Genetic Environment of Plasmid Mediated CTX-M-15 Extended Spectrum Beta-Lactamases from Clinical and Food Borne Bacteria in North-Eastern India

Background The study investigated the presence of CTX-M-15 type extended spectrum beta-lactamases (ESBL), compared their genetic arrangements and plasmid types in gram negative isolates of hospital and food origin in north-east India. From September 2013 to April 2014, a total of 252 consecutive, non-duplicate clinical isolates and 88 gram negative food isolates were selected. Phenotypic and molecular characterization of ESBL genes was performed. Presence of integrons and gene cassettes were analyzed by integrase and 59 base-element PCR respectively. The molecular environments surrounding bla CTX-M and plasmid types were investigated by PCR and PCR-based replicon typing respectively. Transformation was carried out to assess plasmid transfer. Southern blotting was conducted to localize the bla CTX-M-15 genes. DNA fingerprinting was performed by ERIC-PCR. Results Prevalence of ESBL was found to be 40.8% (103/252) in clinical and 31.8% (28/88) in food-borne isolates. Molecular characterization revealed the presence of 56.3% (58/103) and 53.5% (15/28) bla CTX-M-15 in clinical and food isolates respectively. Strains of clinical and food origin were non-clonal. Replicon typing revealed that IncI1 and IncFII plasmid were carrying bla CTX-M-15 in clinical and food isolates and were horizontally transferable. The ISEcp1 element was associated with bla CTX-M-15 in both clinical and food isolates. Conclusions The simultaneous presence of resistance determinants in non-clonal isolates of two different groups thus suggests that the microbiota of common food products consumed may serve as a reservoir for some of the drug resistance genes prevalent in human pathogens.


Introduction
Extended spectrum beta-lactamases conferring resistance to third generation cephalosporins in Gram negative bacteria is a global health threat. In hospital settings infection with ESBL-producing organisms results in poor clinical outcomes, delay in antibacterial therapy, longer hospital stay and increased hospital expenses [1]. Thus, these organisms serve as a potential threat and economic burden for the public health departments and communities. Over the past decade CTX-M has become the most prevalent family of ESBLs encoded by bla CTX-M. More than 147 different variants of CTX-M have been reported (http://www.lahey.org/studies), of which CTX-M-15 is the most predominant one, first reported from the Indian subcontinent in 2001 and expanded worldwide [2,3]. Since then numerous reports documented the presence of bla CTX-M-15 among nosocomial isolates in this subcontinent [4]. The incidence of infections due to such resistant organisms has rapidly increased over the last decade and has become a worldwide epidemic. Besides their resistance to cephalosporins these organisms often display co-resistance to aminoglycosides, tetracyclines and sulfonamides posing serious therapeutic challenge [5].
To make things worse, bla CTX-M-15 is found on plasmid and other mobile genetic elements [2] and is horizontally transferred to other pathogenic bacterial strains or can even cross species barrier. Various types of genetic environments might be involved in the mobilization of bla CTX-M genes, of which ISEcp1 (Insertion sequence) is often associated upstream of the bla CTX-M genes, hence serving as an efficient tool for the lateral transfer and expression [2,[6][7][8]. In addition, several different insertion sequences viz. IS26, IS10, IS5 and IS903 have been detected surrounding the resistance determinant [6,[9][10][11].
For isolates harbouring bla CTX-M-15 , their genetic linkage and transmission dynamics have been reported frequently and very well documented in the nosocomial isolates whereas; in the community numerous reservoirs have been reported which involve non-human pathogens like poultry, food-producing animals, pets and also the raw vegetables [12][13][14][15]. But there is paucity of data pertaining to the persistence and genetic arrangement of CTX-M gene among foodborne isolates. Therefore, it becomes necessary to evaluate if the ready to eat foods available at the stalls and eating outlets or raw food may constitute as reservoir and contribute to the rapid expansion of ESBLs. Also, it is important to know the factors that are involved in maintenance of these determinants in the food isolates and potential risk of transmission of these resistant organisms through the food chain in the environmental conditions. Thus, present study was undertaken to document the prevalence of bla CTX-M-15 type ESBL genes, their genetic arrangement and transmissibility among nosocomial isolates as well as in food-borne isolates obtained from food samples available at eating outlets in north-east India.
Apart from clinical strains, a total of 88 gram negative bacteria [E. coli (n = 48), K. pneumoniae (n = 14), Klebsiella spp. (n = 3), Citrobacter spp. (n = 3), P. mirabilis (n = 7), E. aerogenes (n = 2), E. cloacae (n = 5) Pseudomonas spp. (n = 6)] were isolated from variety of cooked (cakes, sweets, ice-creams etc) and uncooked (vegetables, meat) food samples (Table 1). Samples were collected using sterile vials from various local sweet shops, meat shops, street vendors and other eating outlets situated in remote as well as at the prime locations of the Shillong city from September 2013 to April 2014. Sterile cotton gauze was used to swab properly all over the surface of samples. For further analysis, serial dilutions were prepared starting with 1 mg of food samples diluted in 9ml of saline solution (0.9% NaCl). A volume of 100μl from each well homogenized dilution was inoculated onto the MacConkey agar containing vancomycin (3mg/ L). Plates were incubated at 37°C for 24 h under aerobic conditions. All lactose fermenting and non-lactose fermenting colonies with different coloration and morphology belonging to the members of enterobacteriaceae family as well as non fermenting gram negative rods were picked from the selective plates, subcultured and stored. All the isolates were identified according to Gram stain, cultural characteristics and biochemical reactions [16].

Ethical approval
The work was approved by Institutional Ethics Committee for Human Samples/Participants of North Eastern Hill University, vide reference number: IECHSP/2014/03. The authors confirm that all the patients, patients' parents/guardians provided their written informed consent to participate in the study. Written permission was obtained from the owners of the shops for carrying out experiment from their food samples. The authors hereby also confirm that the field study did not involve any protected or endangered species.

Phenotypic detection of ESBL
All the isolates were selected on the basis of their initial screening followed by the confirmatory combined disc diffusion method for presence of ESBL according to CLSI recommendation [17].

Molecular characterization of bla ESBL genes by Multiplex PCR
DNA was extracted by the boiling centrifugation method. To determine the genotypes of ESBLs, PCR was performed using primers specific to bla TEM , bla CTX-M , bla SHV , bla OXA-2 , bla OXA-10 and bla GES as described previously [18]. Reaction mixture was prepared using Promega 2X PCR mix (Promega, Madison, USA). Reactions were run under the following conditions: initial denaturation 94°C for 5 min, 32 cycles of 94°C for 1 min, 54°C for 1min, 72°C for Multiplex PCR assay for AmpC beta lactamases and carbapenemases encoding genes All the ESBL positive isolates were tested for presence of plasmidic AmpC [19] and class A and B carbapenemases [20][21][22][23]. Reactions conditions were as described previously [19][20][21][22][23] Genetic environment of bla CTX-M-15 Integrase gene PCR was performed for the characterization of class I and class II integrons, using IntI F' CAGTGGACATAAGCCTGTTC, IntI R' CCCGAGGCATAGACTGTA and IntII F' TTGCGAGTATCCATAACCTG, IntII R' TTACCTGCACTGGATTAAGC primers [24]. For association of bla CTX-M with other gene cassettes, 59 base element (59be) PCR was performed using HS 286 and HS 287 primers [24]. In order to determine the variable regions of integron gene cassettes carrying bla CTX-M, cassette PCR was performed amplifying 5' and 3' conserved sequences [25]. Two reactions were performed consequently: in one reaction 5CS and reverse primer of bla CTX-M and in other reaction 3CS and forward primer of bla CTX-M was used [18,25]. PCR reactions were run under the following conditions: initial denaturation 95°C for 2 min, 29 cycles of 95°C for 20 sec, 54°C for 1 min, 72°C for 2 min and final extension at 72°C for 10 min.

Genetic association of the CTX-M gene with insertion sequence ISEcp1
To determine the association of bla CTX-M with ISEcp1, PCR analysis was performed using the forward primer for ISEcp1 (5'-AAAAATGATTGAAAGGTGGT-3') and the reverse primer for CTX-M gene (5'-ACCGCGATATCGTTGGT-3'). The reaction mixture and running conditions were same as described previously for gene cassettes.

Plasmid preparation
All bla CTX-M producing bacterial isolates were cultured in Luria Bertani broth (Hi-Media, Mumbai, India) containing 1mg/L of cefotaxime. Plasmids were purified by standard alkaline lysis method.

Southern hybridization for determination of bla CTX-M-15 carriage
To validate our study, Southern blotting was performed on agarose gel by in-gel hybridization [26] with

Gene Transferability of bla ESBL gene by transformation
Transformation was carried out using E. coli DH5α as recipient. Transformants were selected on cefotaxime (0.5 mg/L) containing LB Agar plates.

Plasmid stability and incompatibiltiy typing
Plasmid stability of bla CTX-M producers as well as their transformants were analyzed by serial passages method for consecutive 75 days at 1:1000 dilutions in LB broth without antibiotic pressure as described previously [27]. PCR for presence of the bla CTX-M-15 was carried out for the isolates after each passage. Incompatibility typing was carried out by PCR based replicon typing [28] among all the transformants carrying bla CTX-M.

Typing of isolates
Enterobacterial repetitive intergenic consensus (ERIC) PCR was performed for typing of all the CTX-M harbouring isolates [29].

Results
Among the tested strains, ESBL production was obtained in 103 (40.8%) clinical and 28 (31.8%) food isolates (S1 Table). Molecular characterization revealed presence of 56.3% (58/ 103) bla CTX-M , 12.6% (13/103) bla TEM and 9.7% (10/103) bla SHV in clinical isolates (S2 Table), while 53.5% (15/28) food isolates showed the presence of CTX-M gene ( Table 2). Few clinical isolates also revealed presence of multiple ESBL genes, while some ESBL producers (n = 27) did not show any amplifications by the primers used in the study. Among studied genes, bla CTX-M was found to be predominantly present in both clinical as well as food isolates. Distribution of CTX-M gene in different clinical isolates was as follows: E. coli (n = 45), K. pneumoniae (n = 6), P. mirabilis (n = 1), P. aeruginosa (n = 6). Among food borne isolates, CTX-M was present in E. coli On observing the antimicrobial susceptibility of the CTX-M encoding isolates, 100% susceptibility was shown towards tigecycline and polymyxin B followed by imipenem (95.8%), ertapenem (95.1%) and amikacin (89.6%), while all other antibiotics including third generation cephalosporins, aminoglycosides and flouroquinolones showed moderate to poor activity (Table 3). In comparison to food isolates, most of the clinical isolates showed multidrug-resistant phenotype. A high MIC in the resistant range was observed against all tested cephalosporins (Table 4).
Transferability assay showed that CTX-M could be horizontally transferred from diverse host range to recipient E. coli DH5α. However, plasmids of P. aeruginosa could not replicate within E. coli host. The plasmids of the transformants could be hybridized with CTX-M-15 specific probe. In replicon typing, it was observed that CTX-M-15 was carried within Inc I1 and Inc F II types. In nine transformants, the plasmid was untypable. All the CTX-M positive isolates were found to harbour a class I integron. Plasmids in clinical isolates were more stable than those of food isolates where plasmid carrying CTX-M was retained after 75 passages for clinical isolates while in case of food isolates plasmid was lost after 62 serial passage. However, transformants did not show high stability as plasmids were lost between 15 to 36 passages. CTX-M-15 was associated with ISEcp1 in the upstream region in 37 clinical and 8 food isolates (Fig 1). Among the rest of isolates, interestingly, integron mediated CTX-M-15 was observed particularly in food isolates. Two different arrangements were observed along with other resistance determinants (Fig 2).

Discussion
The present study demonstrated both the hospital and community based transmission of bla CTX-M-15 , raising a threat to public health. The study simultaneously analyzed the CTX-M-15 type ESBL genes, their plasmid type and genetic environment from food borne and human clinical Gram negative isolates from this part of the world. CTX-M-15 is one of the most widespread ESBL genes and has significant impact on the treatment of hospital and community acquired infections. Majority of the bla CTX-M-15 harbouring nosocomial isolates investigated in the present study were obtained from the females diagnosed with urinary tract infection (UTI). Apart from female patients, the gene was also reported in immuno-suppressed patients of intensive care unit as well as from the out-patient departments. It is well established that such resistant organisms can be frequently transmitted to a wide range of geographical region. Similar to the present findings, several other studies reported the high rate of ESBL producers among UTI patients [30,31]. The study further demonstrated the presence of CTX-M encoding isolates in raw meat and in ready-to-eat foods like salad, cakes and sweets which significantly increases their dissemination and colonization in gut flora. The incidence of CTX-M-15 harbouring isolates in food could serve as potential reservoir of pathogens prevalent in hospital settings. Such situations pose high risk of exposure of healthy humans to these multidrug resistant strains during handling and consumption of meat, sweets and vegetables and may contribute to the colonization of these strains in human gut along with commensals. This suggests that, there should be increased awareness amongst such food handlers regarding the safe and hygienic practices. In contrast to our findings, previous reports from Aligarh, India and from Northern Ireland depicted the dissemination of bla CTX-M in clinical isolates but none of the food borne isolates was harbouring the gene [32,33]. While, similar to our report, a study from Spain in 2003 described CTX-M encoding gram negative isolates in cooked foods, salads and raw chicken [34]. In this study, around 37% of CTX-M-15 harbouring isolates with cefotaxime susceptibility might be attributed to heterogenic resistance property. In the past, reports have also shown similar oxy-imino cephalosporin susceptibility among ESBL producers [35]. Presence of AmpC beta lactamase in our study isolates might be responsible for exhibiting carbapenem nonsusceptibility in few isolates possibly with porin loss as no carbapenamase gene could be traced by PCR assay.
This study advocates diverse source of origin and acquisition of CTX-M-15 in the study area, as the genetic vehicle for transfer of this gene were the two different Inc type plasmids. Similar to earlier studies, bla CTX-M-15 was located on replicon type Inc FII whereas other replicon type Inc I1 was also reported [31,36]. Further, ISEcp1 may be an additional efficient tool for the enhanced mobilization and expression of bla CTX-M-15 which corroborates to previous reports [2,[6][7][8]. In the present study, presence of ISEcp1 in the upstream region of bla CTX-M-15 across the species indicates that the whole insertion sequence or a part of it possibly excised along with CTX-M-15 during horizontal transfer. Integrons play an important role in the dissemination of these resistance determinants among bacterial flora. Another aspect of this study was the location of bla CTX-M-15 on gene cassettes along with other resistance genes which includes dfr7, aadA1, sul1, which enhances the risk of co-selection. The genetic arrangement of CTX-M-15 with ISEcp1 and integrons highlights the role of these mobile elements in the maintenance of this gene in the environment where the antibiotic pressure is invariably considered less compared to hospital settings. The study could highlight a new genetic background of CTX-M-ESBL. This particular finding may be well explained as a reason for expansion of these resistance determinants both vertically and horizontally with better stability and their spread does not confine to a particular geographical location. This study also puts forward the concern for potential threat for treatment failure and maintenance of cephalosporin resistance in a given global situation.

Conclusions
Indiscriminate use, over the counter availability and unethical marketing of antibiotics has already created havoc for treating common infections. Additionally, contamination of foods by food handlers particularly ESBL carriers warrant urgent hygiene measures to slow down the spread of resistance. This demands a need to increase the awareness amongst these food handlers regarding the safe and hygienic practices. In a country with already existing burden of antibiotic pressure in hospital settings, presence and propagation of these resistance determinants in the food borne isolates could serve as an additional reservoir. Transmission of these determinants to healthy individuals, augment the grave situation and contributes to the emergence of drug resistant pathogens in this location. Considering this fact, the presence and persistence of resistance determinants (CTX-M-15) in food isolates underscores the potential threat of their spread within pathogens thereby restricting therapeutic alternatives.