer their resistance trait to the recipient strain. But unlike the
in vitro study, the frequency of transfer from donor to the recipient was between 5.6±2.3×10
-4 to 6.8±3.35×10
-4 trans-conjugants per donor for
blaCTX-M gene. On the other hand, the frequency of transfer from all the donors to recipient was increased 6.6±3.05×10
-5 to 7.2±1.92×10
-5 trans-conjugants per donor under the antibiotic selection pressure. The frequency of transfer of resistance plasmids from donor to recipient may vary depending upon the copy number of the conjugative plasmids in the donor strain. In addition, the interspecies transfer of conjugative plasmids is less efficient than intra-species transfer. Failure or poor rate of transfer of resistance plasmid from donor to recipient by plate mating method is also reported by other workers
(Faure et al., 2009).
Similar kind of experimental data on horizontal gene transfer with variable results are also published by several workers
(Amavisit et al., 2003; Schjorring et al., 2008; Faure et al., 2009) using either gnotobiotic mice model or large animal model or even conventional colonization model to mimic the intestinal environment of human being. Using the animal model containing normal bacteria, flora always gives more realistic results than any
in vitro or gnotobiotic study. The normal bacterial flora barrier and the present immune system give the used animal model advantages in mimicking the human gastrointestinal tract. Transfer of resistance gene under
in vivo condition can’t be calculated from the extrapolation of
in vitro experiments. The indigenous flora can act as a reservoir and transfer resistance genes to pathogenic bacteria that might lead to infections with limited treatment possibilities. Transfer of any microbial resistance genes is threat, but transfer to ESBL resistance genes is in a category of its own, which might result in the limitation of treatment and in worst cases treatment failure (
Pitout and Laupland, 2008;
Schjorring et al., 2008; Dhillon and Clark, 2012;
Sing et al., 2017).
It may be stated that both clonal spread and transfer of transposable genetic elements might contribute to the proliferation of multidrug resistant ESBLs producing bacteria in the environment. Mobile drug resistance genes are capable of crossing bacterial species and are likely to accelerate dissemination of drug resistance between animals and humans through animal proteins or contact. It is, therefore, important to monitor the spread of extended spectrum cephalosporin resistant bacteria. Further studies of the basic mechanism for the evolution and dissemination of resistance as well as improved tools for the risk assessment of the spread of resistant organisms might enable us to control the global spread of antimicrobial resistance under the future influence of globalization.