Characterization of Genomic Diversity among Carbapenem-Resistant Escherichia coli Clinical Isolates and Antibacterial Efficacy of Silver Nanoparticles from Pakistan

The emergence of carbapenem-resistant Escherichia coli (E. coli) is considered an important threat to public health resulting in resistance accumulation due to antibiotics misuse and selection pressure. This warrants periodic efforts to investigate and develop strategies for infection control. A total of 184 carbapenem-resistant clinical strains of E. coli were characterized for resistance pattern, resistance genes, plasmids, sequence types and in vitro efficacy of silver nanoparticles (AgNPs). Carbapenem resistance was prevalent in E. coli isolated from female patients (64.7%), urine samples (40.8%) and surgical wards (32.1%). Polymyxin-B showed higher susceptibility. ESBLs and carbapenemases were produced in 179 and 119 isolates, respectively. Carbapenemase-encoding genes were observed among 104 strains with blaNDM-1 (45.1%), blaOXA-48 (27%), blaNDM-7 (3.8%), blaNDM-1/blaOXA-48 (15.4%), blaNDM-7/blaOXA-48 (2.9%), blaOXA-48/blaVIM (3.8%) and blaNDM-1/blaVIM (2%). ESBL resistance genes were detected in 147 isolates, namely blaSHV (24.9%), blaCTX-M (17.7%), blaTEM (4.8%), blaSHV/blaCTX-M (29.2%), blaSHV/blaTEM (15%) and blaCTX-M/blaTEM (8.8%). ST405 (44.4%) and ST131 (29.2%) were more frequent sequence types with ST101 (9.7%), ST10 (9.7%) and ST648 (7%). The replicon types IncFII, IncFIIK, IncA/C, IncN and IncL/M were detected. The combination of MEM/AgNPs remained effective against carbapenemase-positive E. coli. We reported genetically diverse E. coli strains coharboring carbapenemases/ESBLs from Pakistan. Moreover, this study highlights the enhanced antibacterial activity of MEM/AgNPs and may be used to manage bacterial infections.


Introduction
The repeated exposure of bacterial species to antibiotics results in selection pressure ultimately modifying their antimicrobial physiology. Carbapenem resistance has been accumulated among the Enterobacteriaceae (CRE) due to the significant spread of core carbapenemase-encoding genes present on plasmids or mobile genetic elements such as integrons and transposons. Therefore, carbapenemases presented a stable and transferable form of resistance due to carbapenem hydrolysis and geographic spread such as bla NDM , bla OXA-48 , bla VIM , bla IMP and bla KPC [1,2]. Since the identification of bla NDM , its spread has been commonly observed among Enterobacteriaceae, especially from Asian territories such as Pakistan, India and China. bla IMP -carrying CRE were predominantly detected in Taiwan and Japan with worldwide sporadic reports. Greece was identified as a center of bla VIM -positive CRE leading to the outbreaks in Europe and China. On the other hand, the eruption of bla KPC -producing CRE was recorded in the USA, Europe and China, while Turkey, Europe and the Mediterranean region are the core places for OXA-48-producing CRE [3,4]. This geographical distribution of carbapenemases showed that CRE members have mastered the art of hiding through interchangeable resistance features responsible for the dispersion of carbapenem resistance. Common CRE pathogens responsible for amplified resistance spread through the distribution of carbapenemases, including E. coli, K. pneumoniae and Enterobacter spp. [1,2]. Therefore, the detection and pursuing of such pathogens has been recommended as a critical priority by the CDC and WHO [4].
E. coli is a part of normal human gut microbiota; however, several supremely adapted E. coli clones with multiple resistant genes were shown to accustom the new niches, thus causing far-ranging diseases such as urinary/respiratory tract infections, intestinal infections and sepsis [5,6]. Most studies reported antimicrobial resistance among E. coli in the non-clinical settings from Pakistan such as bla OXA-1 , bla CTX-M15 and bla TEM in chicken meat [7], ST10 with mcr-1 and bla TEM in cattle farm environment [8] and bla NDM , bla OXA-48 , bla TEM and bla SHV from sewage water [9]. However, there are few reports regarding highrisk clones of E. coli with multiple resistance genes in hospital-acquired infections from Pakistan, such as ST131, which has been linked to bla KPC-2 , while ST1196 has been linked to bla NDM-1 [10].
The efficacy of the currently available antibiotics has reduced due to the emergence of resistant bacterial clones, forcing the search for other methods to combat such dangerous clones. In this regard, nanoparticles are attractive candidates due to their stability and biocompatibility, as evidenced by widespread antimicrobial, industrial and biomedical applications [11]. It was shown that nanoparticles, in combination with antibiotics, enhanced the antimicrobial efficiency against resistant microorganisms [12]. A powerful antibacterial response has been observed when myco-synthesized AgNPs in combination with imipenem were used against imipenem-resistant K. pneumoniae isolates. The MIC values of imipenem were reduced for imipenem-resistant K. pneumoniae strains with an FIC index of 0.07 [13]. Biologically synthesized AgNPs from plant extracts have been used with tetracycline against S. aureus and K. pneumoniae, showing significantly increased activity [14]. Another study reported the combined elevated effect of AgNPs with kanamycin [15] and colistin against MDR pathogens [16]. Therefore, the use of AgNPs may be considered as one of the useful therapeutic strategies for treating microbial infections and the reversal of bacterial resistance.
Globally widespread carbapenem-resistant E. coli strains necessitate novel approaches to stop the spread of these dangerous infections. Therefore, continuous surveillance studies and improved antimicrobial usage plans are required so that the proper measures can be adopted to overcome the spread of high-risk clones. The distribution of carbapenemases among Pakistan E. coli is, however, only partially and sparsely studied [17]. Therefore, constant surveillance investigations are required in terms of genotyping, plasmid and sequence typing to address the resistance situation in this country. Keeping this in view, this study aims to investigate the burden of carbapenem resistance, genomic diversity and efficacy of antimicrobials with silver nanoparticles among E. coli clinical isolates from Pakistan, so that applicable strategies can be devised for appropriate infection control. were used to identify ESBL producer strains. The bacterial cultures were characterized by Gram's staining and API-20E (BioMerieux, Marcy-l'Étoile, France).

Detection of Antimicrobial Resistance-Encoding Genes
Genomic DNA was isolated by the heat lysis method [19]. Briefly, 2 to 3 colonies of bacterial culture were mixed with 500 µL sterile distilled water and heated at 98 • C for 10 min at 300 rpm (ThermoMixer, Fischer Scientific, Waltham, MA, USA). Tubes were centrifuged at 1000 rpm for 10 min and supernatant collected in newly labeled tube. DNA was stored at −80 • C. The carbapenemase-encoding genes bla NDM-1 , bla OXA-48 , bla IMP , bla VIM and bla KPC-2 and the ESBLs bla CTX-M , bla SHV and bla TEM were detected by using a 50 µL of PCR reaction mixture containing 25 µL of 2 × PCR Master Mix (Cat # K0171, Thermoscientific, Waltham, MA, USA), 1 µL of each primer (10 µM), 2 µL of DNA and dH 2 O up to 50 µL in PCR (Thermal Cycler, Proflex, ABI). PCR cycling conditions were 95 • C for 40 sec, melting temperature (Tm) 30 sec and 72 • C for 30 sec (30 cycles). Agarose gel electrophoresis (1-1.5%) was used to resolve PCR products. Allelic discrimination of bla NDM was performed by Sanger's sequencing method. The cycle sequencing was performed by using BigDye terminator v3.1 kit with 10 µL PCR reaction mixture containing BigDye terminator 3.1 Ready Reaction Mix 4 µL, forward primer (3.2 pmol) 0.5 µL, purified DNA template (5-20 ng) 2 µL and dH 2 O 3.5 µL. PCR cycling conditions were 96 • C for 1 min, 96 • C for 10 sec, 50 • C for 5 sec and 60 • C for 2 min (35 cycles). PCR product was purified by using BigDye XTerminator purification kit as per kit instructions and capillary electrophoresis was conducted by Genetic Analyzer (ABI-3500, Thermo Fischer, Waltham, MA, USA). Sequencing analysis software v6.1 and basic local alignment tool (BLAST, NCBI) was used for data analysis and interpretation. The primer sequences with Tm are given in Table S1.

Determination of Efficacy of Silver Nanoparticles (AgNPs)
The activity of AgNPs and MEM was evaluated by the broth microdilution checkerboard method. Commercially available AgNPs (particle size: 10 nm, solution concentration 20 µg/mL in aqueous buffer containing sodium citrate as stabilizer) were used (Cat # 730785, Sigma-Aldrich, St. Louis, MO, USA). The dilutions of MEM and AgNPs were prepared in Mueller Hinton broth with final concentration ranges of 1024, 512, 256, 128, 64, 32, 16 and 8 µg/mL for MEM and 10, 5, 2.5, 1.25, 0.625 and 0.312 µg/mL for AgNPs (Table S2). Bacterial cultures were prepared at a concentration of 0.5 McFarland (equivalent to 10 8 CFU/mL) and diluted to 1:100 (10 6 CFU/mL). In sterile 96-well microtiter plate, each well was inoculated with 100 µL of diluted bacterial suspension and mixed with equal volumes of antibiotic solution. All tests were conducted in duplicate with a growth control without addition of antibiotics and with sodium citrate addition. The inoculated microtiter plate was incubated at 37 • C for 18 h. After incubation, the fractional inhibitory concentration index (ΣFIC) was calculated by dividing individual MIC of treatments with MIC of the combination drugs. ΣFIC value ≤1 was considered to have a synergistic, 1.1 to 2.0 indifferent and ≥2 antagonistic effect [22].

Characteristics of Sample
In the current study, the isolates were screened and included on the basis of their carbapenem susceptibility profile. All the isolates were resistant to at least one of the carbapenems (MEM, IMP and ETP). On the basis of this selection criterion, 184 carbapenemresistant E. coli strains were retrieved out of total 650 collected samples. Carbapenemresistant samples were mainly obtained from females (119/184, 64.7%), the major sample type was urine (75/184, 40.8%) and the main hospital section was surgery (59/184, 32.1%), followed by nephrology (36/184, 19.6%). Carbapenemase production was recorded in 64.7% (119/184) of the carbapenem-resistant E. coli isolates, while 35.3% (65/184) did not produce carbapenemase, indicating that other mechanisms are involved for resistance development against carbapenems in these isolates. On other hand, ESBL production was observed in 97.3% (179/184) of the isolates.
Furthermore, the correlation of carbapenemase-resistance genes with sequence types was analyzed, by which we identified different combinations of resistance genes within different detected sequence types. Most importantly, we found that ST405 and ST131 were more prevalent comprising isolates with bla NDM-7 /bla OXA-48 and bla NDM-1 /bla OXA-48 which coharbored different combinations of ESBLs, while ST10, ST101 and ST648 were prevalent among the bla NDM-1 harboring isolates. The distribution of sequence types in relation to resistance genes is given in Table 3.
Carbapenem-resistant strains of E. coli containing carbapenemases (n = 16) were selected randomly for the determination of the AgNPs' efficacy. Carbapenem-sensitive strains of E. coli (n = 6) were used as controls. The MIC values of bacterial cultures grown in the presence of MEM, AgNPs and MEM/AgNPs were determined. The MIC values of the cultures were the highest in the presence of MEM and AgNPs alone. However, when a combination of MEM/AgNPs was used, there was a reduction in the MIC values. The detailed results are described in Table 4.  Table 3. Distribution of sequence types with resistance gene pattern.

Discussion
The eventual outcome of carbapenem resistance among Enterobacteriaceae was evidenced globally by the rapid circulation of plasmid-encoded carbapenemases, making them critical for nosocomial outbreaks. Although carbapenemase detection is essential for infection control purposes, their precise characterization among different species is helpful in clinical practice as it impacts therapeutic decisions. The presence of different clinical strains in the genetic context among E. coli has not been revealed in the study area. In the present study, carbapenemase production was observed in 64.7% of the E. coli isolates. Previously, a variable range of carbapenemase production among E. coli strains has been documented from Pakistan, such as 37.1% in 2015, 93% in 2018, 81% in 2019 and 22.02% in 2021 [5,10,23,24] with the observation of similar trends globally, such as 89% from China, 9.82% from Morocco and 20.2% from Germany [20][21][22]. Similarly, reports from Pakistan showed higher rates of carbapenemase production among other Enterobacteriales such as K. pneumoniae 27.2% (34/125) [10], 61.8% (68/110) [25], 77.7% (91/117) [26] and 82.8% [27]. On the other hand, we found 28.3% of carbapenem-resistant E. coli clinical isolates. Previous studies from Pakistan demonstrated an increasing trend of carbapenem resistance in E. coli with 6% to 37.9% resistance from 2014 until 2018, and thus identified E. coli as a major contributor to the carbapenem resistance in Pakistan [23,28,29]. The leading cause for the growing trend of increased carbapenem resistance is the excessive use of carbapenems that result in the survival of complex clones with highly endured resistant strains [30].
Our results indicate a high resistance to third-generation cephalosporins, while PB displayed a higher susceptibility. In agreement with our results, previous studies from Pakistan also reported higher susceptible rates for colistin and fosfomycin [10,31,32]. On the other hand, avian-derived E. coli isolates from Pakistan showed resistance to SXT, TE, CTX and CAZ, while chicken-originated isolates were more susceptible to chloramphenicol and lower levels of resistance against third-generation cephalosporins [33,34]. Moreover, we found that surgery (32.1%) and nephrology (19.6%) were the main hospital sections responsible for the spread of E. coli infections. However, the ICU has been linked to the dissemination of carbapenem-resistant strains in China [6,35].
Antimicrobial pressure has the ability to single-out clonal lineages and plasmids with resistance determinants, resulting in an enhanced transmission capacity. In our study, the clonal lineage analysis showed the sequence types ST405 and ST131 predominantly coharboring bla NDM-7 /bla OXA-48 and bla NDM-1 /bla OXA-48 , while ST101, ST10 and ST648 were prevalent among bla NDM-1 harboring isolates. Previously, ST131 and ST405 in bla NDM-1 and bla KPC-2 positive E. coli strains with IncH12 and IncN replicon types have been reported, while ST648 has been described in bla NDM-7 containing E. coli isolates. [10,36,40]. ST101 and ST648 were reported in NDM-positive E. coli isolates [41]. On the other hand, ST10 was reported in avian-derived E. coli isolates [33] and ST131 among poultry birds from Pakistan [42], while we observed ST10 among clinical isolates for the first time. Furthermore, we detected IncFII, IncFIIK, IncA/C, IncN and IncL/M replicon types among our study isolates. Other replicon types reported from Pakistan include IncL/M, IncA/c, Inc and IncF-II [26,33,36,43,44].
Regardless of the significant efforts for the improvement and control of infections, carbapenem-resistant bacteria remain an alarming threat. Thus, few treatment options are left due to limited resources. Mostly, β-lactam antimicrobials showed inconsequential treatment effects against carbapenem-resistant microbes [45]. Emerging clinical evidence suggests that treatment with combination therapy may be beneficial against carbapenemresistant pathogens [46]. Since ancient times, silver is known for its antimicrobial effects; therefore, in order to overcome the resistance development by the extensive use of antibiotics, silver nanoparticles can be used as an alternative approach to antibiotic combination therapy against MDR organisms [15,47,48]. Our data indicated that the combination of MEM/AgNPs resulted in the reduction of MIC values as compared to the presence of MEM and AgNPs alone against NDM-positive E. coli isolates. It has been shown that AgNPs and ciprofloxacin have better antimicrobial efficiency against E. coli [49]. Moreover, it has been suggested that biosynthesized AgNPs may work as antimicrobials to control E. coli infections [50][51][52][53]. However, we reported for the first time the effect of AgNPs in combination with MEM against E. coli clinical isolates.

Conclusions
We reported the co-existence of bla NDM-7 /bla OXA-48 and bla OXA-48 /bla VIM in E. coli isolates from Pakistan with a novel ST405 E. coli strain coharboring bla NDM-7 /bla OXA-48 / bla CTX-M /bla SHV /bla TEM . Moreover, ST10 was identified in clinical isolates coharboring bla NDM-1 /bla CTX-M /bla SHV /bla TEM for the first time. The resistance pattern observed in our study suggested that surprisingly powerful strains evolved in Pakistan with time, which may indicate a complicated survival mechanism, particularly in the scenario wherein antibiotics misuse is rising. However, due to funding issues, we could not explore other resistance mechanisms and invasive genes. Moreover, a large number of strains may be tested further for AgNPs synergism by time kill assay. Our results also show that the antimicrobial efficacy can be improved when used in combination with silver nanoparticles.