Detection of TEM and CTX-M Genes in Escherichia coli Isolated from Clinical Specimens at Tertiary Care Heart Hospital, Kathmandu, Nepal

Background: Antimicrobial resistance (AMR) among Gram-negative pathogens, predominantly ESBL-producing clinical isolates, are increasing worldwide. The main aim of this study was to determine the prevalence of ESBL-producing clinical isolates, their antibiogram, and the frequency of ESBL genes (blaTEM and blaCTX-M) in the clinical samples from patients. Methods: A total of 1065 clinical specimens from patients suspected of heart infections were collected between February and August 2019. Bacterial isolates were identified on colony morphology and biochemical properties. Thus, obtained clinical isolates were screened for antimicrobial susceptibility testing (AST) using modified Kirby–Bauer disk diffusion method, while ESBL producers were identified by using a combination disk diffusion method. ESBL positive isolates were further assessed using conventional polymerase chain reaction (PCR) to detect the ESBL genes blaTEM and blaCTX-M. Results: Out of 1065 clinical specimens, 17.8% (190/1065) showed bacterial growth. Among 190 bacterial isolates, 57.4% (109/190) were Gram-negative bacteria. Among 109 Gram-negative bacteria, 40.3% (44/109) were E. coli, and 30.2% (33/109) were K. pneumoniae. In AST, 57.7% (n = 63) Gram-negative bacterial isolates were resistant to ampicillin and 47.7% (n = 52) were resistant to nalidixic acid. Over half of the isolates (51.3%; 56/109) were multidrug resistant (MDR). Of 44 E. coli, 27.3% (12/44) were ESBL producers. Among ESBL producer E. coli isolates, 58.4% (7/12) tested positive for the blaCTX-M gene and 41.6% (5/12) tested positive for the blaTEM gene. Conclusion: Half of the Gram-negative bacteria in our study were MDR. Routine identification of an infectious agent followed by AST is critical to optimize the treatment and prevent antimicrobial resistance.


Background
Escherichia coli and Klebsiella species comprise the largest portion of the Gram-negative pathogens in several nosocomial and community-acquired infections, such as intra-abdominal infection, bloodstream infection (BSI), meningitis, and pyogenic liver abscess (PLA) [1]. E. coli is a commensal of the mammalian gastrointestinal tract, but can also be found in water, soil, and vegetation [2]. Virtually all pathogenic strains of Gram-negative bacteria are responsible for several infections, including gastroenteritis, urinary tract infection (UTI), septicemia, nosocomial infections, pneumonia, brain and abdominal abscess, and neonatal meningitis [3].

Sample Size and Sample Type
A total of 1065 non-duplicated clinical specimens were collected from the patients attending Shahid Gangalal National Heart Center. Among 1065 clinical specimens analyzed, 246 were blood, 300 were urine, 280 were sputum, 151 were pus or wounds, 59 were tips (catheter tips, central venous pressure tips, endotracheal tube, and secretion and suction tips), and 25 were fluids (pericardial fluids, peritoneal fluids, pleural fluids, and other body fluids). Demographic information was collected using a structured questionnaire.

Sample Collection and Transportation
All of the samples were collected in a clean, leak-proof container by the trained medical personnel. After aseptic collection of the samples, they were well-labeled and immediately sent to the microbiology department for further assay. In case of delay in delivery, samples were either refrigerated or preserved at 4 • C with appropriate preservatives. Samples were collected and processed by following standard microbiological methods [21,22].

Laboratory Processing of the Specimens 2.4.1. For Blood Samples
Five ml of blood were mixed with 45 mL of brain heart infusion (BHI) broth for adults and 1 mL of blood with 9 mL of BHI broth for children. The bottle was incubated at 37 • C for seven days. After sufficient incubation, bottles showing turbidity were further sub-cultured aerobically in MacConkey agar (MA) and blood agar (BA) at 37 • C for 24 to 48 h. If growth occurred, then the isolated colony was identified by colony morphology, Gram staining, and a biochemical test [23].

For Urine Samples
The urine samples were cultured into cysteine lactose electrolyte deficient (CLED) agar by semi-quantitative culture techniques using a standard calibrated loop (4 mm). A loop full of urine was streaked on the plate and then incubated at 37 • C in aerobic conditions for 18-24 h. Colony count was performed, in which a positive result was considered for plates having more than or equal to 10 5 colony-forming units (CFU)/mL of urine based on Kass, Marple, and Sanford criteria [21,24] These specimens were inoculated in BA and MA plates. The plates were incubated aerobically at 37 • C overnight. If growth occurred, then the isolated colony was identified by colony morphology, Gram staining, and a biochemical test [21,24].

Identification of the Isolates
After incubation, the culture plates were examined for the growth of organisms. Presumptive identification of the isolates was made solely on the basis of colony morphology and Gram staining. Isolates were then subjected to other confirmatory biochemical tests (catalase, oxidase, sulphur indole motility, methyl red/voges-proskauer, triple sugar iron, citrate, urease, and oxidative fermentative test) [24].

Screening of Multidrug Resistant (MDR) and Potential ESBL Producers
Isolates showing resistance to at least one agent of three or more classes of antimicrobial agents were classified as multi-drug resistant (MDR) [26]. If the zone of inhibition was ≤23 mm ceftriaxone, ≤21 mm ceftazidime, and ≤26 mm cefotaxime, the isolates were considered as a potential ESBL producer [25], and thus further subjected to the confirmatory tests.

Phenotypic Confirmation of ESBL Production
A combination disk test (CDT), as recommended by the CLSI (2018), was performed as the confirmation test for ESBL producers [27]. In this test, ceftazidime (30 µg) and cefotaxime (30 µg) disks alone and in combination with clavulanic acid (ceftazidime plus clavulanic acid, 30/10 µg; cefotaxime plus clavulanic acid, 30/10 µg) disks were applied onto an MHA plate (with an already inoculated test strain), and then incubated at 16-18 h for 35 ± 2 • C. Isolates showing an increase of ≥5 mm in the zone of inhibition of the combination disks in comparison to that of the ceftazidime/cefotaxime disk alone were confirmed as ESBL producers [27].

Preservation of the Isolates
An axenic culture of ESBL producer E. coli isolates were preserved in 20% glycerol containing tryptic soya broth and kept at −70 • C until further processing for molecular assay [28,29].

Plasmid DNA Extraction and Amplification
All of the phenotypically confirmed ESBL producers E. coli were analyzed for detection of β-lactamase genes (bla TEM and bla CTX-M ). The test organisms were inoculated in Luria-Bertani (LB) broth and incubated overnight at 37 • C with aeration by using a water bath shaker. After incubation, the plasmid DNA was extracted by using the alkaline lysis method [30]. After extraction of the DNA, the DNA was suspended in 50 µL of TE buffer and stored at deep freeze (4 • C) or at −20 • C [30].

DNA Amplification and Detection
Conventional PCR was used to detect the plasmid genes. PCR amplification reactions were carried out in a 21 µL volume, in which a master mix containing 200 µM of dNTP s (dATP, dCTP, dGTP, and dTTP), 120 nM of each primer (forward and reverse), 0.5 U/µL of Taq polymerase in 1× PCR buffer, 25 mM of MgCl 2 , and 3 µL of DNA template was added. Amplification reactions were performed in a DNA thermal cycler under the following thermal and cycling conditions for the bla TEM (F.P:5 -GAGACAATAACCCTGGTAAAT-3 R.P:5 -AGAAGTAAGTTGGCAGCAGTG-3 ) [16] and bla CTX-M (Forward Primer: 5 -TTTGCGATGTGCAGTACCAGTAA-3 , Reverse Primer: 5 -CGATATCGTTGGTGGTGCCATA-3 ) [31] genes: initial denaturation at 94 • C for 1 min, denaturation at 95 • C for 5 min of 35 cycles, annealing at 55 • C for 1 min of 35 cycles for bla CTX-M and 56 • C for 45 s of 35 cycles for bla TEM , extension at 72 • C for 1 min of 35 cycles, and final extension at 72 • C for 10 min [16,31].
After PCR amplification, 15 µL of each reaction were separated by electrophoresis in 1.5% agarose gel for 60 min at 100 V in 0.5× TBE buffer. Ten µL of DNA was stained with ethidium bromide (5 µg/mL). Thus, amplified DNA bands were visualized by using a UV transilluminator [28,29]. The amplicon size for bla TEM was 459 bp [16] and bla CTX-M was 544 bp [31].

Quality Control
For the standardization of the drug susceptibility test, control strains of K. pneumoniae (ATCC 700603) and E. coli (ATCC 25922) were used. For, PCR control sterile water (negative) and the known positive DNA and negative controls from the previous extraction (positive) were processed to ensure the correctness of PCR process.

Statistical Analysis
Collected data were entered into SPSS version 24.0 software and analyzed. The chisquared test was used to explore the association between categorical variables. A p-value of < 0.05 was considered to be significant in determining the relationship between the variables.

Distribution of Culture-Positive Bacterial Isolates
Out of the total samples, 17

Distribution of ESBL Producers ESBL-Producing E. coli
Among 44 E. coli isolates, 22 isolates were screened positive for ESBL production using ceftazidime, whereas 24 isolates were screened positive for ESBL production using cefotaxime. In the confirmatory assay, 27.3% (12/44) were confirmed as ESBL-producing E. coli. Among 12 ESBL producers, the majority (66.7%; 8/12) were from females. All of the ESBL producers (n = 12) were detected in urine samples. Statistically, there was no significant association between the type of specimen and ESBL producer (p = 0.54) ( Table 4).

Distribution of ESBL Producers ESBL-Producing E. coli
Among 44 E. coli isolates, 22 isolates were screened positive for ESBL production using ceftazidime, whereas 24 isolates were screened positive for ESBL production using cefotaxime. In the confirmatory assay, 27.3% (12/44) were confirmed as ESBL-producing E. coli. Among 12 ESBL producers, the majority (66.7%; 8/12) were from females. All of the ESBL producers (n = 12) were detected in urine samples. Statistically, there was no significant association between the type of specimen and ESBL producer (p = 0.54) ( Table 4).

Molecular Detection of ESBL Producer Genes
Among 12 ESBL-producing E. coli, 58.4% (n = 7), 41.6% (n = 5), and 25.0% (n = 3) isolates were tested positive for the bla CTX-M gene, bla TEM gene, and both bla CTX-M and bla TEM genes, respectively ( Figure 3). The bla CTX-M gene and bla TEM gene on 1.5% agarose gel under UV light was detected with a band of product size of 544 bp and 499 bp, respectively (Figure 4).

Overall Findings
Infections mediated by Gram-negative bacteria (GNB) constitute the major burden in low-and middle-income countries (LMICs) due to acquisition of various resistant genotypes. In particular, E. coli, K. pneumoniae, E. cloacae, and non-fermentative bacteria, such as P. vulgaris, A. baumannii, and Salmonella species, represent the major members of GNB that are associated with frequent and more severe forms of clinical manifestations, including urinary tract infection, bacteremia, and pneumonia [10,32]. More specifically, E. coli and K. pneumoniae represent the most predominant pathogens isolated from such infections [15,33]. Antibiotics are the choice of therapeutic options for the management of these infectious diseases. However, recent findings from Nepal on the AMR suggest the emerging resistance to virtually all classes of frontline drugs in use [5]. Acquisition and horizontal transfer of resistant genes from multiple sources of pathogenic microbes, the environment, and animals are suggested as the principal drivers of the uncontrolled spread of resistance [5,34]. Urinary tract infection (UTI) was the major clinical complaint among pa-

Overall Findings
Infections mediated by Gram-negative bacteria (GNB) constitute the major burden in low-and middle-income countries (LMICs) due to acquisition of various resistant genotypes. In particular, E. coli, K. pneumoniae, E. cloacae, and non-fermentative bacteria, such as P. vulgaris, A. baumannii, and Salmonella species, represent the major members of GNB that are associated with frequent and more severe forms of clinical manifestations, including urinary tract infection, bacteremia, and pneumonia [10,32]. More specifically, E. coli and K. pneumoniae represent the most predominant pathogens isolated from such infections [15,33]. Antibiotics are the choice of therapeutic options for the management of these infectious diseases. However, recent findings from Nepal on the AMR suggest the emerging resistance to virtually all classes of frontline drugs in use [5]. Acquisition and horizontal transfer of resistant genes from multiple sources of pathogenic microbes, the environment, and animals are suggested as the principal drivers of the uncontrolled spread of resistance [5,34]. Urinary tract infection (UTI) was the major clinical complaint among patients in this study. Overall, more than half of the isolates showed multidrug resistance (MDR) to the common antibiotics in use. Similarly, half of the E. coli isolates were ESBL producers, of which almost one-third of the isolates had ESBL genes bla CTX-M and bla TEM . A high prevalence of MDR and acquisition of resistant genotypes is attributed to the poor infection control in the country, and warrants urgent efforts to tackle the burgeoning AMR.
In this study, urine constituted the highest frequency (28.2%) among all of the clinical samples. Our samples echo previous studies conducted in the International Friendship Hospital, Kathmandu [3], Universal College of Medical Sciences Bhairahawa [10], Human Organ Transplant Center [33], and Nobel Medical College, Biratnagar [35]. This may be due to the high association of Gram-negative bacteria with UTI, which is the most common among patients attending a hospital in Nepal [16]. Similarly, E. coli was the most common organism, followed by Klebsiella spp., A. baumannii, and Pseudomonas spp. E. coli alone constituted almost one-fifth of the total isolates and more than one-third of the GNB alone. Our findings are consistent with the previous studies from Everest Hospital, Kathmandu [20], Alka Hospital, Lalitpur [28], International Friendship Hospital, Kathmandu [3], Padma Nursing Hospital, Pokhara [36], National Public Health Laboratory, Kathmandu [37], B.P. Koirala Institute of Health Sciences, Dharan [38], and Alka Hospital, Lalitpur [39].

Antibiotic Resistance and Multidrug Resistance
In this study, most of the GNB showed resistance to the commonly prescribed broadspectrum antibiotics. In addition to this, almost all of them were sensitive to the carbapenem drugs. More specifically, E. coli and K. pnemoniae were highly sensitive towards carbapenems (imipenem, meropenem) and amikacin. E. coli showed high sensitivity (>80%) towards meropenem and imipenem in our study. These findings are in line with the study reported from Manmohan Memorial Medical College and Teaching Hospital (MMCTH) [18], Alka Hospital, Lalitpur, Kathmandu [28]. The antibiogram of this study suggests the utility of carbapenems as a secondary therapeutic option for infections caused by MDR Gramnegative organisms. However, recent studies have shown the growing resistance to the last resort antimicrobials, carbapenems [40].
Overall, more than half (51.3%) of the GNB isolates were reported as MDR in our study. A high prevalence of MDR is reported from other studies from the International Friendship Hospital [3], Everest Hospital [20], Alka Hospital [28], Human Transplant Center, Kathmandu [33], Manmohan Memorial Medical College, and Teaching Hospital (MMCTH), Kathmandu [18]. The difference in the prevalence of MDR in Gram-negative bacteria in this study and previous studies is attributable to the differences in the source and types of samples, sample size, growth of organisms, and drug resistance pattern [10].

ESBL Producers and Acquisition of Resistant Genotypes
In this study, more than one-fourth (27.3%) of the total E. coli isolates were confirmed as ESBL producers. This finding is consistent with the previous studies reported from Model Hospital, Bagbazar, Kathmandu [15], Everest Hospital, Baneshwor, Kathmandu [20], and International Friendship Hospital, Maharajgunj, Kathmandu [3], and lower than a study reported from Manmohan Memorial Medical College and Teaching Hospital (MMCTH) [16,18] and Om Hospital and Research Centre, Kathmandu [41]. The higher rate of ESBL in this organism compared to the previous studies may indicate the temporal impact of increasing infections and AMR over the years. For instance, together with the growing population, the trend in self-medication, distribution of suboptimal quality of antibiotics, poor infection control, sanitation, and hygiene practices (in the personal and community level) are the driving mechanisms for the unabated spread of resistant pathogens [11,42].
In our study, all of the ESBL-producing E. coli (27.3%) harbored at least one of the tested genes (bla CTX-M (58.4%) and bla TEM (41.6%)), while one-fourth of them showed the co-expression of both genotypes. This result indicates a slight decrease in prevalence of bla TEM and bla CTX-M genes than a previous study reported from Annapurna Neurological Institute and Allied Sciences, Kathmandu (33.2% ESBL producers, 83.8% bla TEM , and 30.8% bla CTX-M ) [31]. Another study reported 40.3% ESBL producers, 83.8% bla TEM and 66.1% bla CTX-M from MMCTH [16]. The dominance of the bla CTX-M gene has been reported from previous studies from Nepal and overseas [43][44][45][46]. The difference in the prevalence rate of TEM and CTX-M genes between the present study and other studies might be due to the difference in the sample population and antibiotic susceptibility pattern of E. coli, the type of study design, and sample size. Similarly, co-expression (or multiple occurrences) of the genes were common in other studies too. These genes are often present in large plasmids, and are capable of conferring resistance to the organisms [47].
All of the ESBL-producing E. coli isolates were resistant towards ampicillin, cefotaxime, and ceftazidime, while the entire isolates were also susceptible to the carbapenems, such as meropenems and imipenems. Our findings are consistent with some previous findings reported from Manipal Teaching Hospital, Pokhara [48], International Friendship Hospital, Maharajgunj, Kathmandu [3], Chitwan Medical College, Bharatpur [49], International Friendship Children Hospital, Kathmandu [50], Kathmandu Medical College Teaching Hospital, Sinamangal, Kathmandu [14], and Shahid Gangalal National Heart Centre, Bansbari, Kathmandu [51,52]. Higher resistance to penicillin and third generation cephalosporin in this study can be solely attributable to their ability to produce ESBL.
A higher fraction (respectively, two-thirds and slightly less than two-thirds) of ESBL producers and MDR were isolated from the female patients. The slight preponderance in females is explained by the higher prevalence of UTIs among females. Similarly, women of the age group below adulthood and in the post-menopausal stage are infected due to hormonal changes and poor sanitation practices that are often prevalent in LMICs [36,37]. GNB are the potential agents for nosocomial infections, and the outcome of such infections are associated with prolonged hospital stays, increased ICU admissions, and unwanted morbidities and mortalities [3,53].

Strengths and Limitations
The prevalence of GNB, their antibiogram, and the status of MDR among clinical samples of the study hospital can serve as an important reference tool for future studies, clinicians, and policy makers. Exploration of the prevalence of resistant genotypes of ESBLproducing E. coli in this study highlights the importance and need of molecular diagnostic facilities for a more precise detection of the infectious diseases. However, our study also suffers from a number of limitations. Our study design was limited to a single hospital, and had small clinical samples, which cannot generalize the prevalence of AMR elsewhere. Moreover, due to the limitation of the laboratory resources and funding, we could not characterize the resistant genotypes of other Gram-negative organisms. In addition to this, we could not characterize the other members of the ESBL genes, including SHV within the same family, and the genotypes of β-lactamases of other Ambler classes. The study is limited by the inability to explore the origin of the resistant genotypes and their potential transferability. Therefore, future studies are recommended to include the larger populations and provide an extensive characterization of the resistant genotypes to yield an accurate burden of AMR.

Conclusions
More than half of the Gram-negative isolates were detected as the multidrug resistant strains in our study. Half of the pathogenic strains of E. coli were potential ESBL producers, of which half of them harbored the ESBL genes under question. The findings of this study imply an urgent need for early suspicion, identification, and AST to optimize treatment and limit the spread of AMR.

Acknowledgments:
We are grateful to the staff and faculty members of the Sahid Gangalal National Heart Center, Kathmandu, Nepal, and Central Department of Microbiology, Kirtipur, Kathmandu for their support and coordination to accomplish the study. We express our sincere gratitude to all the patients for their involvement in the study. We would like to thank UGC (University Grants Commission) for providing financial support for thesis preparation through a faculty research grant under Nabaraj Adhikari (UGC Grant No.FRG-74/75-HS-13).

Conflicts of Interest:
The authors declare no conflict of interest.