Investigating the molecular transmission dynamics of bla_NDM in antibiotic-selective environments

ABSTRACT

Carbapenem resistance mediated by bla_NDM-encoded metallo-beta-lactamases is often linked to ISAba125, an insertion sequence from the IS30 family, which is widely distributed among critical and high-priority bacterial pathogens. The rapid dissemination of ISAba125-linked bla_NDM in both nosocomial and community-acquired infections presents a serious challenge to healthcare systems and pharmaceutical industries. Despite the urgency of this issue, the factors driving bla_NDM spread and the molecular mechanisms governing ISAba125 mobility remain poorly understood. In this study, we engineered the genomes of Vibrio cholerae and Escherichia coli to investigate the mobility of bla_NDM under controlled conditions both with and without the genetically linked ISAba125. We also examined the transmission efficiency and the stability of bla_NDM in environments with and without sublethal antibiotic concentrations. Our in vitro findings were validated in a rabbit ileal loop model. The results revealed that antibiotic pressure significantly influences the mobility of bla_NDM, shedding light on the molecular dynamics of its transmission. These insights are crucial for developing strategies to curb the spread of bla_NDM and mitigate the growing threat of carbapenem resistance in bacterial pathogens.

IMPORTANCE

Insertion sequences are the simplest form of mobile genetic elements that play a critical role in the adaptation of bacteria, allowing them to rapidly acquire new traits like resistance genes that enhance their survival. ISAba125 is one such insertion sequence that facilitates the spread of bla_NDM, contributing to the global challenge of carbapenem resistance. In this study, we developed reporter strains that could be used as a valuable tool for investigating the dynamics of ISAba125-linked bla_NDMsh-ble and evaluated the transposition frequency of ISAba125-linked bla_NDMsh-ble in the presence and absence of sublethal concentration of antibiotics. Our results demonstrated that ISAba125 enhances the spread of bla_NDMsh-ble under sublethal concentration of antibiotics that induces SOS response.

INTRODUCTION

Antimicrobial resistance (AMR) poses a growing and urgent threat to both human and animal health, with the recent rise of bacterial pathogens that produce carbapenemases further complicating treatment options. Carbapenems, being one of the last-resort antibiotics, are crucial in combating severe hospital-acquired infections, making this emergence particularly concerning (1). The New Delhi metallo-beta-lactamases (NDM), which fall under the Amber Class B beta-lactamase, are the most widely distributed carbapenemases in enteric pathogens conferring resistance to nearly all beta-lactam antibiotics, except aztreonam. Previously, we demonstrated the difference in genetic arrangements between bla_NDM-1 and bla_NDM-5 across different organisms and compared the different mobile genetic elements (MGEs) associated with these genes. Our findings revealed the association of bla_NDM genes to IS30-like ISAba125 or IS91-like ISVSa3 insertion sequences (2). Insertion sequences (IS) are small, highly recombining, and widely distributed transposable elements that are among the most frequent MGEs, facilitating genetic exchange between pathogens and enhancing their adaptability to diverse environments (3). These elements are integrated into chromosomes as well as on plasmids, underscoring their critical role in the dissemination of associated genes. The ISAba125 element commonly associated with the Tn125 transposon typically contains terminal inverted repeats (TIRs) at both ends, encoding integrase, recombinase, and transposase enzymes crucial for chromosomal integration and gene transfer (4). Moreover, it has been reported that a decanucleotide, 5′-GAGATAATTG-3′ in the right end of the ISAba125 insertion sequence enhances the transposition activity in Escherichia coli and exhibits a preference for “natural” hot spots characterized by a 24 bp symmetric consensus in the bacterial genome (5, 6). Additionally, several other insertion sequences in close vicinity of the ISAba125 are known to influence its transposition efficacy and gene expression (7). Thus, among the MGEs implicated in bla_NDM dissemination, ISAba125 has emerged as a particularly influential element.

Clinically significant antibiotics can directly and indirectly influence the transmissibility and stability of resistance genes in bacteria (8). Previous studies have shown that sublethal antibiotic concentrations in the environment act as a major driver of resistance development, promoting mutations that enhance bacterial fitness (9, 10). However, the mechanisms through which sublethal antibiotic concentrations promote horizontal gene transfer (HGT) remain poorly understood. Understanding the impact of sublethal antibiotic exposure on MGEs and their transmission dynamics is important for developing effective strategies to combat the spread of AMR. This study aims to investigate how the MGE ISAba125 influences the propagation of bla_NDM in bacterial communities exposed to both normal and sublethal antibiotic concentrations, emphasizing the urgent need for responsible antibiotic use and novel approaches to mitigate resistance development.

Vibrio cholerae is a unique gram-negative pathogen that occupies both aquatic environmental niches and human hosts during its life cycle. This dual lifestyle brings it into frequent contact with a wide range of microbial communities, including closely and distantly related species. Such interactions increase the likelihood of horizontal gene transfer (HGT), particularly involving MGEs like insertion sequences, integrons, and transposons. Among these, the ISAba125 element associated with the bla_NDM gene is of particular concern due to its strong linkage with carbapenem resistance—a growing threat in clinical settings globally. The presence of the bla_NDM-sh-ble element in V. cholerae is particularly noteworthy given the bacterium’s genomic structure. Unlike many other gram-negative pathogens that typically possess a single circular chromosome, V. cholerae contains two chromosomes. This genomic organization not only reflects the evolutionary divergence of V. cholerae but also potentially offers multiple integration sites for MGEs. The possibility that ISAba125-bla_NDM-sh-ble-linked elements could be stably maintained across both chromosomes increases the risk of dissemination and persistence of antibiotic resistance within the species and across microbial populations. Furthermore, the high sequence similarity of the ISAba125-bla_NDM-sh-ble element in V. cholerae to similar elements found in other gram-negative pathogens, such as Klebsiella pneumoniae, E. coli, and Acinetobacter baumannii, suggests a shared evolutionary trajectory or common HGT mechanisms. This highlights the potential role of V. cholerae as both a reservoir and a vector for the spread of multidrug resistance genes across different ecological and clinical settings.

Our results demonstrated that ISAba125 enhances the spread of bla_NDM sh-ble under sublethal concentrations of antibiotics. Notably, the study reveals that ISAba125-linked bla_NDM sh-ble genes can integrate into a diverse range of chromosomal loci, particularly near actively transcribed operons likely to take advantage of the elevated transcriptional activity for enhanced gene expression. Furthermore, this is the first study to systematically quantify the influence of the ISAba125 copy number on the stability and transmissibility of resistance genes both in vitro and in vivo.

RESULTS

Genome engineering of V. cholerae N16961 for monitoring ISAba125 mobility linked to bla_NDM sh-ble

In our previous work, through comprehensive sequencing and bioinformatic analysis, we identified a significant association between the insertion sequence ISAba125 and the bla_NDM-sh-ble resistance gene cassette in several clinically important bacterial pathogens (2). As depicted in Fig. 1, we present a schematic representation of the diverse genetic organizations of the ISAba125 element in association with the bla_NDM-sh-ble gene cassette across various bacterial species. Despite advancements in understanding the role of the transposase activity in the dissemination of ISAba125-associated bla_NDM sh-ble resistance genes in Enterobacteriaceae and non-fermenter pathogens, the rate of mobility—encompassing both excision and integration of these elements under normal and antibiotic pressure—remains largely unexplored.

Fig 1.

Schematic representation of the genetic organization of the ISAba125 element associated with the bla_NDM sh-ble gene cassette in high-priority bacterial pathogens. The ISAba125 element is genetically linked to the bla_NDM sh-ble gene cassette, illustrating the diverse genetic configurations across various pathogens.

To address this, we initially constructed an array of genetically engineered V. cholerae reporter strains harboring the ISAba125 element linked with the metallo-beta-lactamase encoding bla_NDM gene both independently and in combination with other closely associated insertion sequences. For the construction of the reporter strains, we utilized a highly sensitive V. cholerae strain, N16961, devoid of resident ISAba125 or bla_NDM sh-ble. The recombinant vectors were constructed by amplifying ISAba125, including its TIRs region, along with bla_NDM sh-ble resistance genes and its native promoter from the genome of different clinical isolates using specific set of primers (Table S1). These amplicons were then cloned into two vectors: a replicative temperature-sensitive suicide vector pFX524 having pSC101 origin of replication; and an in-house constructed integrative vector pBD60 (Table S2). The pFX524 vector was chosen for this study for two key reasons. First, it is temperature-sensitive, allowing for its removal at a non-permissive temperature and exhibiting low insertion bias, which facilitates random excision and integration of the insertion sequence along with the bla_NDM resistance gene into various genomic sites under temperature stress. Second, the pFX524 vector carries the R6K origin of replication, enabling plasmid replication solely in bacterial hosts capable of producing the lambda pir protein. Since the recipient cells do not produce the pir protein, the plasmid will not be maintained in the recipient strains, and only those recipients that have the insertion sequence-linked bla_NDM gene integrated into their chromosome grow under antibiotic selection. The advantage of using the pBD60 vector is that it integrates site specifically at the ‘dif’ site of the recipient bacterial chromosome and is stably expressed. The dif site is a conserved chromosomal locus where XerC and XerD recombinases mediate site-specific recombination to resolve chromosome dimers; it comprises two 11 bp binding regions flanking a 6 bp overlap site, facilitating strand exchange and integration of mobile genetic elements into bacterial chromosomes. The reporter strains conjugated with both recombinant pFX524 and pBD60 vectors were used to study the excision/integration frequency and in vitro stability of ISAba125-linked bla_NDM sh-ble by plasmid curing and serial passaging experiments. The different reporter strains thus constructed are listed in Table S2.

Transposition efficiency of ISAba125 in different genetic backgrounds

To examine the transposition efficiency of ISAba125 from plasmid to chromosome, a series of reporter strains (N:SK4, N:SK5, N:SK6, N:SK7, N:SK8, E:SK4, E:SK6, E:SK7 and E:SK8) carrying the bla_NDMsh-ble resistance cassette linked to ISAba125 along with its right- (RIR) and left-inverted repeats (LIR) were constructed by cloning the desired DNA fragments into pFX524. Later on, the recombinant pFX524 vectors were subjected to a plasmid curing assay. Plasmid curing was done at a non-permissive temperature of 42°C. Although no transposition was observed from plasmid to chromosome in either V. cholerae or E. coli derivatives, differences in the plasmid curing efficiency between the two species were noted, with V. cholerae strains curing the plasmid faster than E. coli derivatives. This was confirmed, as the plasmid cured reporter strains did not grow in agar plates supplemented with either ampicillin, imipenem, or zeocin. Interestingly, a difference in the plasmid curing efficiency was identified between the V. cholerae and E. coli reporter strains, with the complete curing of plasmid confirmed at 96 h for the V. cholerae derivatives and 108 h for the E. coli derivatives.

Effect of the copy number of transposases on bla_NDM sh-ble mobility

A key finding from this study was the impact of the copy number of ISAba125 on the expression and excision frequency of the bla_NDM sh-ble gene. Reporter strains with a single copy of ISAba125 (monomer) exhibited higher resistance to antibiotics compared to strains with two copies (dimer) of ISAba125 (Fig. 2). Additionally, a gene excision assay was done using reporter strains at sub-lethal concentrations of different antibiotics to evaluate its role in transmissibility. The different antibiotics and their sub-lethal concentrations used in the experiment are listed in Table S3. Interestingly, it was observed that N:SK9, which harbored a single copy of ISAba125, exhibited higher MIC than that of N:SK10, which had two copies of ISAba125 against the tested beta-lactam antibiotics (Fig. 2). Interestingly, while no significant difference was observed in the excision frequency between the two strains under normal conditions, the strain with a single copy of ISAba125 demonstrated a higher excision frequency when exposed to sublethal concentrations of antibiotics like mitomycin C, ciprofloxacin, and gentamicin (Table 1). This suggests that the copy number of ISAba125 affects its mobility and may influence the transmissibility of resistance genes. Spontaneous mutations which may lead to alteration in the catabolic pathway could potentially result in concomitant adaptation of phenotypic traits, such as gene inactivation, gene transfer, and changes in gene expression (11). Also, this could be due to auto-repression of the transposase at high concentration like those reported for Tn10, Tn5, and IS10 (12). Nevertheless, this report highlights the impact of the copy number of ISAba125 in the expression and excision frequency of the bla_NDM sh-ble gene.

Fig 2.

Minimum inhibitory concentration of beta-lactam and zeocin antibiotics against N16961 (WT), NZ-1, NSK9, and NSK10, which are N16961-derived reporter strains carrying either bla_NDM sh-ble or bla_NDM sh-ble along with single or two copies of ISAba125, respectively.

TABLE 1.

Excision frequencies of ISAba125 transposon-linked bla_NDM sh-ble in V. cholerae reporter strains NZ-1, N:SK9, and N:SK10 in vitro^a

Antibiotic treatment	Strains (host)	Total CFU/mL	Screened colonies	Mean ± SD (%)	Excision frequency (%)
Untreated	NZ-1 (Tnp)	1.20 * 109	216	NDc	ND
	NSK:9 (monomer)	2.98 * 109	227	ND	ND
	NSK:10 (dimer)	2.03 * 109	202	ND	ND
Ciprofloxacin (CIP)b	NZ-1 (Tnp)	2.49 * 109	241	ND	ND
	NSK:9 (monomer)	1.05 * 109	319	1.5 ± 0.94	0.009114
	NSK:10 (dimer)	1.20 * 109	196	ND	ND
Doxycycline hyclate (DOX)	NZ-1 (Tnp)	1.33 * 109	163	ND	ND
	NSK:9 (monomer)	1.06 * 109	195	ND	ND
	NSK:10 (dimer)	1.02 * 109	158	ND	ND
Gentamicin (GEN)	NZ-1 (Tnp)	1.22 * 109	183	ND	ND
	NSK:9 (monomer)	1.09 * 109	144	1.0 ± 1.39	0.002642
	NSK:10 (dimer)	1.05 * 109	151	ND	ND
Kanamycin (KAN)	NZ-1 (Tnp)	1.11 * 109	320	ND	ND
	NSK:9 (monomer)	2.01 * 109	275	ND	ND
	NSK:10 (dimer)	2.02 * 109	288	ND	ND
Mitomycin C (MMC)	NZ-1 (Tnp)	2.56 * 109	369	0.5 ± 0.81	0.004324
	NSK:9 (monomer)	2.77 * 109	353	3.5 ± 1.98	0.008921
	NSK:10 (dimer)	2.69 * 109	246	ND	ND
Neomycin (NEO)	NZ-1 (Tnp)	2.19 * 108	140	ND	ND
	NSK:9 (monomer)	2.32 * 108	103	ND	ND
	NSK:10 (dimer)	2.00* 108	110	ND	ND
Rifampin (RIF)	NZ-1 (Tnp)	2.36 * 109	193	ND	ND
	NSK:9 (monomer)	2.81 * 109	200	ND	ND
	NSK:10 (dimer)	3.22 * 109	190	ND	ND
Tetracycline (TET)	NZ-1 (Tnp)	1.71 * 109	145	ND	ND
	NSK:9 (monomer)	1.89 * 109	113	ND	ND
	NSK:10 (dimer)	1.72 * 109	110	ND	ND
Trimethoprim (TMP)	NZ-1 (Tnp)	1.06 * 109	184	ND	ND
	NSK:9 (monomer)	1.91 * 109	166	ND	ND
	NSK:10 (dimer)	1.22 * 109	132	ND	ND

^a Tnp—transposon; monomer—single copy; dimer—two copies.

^b Boldface indicates experiments in which excision efficiency was determined.

^c “ND” indicates not detected.

Identification and functional analysis of integration sites for ISAba125 in the V. cholerae genome

A chitin-induced natural transformation was conducted using V. cholerae clinical isolate C6709 and lab-generated N16961:HapR⁺ (streptomycin-sensitive) strains to investigate environmental acquisition of the ISAba125-bla_NDM-sh-ble element. Both HapR⁺ strains initially lacking the bla_NDM-sh-ble element successfully acquired and chromosomally integrated the gene linked to ISAba125 following natural transformation with 10 µg of V. cholerae IDH06781 genomic DNA (Fig. 3A). Antibiotic susceptibility profiling confirmed that transformants exhibited resistance to β-lactam antibiotics, including penicillin, ampicillin, carbenicillin, and imipenem, compared to their wild-type counterparts (Fig. S1), supporting the role of natural transformation in disseminating clinically relevant resistance determinants in V. cholerae under environmental conditions.

Fig 3.

Transposition efficiency of the ISAba125-linked bla_NDM sh-ble genes in different V. cholerae strains. (A) Chitin-induced natural transformation assay to investigate the transfer of the ISAba125-linked bla_NDM sh-ble genes from V. cholerae IDH06781 strain to bla_NDM-negative V. cholerae strains, HapR⁺ N16961 and C6709. (B) Genetic arrangement of the integration site of the ISAba125-linked bla_NDM sh-ble in V. cholerae strains, HapR⁺ N16961 and C6709 transformants of the chitin assay. The ISAba125 sequence was identified through BLAST in bla_OXA-10 (yellow) sequence. Cluster alignments are drawn to scale using the Clinker tool version 0.0.26. Arrows indicate the direction of the open reading frame.

To investigate the integration site of the ISAba125 element linked to the bla_NDM-sh-ble gene cassette in the V. cholerae genome, N16961 HapR⁺ and C6709 transformants were subjected to whole-genome sequencing. The analysis of the genome sequences of both the transformants revealed the integration site of the bla_NDM sh-ble cassette into its chromosome. The analysis also identified a 130 bp nucleotide sequence homologous to the attL and attR attachment sites of ISAba125 in C6709’s genome, specifically between the 23S and 16S rRNA genes in chromosome I of the isolate (Fig. 3B). The integration site of the ISAba125-bla_NDM-sh-ble element in the N16961:HapR⁺ transformant could not be resolved using Illumina short-read sequencing due to the fragmented nature of the assembly and the small size of contigs carrying the resistance cassette. To precisely identify the chromosomal integration site, whole-genome sequencing was performed using Oxford Nanopore Technology (ONT), which generates long reads. These long reads were hybrid-assembled with the Illumina short reads to reconstruct a complete, high-quality genome assembly, which was then aligned to the V. cholerae N16961 reference genome. Unlike the C6709 transformant, where the ISAba125-bla_NDM-sh-ble element integrated into chromosome I, in the N16961 transformant, the cassette was found integrated into chromosome II proximal to the infC-rpmI-rplT operon. This operon encodes essential proteins involved in translation and ribosome assembly and is known to be actively transcribed in V. cholerae, similar to the rRNA operons—suggesting a potential preference for integration near transcriptionally active regions.

To assess the functional consequences of this integration near actively transcribed operons, qRT-PCR analysis was conducted on C6709 transformants with the bla_NDM sh-ble linked to ISAba125 integrated either at the 23S rRNA-16S rRNA operon or at the ‘dif’ site using specific primers bla_NDM-F/bla_NDM-R (Table S1). Results revealed a approximately six fold higher expression of the bla_NDM gene when integrated at the 23S rRNA–16S rRNA operon compared to the ‘dif’ locus (Fig. 4A and B). This difference was further supported by CFU and turbidity assays under antibiotic pressure, where the transformant with integration at the 23S rRNA–16S rRNA operon exhibited significantly higher bla_NDM expression at earlier time points (Fig. 4C and D). These findings highlight the influence of the genomic integration site on the expression and functionality of resistance genes in V. cholerae.

Fig 4.

Activity of bla_NDM sh-ble in accordance with the site of integration into the genome of transformants. (A) Schematic representation of genomic locations of bla_NDM sh-ble in different V. cholerae strains (WT, reporter strains, and transformants). (B) Transcript analysis of the bla_NDM gene reporter strains and transformants having the bla_NDM sh-ble integrated at different genomic locations. The experiment was done in triplicates (n = 3). The expression levels of the target genes were normalized to that of the housekeeping gene, recA. Data were analyzed by unpaired t-test; values are expressed as mean ± SEM of all the three experiments using the GraphPad Prism version 10.9. (C and D) Graph illustrating the growth dynamics of C6709 (WT) and its transformants with bla_NDM sh-ble integrated at different genomic locations, as assessed by colony-forming units and turbidity assay (OD₆₀₀) at 2 and 16 h in the presence of ampicillin and imipenem antibiotics.

Analysis of sublethal concentration of antibiotics as an inducer of ISAba125 linked with bla_NDM sh-ble transmission

The role of sublethal antibiotic concentrations in inducing the transmission of ISAba125 linked with bla_NDM sh-ble was explored, focusing on their impact on the SOS response and gene transfer. Excision assays using the N:SK9 and N:SK10 reporter strains under sublethal antibiotic pressure (mitomycin C, ciprofloxacin, and gentamicin) demonstrated a significant increase in the excision frequency of ISAba125-linked bla_NDM. To further investigate, qRT-PCR analysis was performed to evaluate the expression of key SOS response genes (recA, recN, lexA, and yebG) in the presence and absence of sublethal antibiotic concentrations. The results revealed that exposure to antibiotics like ampicillin and gentamicin led to increased expressions of recA, recN, lexA, and yebG in N16961 (Fig. 5A and B). Additionally, exposure to ciprofloxacin and gentamicin enhanced the expressions of recN, lexA, yebG, and recA, respectively, in C6709.

Fig 5.

Relative gene expression levels of SOS genes in (A) N16961 and (B) C6709 under pressure of sublethal concentration of antibiotics compared to untreated (UT) control. The expression levels of the target genes (recA, recN, lexA, and yebG) were normalized to that of the housekeeping gene, rpoB. Statistical analysis was performed by unpaired t-test, and the data are presented as mean ± SEM (n = 3) (*P < 0.05, **P < 0.001, ***P < 0.0001) using GraphPad Prism version 10.9.

To confirm that these antibiotic-induced changes were associated with the spread of ISAba125-linked bla_NDM sh-ble, chitin-induced natural transformation assays were performed. The results indicated that sublethal concentrations of ampicillin and gentamicin enhanced the transformation efficiency and dissemination of bla_NDM sh-ble in N16961. This suggests that the SOS response activated by these antibiotics promotes the transmission of bla_NDM sh-ble (Fig. S2). However, no significant increase in transformation efficiency was observed for the C6709 strain under sublethal antibiotic pressure. These findings indicate that the effect of antibiotics in the spread of resistance genes primarily through SOS response activation may vary depending on the genetic makeup of the bacterial species.

In vivo transmission and stability of ISAba125 linked with bla_NDM sh-ble in V. cholerae in a rabbit model

To assess the in vivo stability of the bla_NDM sh-ble resistance gene associated with ISAba125, we used a rabbit ileal loop model. The use of the rabbit ileal loop model provides a physiologically relevant in vivo system for studying V. cholerae colonization, gene expression, and interaction with host tissues under conditions that closely mimic the human intestinal environment. In the context of studying the dynamics of ISAba125-bla_NDM-sh-ble element, this model is particularly valuable for assessing the stability, expression, and possible fitness costs or benefits of carrying the MGE in a host-associated context. This model also allows for the evaluation of the potential impact of the genetic element on the virulence or colonization ability of V. cholerae, which may differ significantly in vivo compared to in vitro observations. Given the global health implications of bla_NDM dissemination, understanding how such elements behave and persist within V. cholerae during infection can provide critical insights into the transmission dynamics and evolutionary fitness of antibiotic-resistant strains.

Reporter strains carrying the bla_NDM-sh-ble-ISAba125 construct were inoculated into ligated rabbit ileal loops and incubated for 16–18 h. The fluid from the ileal loops was collected, and CFUs were counted by plating on LA agar and TCBS selective plates. The presence of bla_NDM sh-ble was confirmed through resistance profiling and PCR amplification using gene-specific primers.

In vivo experiments revealed a higher frequency of bla_NDM sh-ble loss in the NZ-1 strain, which lacks the transposon, compared to strains containing the transposon (Table 2). Notably, the NSK:10 strain, which carries two copies of ISAba125, showed no gene loss, consistent with in vitro findings. Gene loss frequencies were determined by plating on selective media containing ampicillin, zeocin, and streptomycin. The frequency of bla_NDM sh-ble loss was significantly higher in the NZ-1 strain under in vivo conditions, as opposed to in vitro conditions. Conversely, the NSK:9 strain exhibited greater gene loss in vitro than in vivo. These results suggest that the presence of multiple ISAba125 copies enhances bla_NDM sh-ble stability in vivo.

TABLE 2.

Excision frequencies of the ISAba125 transposon-linked bla_NDM sh-ble in V. cholerae reporter strains NZ-1, N:SK9, and N:SK10 in vivo^a,b

In vivo	Strains (host)	Total V. cholerae CFU/mL	Total V. cholerae screened colonies	Mean ± SD (%)	Excision frequency (%)
Infection incubation period 18 h	NZ-1 (Tnp)	2.8 * 107	1230	14 ± 2.276	12.300000
	NSK:9 (monomer)	1.1 * 107	951	1 ± 0.210	1.729091
	NSK:10 (dimer)	7.0 * 107	1110	ND	ND

^a Tnp—transposon; monomer—single copy; dimer—two copies.

^b “ND” indicates not detected.

DISCUSSION

Mobile genetic elements (MGEs) are pivotal in the horizontal transfer of resistance genes among bacterial populations. Among these, insertion sequences (ISs) are the simplest and most commonly encountered MGEs, playing a significant role in the dissemination of resistance genes. Notably, previous studies have highlighted a strong genetic association between the insertion sequence ISAba125 and the carbapenemase gene bla_NDM. ISAba125 has been identified as a key contributor to the rapid spread of bla_NDM, underscoring its critical role in the expansion of antimicrobial resistance (2, 4, 13). This study aimed to investigate the mobility and integration of the ISAba125 element linked to the bla_NDM sh-ble resistance gene cassette, specifically in V. cholerae N16961. Our previous work demonstrated a strong genetic association between ISAba125 and bla_NDM sh-ble in clinically significant pathogens (2, 14). In this study, we sought to explore the dynamics of ISAba125 mobility, focusing on transposition efficiency, integration sites, and the influence of antibiotic pressure on gene transfer.

V. cholerae was strategically selected as the model organism for investigating the transmission dynamics of the ISAba125-associated bla_NDM-sh-ble element due to multiple compelling factors. Initially, the direct detection of this genetic linkage in V. cholerae isolates from diarrheal patients provided a clinically relevant foundation for the study. Furthermore, emerging evidence indicates that V. cholerae possesses an enhanced ability to acquire and integrate novel antimicrobial resistance determinants, underscoring its role as a dynamic genetic reservoir. Its ecological niche in aquatic environments characterized by extensive microbial diversity and frequent HGT events further positions V. cholerae as a critical conduit for resistance gene dissemination. The clinical significance of V. cholerae as the causative agent of cholera amplifies the potential public health impact of resistance development within this species. Importantly, the well-characterized and experimentally tractable HGT systems in V. cholerae provide a robust and versatile model to dissect the molecular mechanisms underlying the acquisition, integration, and spread of ISAba125-linked bla_NDM-sh-ble. Collectively, these features establish V. cholerae as an ideal chassis for elucidating the transmission dynamics of this mobile genetic element, with direct implications for understanding and mitigating antimicrobial resistance in both environmental and clinical contexts.

We successfully engineered the genome of the V. cholerae strain, N16961, to develop reporter strains that facilitate the monitoring of the ISAba125 mobility linked with the bla_NDM-sh-ble resistance gene. This innovative approach not only provides a valuable tool for investigating the dynamics of antibiotic resistance elements but also enhances our understanding of the mechanisms underlying the spread of bla_NDM among enteric bacteria.

Our analysis of the transposition efficiency of ISAba125 across different genetic backgrounds revealed no significant variability, as none of the V. cholerae N16961 and E. coli reporter strains carrying different combinations of insertion sequences, including IS3/ISAba14, IS911, IS91, and IS630, along with ISAba125, exhibited transposition of the resistance cassette from plasmid to the chromosome. However, a difference was observed in the plasmid stability among the different V. cholerae and E. coli reporter strains. Plasmids were less stable in the V. cholerae strain possibly due to the poor domestication of the exogenous plasmid in the isolate due to the presence of two chromosomes (15). Understanding these interactions is vital for predicting the potential for HGT, particularly in clinically relevant strains.

However, our investigation into the effect of the transposase copy number on the mobility and activity of ISAba125 linked with bla_NDM sh-ble demonstrated that different transposase copy numbers modulate the transposition frequency and activity of the associated resistance genes. This modulation of transposase activity might serve as a strategic approach to control the dissemination of bla_NDM in pathogenic bacteria. The identification and analysis of integration sites for ISAba125 within the V. cholerae genome revealed that it can integrate in a diverse array of loci, emphasizing the plasticity of bacterial genomes in accommodating MGEs. Understanding these integration hotspots is crucial for mapping the pathways of antibiotic resistance gene dissemination and for developing predictive models of resistance evolution. Such insights may inform the design of effective surveillance strategies in healthcare settings.

In vivo stability of the bla_NDM sh-ble cassette linked with ISAba125 was assessed using a rabbit ileal loop model, revealing that strains with multiple copies of ISAba125 exhibited the highest stability of the bla_NDM sh-ble gene compared to strains with a single copy or no transposase. This finding suggests that the presence of multiple copies of ISAba125 enhances gene stability, even in the absence of antibiotic pressure. It is likely that the transposase efficiently reintegrates the excised cassette before cell division, maintaining stability. In contrast, strains lacking transposase exhibited no integration events, leading to a higher frequency of bla_NDM sh-ble cassette loss. However, the precise molecular mechanisms behind the increased excision and loss frequencies in the absence of transposase remain unclear. Similarly, the exact mechanism by which the transposase copy number influences the transposition efficiency could not be fully elucidated in the present study. These observations highlight the need for further investigations to better understand the dynamics of ISAba125 mobility and integration, which could offer valuable insights into controlling the spread of antibiotic resistance gene cassettes in clinically relevant bacterial species. Future studies could focus on evaluating the potential auto-repression of ISAba125. Understanding these mechanisms may be crucial for developing strategies to mitigate the transmission of antimicrobial resistance.

Conclusion

In conclusion, this study reveals the complex dynamics of bla_NDM mobility linked with ISAba125, highlighting the role of antibiotic exposure in modulating transposition efficiency and genetic adaptation. Our findings demonstrate that ISAba125’s behavior is influenced by genetic context, host background, and antibiotic pressure, which can enhance the mobility of resistance genes. The integration of ISAba125 in the V. cholerae genome further suggests its potential for horizontal gene transfer, functioning similarly to site-specific recombination systems. These insights underscore the importance of targeting ISAba125 to combat the spread of antibiotic resistance and emphasize the need for ongoing research to develop effective strategies against resistant pathogens.

MATERIALS AND METHODS

Bacterial strains, plasmids, and growth conditions

The bacterial isolates used in this study are listed in Table S4. For the liquid culture, the strains were grown in Luria-Bertani broth (LB) or Mueller Hinton broth (MHB) at 37°C in a shaker with 180 RPM, while LB agar plates were used for the solid culture. Bacterial strains were preserved at −80°C in LB containing 15% sterile glycerol. Antibiotics (Sigma-Aldrich, USA) were used at the following concentrations, unless mentioned otherwise: ampicillin: 100 µg mL⁻¹; kanamycin: 40 µg mL⁻¹; streptomycin: 100 µg mL⁻¹; and chloramphenicol: 3 µg mL⁻¹ for V. cholerae. Before the initiation of any experiment, bacterial strains were freshly inoculated from the −80°C stock. Bacterial growth was monitored by measuring the optical density (OD) at 600 nm using a biospectrophotometer (Eppendorf, Hamburg, Germany).

Molecular biological methods

Standard molecular biological methods for chromosomal and plasmid DNA preparations, restriction enzyme digestion, DNA ligation, bacterial transformation, conjugation, agarose gel electrophoresis, etc., were followed, unless stated otherwise. Restriction and nucleic acid-modifying enzymes were purchased from New England Biolabs, Inc. (NEB, USA) and used as directed by the manufacturer. The T4 DNA ligase enzyme used in this study was procured from NEB (USA). Ligation reactions were carried out essentially as directed by the manufacturer.

Construction of plasmids and reporter strains

The recombinant plasmids were constructed by the PCR amplification of the bla_NDM sh-ble ISAba125 ORF with its natural promoter (size 2.7 kb) alone and in conjunction with other insertion sequences, as listed in Table S2. The genomic DNA of clinical isolates listed in Table S4 harboring the ISAba125 linked bla_NDM was used as the template. The amplicons were purified, restriction enzyme digested, and ligated into vectors pFX524 and pBD60 (16). Furthermore, the ligation mixture was transformed into FCV14 E. coli cells, and the positive transformants with the recombinant plasmid were selected on ampicillin (100 µg mL⁻¹) plates. The recombinant plasmid was then isolated from the positive transformants and further transformed into the intermediate host E. coli β2163 to conjugate it with N16961 and E. coli ATCC 25,922 cells to generate the genetically engineered reporter strains. The conjugation plate was supplemented with 0.3 mM DAP for selecting β2163 transformants. Exponentially growing (optical density at 600 nm, ∼0.3) donor and recipient cells were mixed in a 1/2 ratio and incubated at 30 and 37°C overnight for the replicative (pFX524) and integrative vectors (pBD60), respectively. The transconjugants were selected on agar plates supplemented with ampicillin and zeocin at concentrations of 100 and 25 µg mL⁻¹.

Transposition assays

To understand the transposition efficacy of ISAba125-linked bla_NDMsh-ble from plasmids, the transposition assay was performed as previously described with some modifications (17). Briefly, N16961 and E. coli ATCC 25922 reporter strains N:SK4, N:SK5, N:SK6, N:SK7, N:SK8, E:SK4, E:SK6, E:SK7, and E:SK8 with the recombinant pFX524 plasmid were initially grown overnight in the LB containing 100 µg mL⁻¹ ampicillin. Overnight cultures were then diluted by a factor of 100 in the absence of antibiotic selection and grown at 42°C, with shaking for approximately 300 generations. At every 18 h, the cultures were serially diluted and spread on LB agar plates and incubated overnight at 30°C. The next day, around 100 colonies were picked at random, patched onto selective media containing ampicillin, and grown overnight at 30°C. After incubation, colonies with optimum growth were counted and confirmed by bla_NDM-specific PCR, and the transposition frequency was evaluated. The transposition of ISAba125-linked bla_NDM from plasmid togenome was confirmed by two sets of PCR reactions. A PCR using plasmid-specific primers was performed to confirm the successful elimination of the plasmid at non-permissive temperature from the bacterial cell. In addition, a PCR using bla_NDM specific primers was performed to confirm a successful integration of the genome into the bacterial chromosome. Additionally, the growth of the plasmid-cured reporter strains was assessed on agar plates supplemented with either ampicillin (100 µg mL⁻¹), imipenem (10 µg mL⁻¹), or zeocin (25 µg mL⁻¹).

Antibiotic susceptibility testing and minimum inhibitory concentration (MIC) assay

The antibiotic susceptibility (AST) profiling of the wild type and transformants was performed by disc diffusion and broth microdilution assays. The MIC of reporter strains and transformants of the chitin assay was determined by broth dilution method of antimicrobial susceptibility testing according to CLSI guidelines (18). Briefly, overnight bacterial culture grown in the MHB medium was diluted at 1:100 in fresh MHB and cultured at 37°C with shaking at 180 RPM to an optical density of OD₆₀₀ 0.5. Furthermore, the cells were diluted at 1:1,000 in fresh MHB again, and 100 µL of this diluted culture was dispensed into each well of a 96-well microtiter polystyrene plate. A series of twofold dilutions of an antibiotic was added into these wells, and the mixtures were incubated at 37°C for 16–18 h.

Chitin-induced natural transformation

Chitin-induced natural transformation was carried out as previously described with few modifications (19). Briefly, the recipient HapR⁺ V. cholerae bacterial strains N16961 (HapR⁺, str^s) and C6709 were grown on 50–80 mg autoclaved chitin flakes (Sigma cat. C9213) immersed in M9 medium supplemented with MgSO₄ and CaCl₂ for 16–20 h at 30°C. Next day, the spent media were replenished with fresh M9 media with MgSO₄ and CaCl₂, and the donor DNA, 10 µg of genomic DNA from a highly resistant clinical V. cholerae isolate, IDH06781, was added as the transforming material. The genomic DNA of the V. cholerae reporter strain, NZ-1, with bla_NDM sh-ble cassette was used as the positive control. Cells were further incubated for 24 h and subsequently detached from the chitin by vigorous vortexing for 30 s to 1 min. The transformants were selected on LB agar plates supplemented with ampicillin (100 µg mL⁻¹) antibiotic. The antibiotics and their sublethal concentrations against the V. cholerae HapR + str^s N16961 and str^r C6709 strains used in chitin assay are presented in Table S5.

Quantitative RT-PCR (qRT-PCR) analysis of SOS (save our soul) gene expression

Real-time reverse transcriptase PCR was performed to detect the differential expression of the bla_NDM gene and the SOS response genes (recA, recN, lexA, and yebG) in the transformants of chitin-induced natural transformation. The primer pairs were designed using Primer 3 software version 0.4.0. For sample preparation, three fresh colonies of each strain were inoculated separately into LB broth and incubated overnight at 30°C with shaking condition (180 rpm). The next day, overnight cultures were diluted 100-fold in 5 mL of fresh LB broth and further grown to mid-log for 1–2 h (OD₆₀₀ = 0.3) at 30°C with shaking at 180 RPM. At this point, sublethal concentrations of antibiotics were added, and the cells were grown to late-log phase to reach OD₆₀₀ = 0.7–0.8 (∼2 × 10⁸ CFU/mL). One milliliter from each culture was aliquoted, and the cultures were washed with cold PBS. Furthermore, the bacterial cells were harvested by centrifugation at 5,000 RPM before RNA extraction using Trizol. First-strand cDNA was synthesized using Qiagen’s QuantiTect Reverse Transcription Kit from a 100 ng RNA template. Real-time PCR was carried out in a two-step method on a QuantiStudio 6 Real-Time PCR Machine of Thermo Fisher Scientific using the Power Track SYBER Green Master Mix (Applied Biosystem). The primers used for qRT-PCR are listed in Table S1. The expression of genes was normalized to housekeeping gene rpoB as internal control. Each assay was performed in technical triplicates and biological duplicates.

Persistence assay

To test the stability of ISAba125-linked bla_NDM in the genome of V. cholerae isolates, reporter strains N:SK9, N:SK10, and NZ-1 were used, which had the gene cassette integrated into the ‘dif’ loci using the pBD60 vector. Serial passaging of the reporter strains in the presence and absence of a sublethal concentration of antibiotics was done to evaluate the excision frequency of the resistance gene linked to ISAba125. Briefly, overnight cultures of the reporter strains were diluted 100-fold in 5 mL of fresh LB medium and incubated in the presence and absence of sublethal concentration of antibiotics at 37°C with shaking at 180 RPM/min for 16 h. The next day, 100 µL of the cultures from each tube was further diluted in fresh LB broth with the same conditions and incubated further. The serial passaging was continued every 12 h and, at each passage, 100 µL of the cultures was also serially diluted and plated on LB agar medium having appropriate antibiotics to determine the CFU. To estimate the excision frequency in vitro, at least 100 colonies derived from each culture were plated onto fresh nutrient agar with or without ampicillin plus zeocin. Clones that had become ampicillin 100 µg mL⁻¹ plus zeocin 12.5 µg mL⁻¹-sensitive were screened by PCR using primers listed in Table S1. Nine different antibiotics primarily acting on cell wall synthesis and permeability were selected for the survival assay. The sublethal concentrations of the antibiotics were optimized using the double-dilution method.

Wholegenome sequencing and bioinformatics analysis

The WGS of the N16961 (HapR⁺, str^s) and C6709 transformants selected from the chitin-induced natural transformation of the IDH06781 genome was performed in-house by a high-throughput Illumina MiSeq sequencing platform using the Nextera XT DNA Library Preparation Kit (Illumina, Inc.). To obtain the long-read sequences of the N16961:HapR⁺ transformant, whole-genome sequencing was performed using Oxford Nanopore Technology. The genomic DNA was end-polished and A-tailed (NEBNext Ultra II End Repair Kit, New England Biolabs, MA, USA). The end-prepared samples were barcoded using Blunt TA ligase master mix (M0367L). The equimolar concentration of the barcoded sample was pooled, and the library was prepared by ligating sequencing adapters (SQK-NBD114.96 Kit (Oxford Nanopore Technology, UK) onto double-stranded DNA fragments using NEB Quick T4 DNA ligase (New England Biolabs, MA, USA). After purification with AMPure XP beads, the prepared library was ready for library quality control.

The experimental procedures of the DNA library preparation

The library was checked with Qubit for quantification and pooled and sequenced on Nanopore PromethION System (PromethION P24 and Data Acquisition Unit, ONT, Oxford, UK) using a PromethION flow cell (FLO-PRO114M) according to effective library concentration and the data amount required. High-quality raw reads were used for assembly using a hybrid assembly pipeline Unicycler version v0.4.8 (20). The assembled genome quality was assessed for contamination using the CheckM Program version v1.1.3 (21). Genomes with contamination levels < 5% were included for analysis. A hybrid genome assembly was performed using Unicycler, which combines both short- and long-read sequencing data to generate high-quality contigs. The resulting de novo assembled genome was aligned against the reference V. cholerae strain N16961 using minimap2 (v2.26). The SAM file obtained from the alignment was subsequently converted to a sorted BAM format and indexed using SAMtools. To validate the presence and determine the precise coordinates of the bla_NDM insertion, we conducted a nucleotide BLAST (BLASTn) of the gene against the assembled genome. Then, Prokka and Rapid Annotation using Subsystem Technology (RAST) version 2.0 were used for the functional annotation of the draft assembly, enabling the identification of coding sequences and the annotation of the flanking regions surrounding the insertion (22).

Rabbit ileal loop assay

The rabbit ileal loop assay was carried out on young New Zealand white rabbits weighing approximately 2.0 kg. The rabbits were starved for 24 h before the surgery was done, but water was given ad libitum. V. cholerae strains N16961, NZ-1, NSK:9, NSK:10, and NSK12 were cultured in LB until the mid-log phase (OD₆₀₀ 0.4 to 0.5) corresponding to 10⁸ CFU/mL, followed by washing with sterile phosphate-buffered saline (PBS) and resuspended in 1 mL PBS. After administering anesthesia (ketamine and xylazine, as per bodyweight) and performing a laparotomy, the ileum of the rabbits was sectioned into separate ligated loops, each measuring approximately 10 cm and interloops of 2 cm. V. cholerae strains N16961, NZ-1, NSK:9, NSK:10, or NSK12 were then inoculated into the ligated loops separately, and the loop containing only PBS was used as the negative control. The intestine was repositioned into the peritoneum, and the abdomen was sutured back. After 16 to 18 h post-infection, the rabbits were euthanized; the abdomen was opened; and the length and volume of fluid that had accumulated in each loop were measured. The ileal fluid accumulation (FA) ratio was calculated by the following formula: volume of fluid accumulated (mL)/length of intestine (centimeters). The FA ratio of greater than 1 was considered a strong positive response, while that less than 0.2 indicated a negative response. To determine the in vivo excision frequency, the ileal fluid was first enriched overnight in LB broth at 37°C with shaking at 180 RPM. The following day, serial dilutions (10⁰–10⁶) were performed and plated onto LA plates. Subsequently, 1,500 colonies from each culture were screened onto fresh agar plates with or without ampicillin and zeocin. Clones that had become sensitive to ampicillin 100 µg mL⁻¹ and zeocin 12.5 µg mL⁻¹ were further confirmed by PCR using primers listed in Table S1. The experiment was done in biological triplicate in three rabbits, and the average of the independent results was compiled.

Statistical analysis

All statistical analyses were performed using GraphPad Prism version 10.9. (GraphPad Software, Inc.) and display the experimental data of at least two independent experiments performed in triplicate, displaying the mean and the standard error of mean (SEM) throughout the manuscript, except for animal-related experiments, which display mean ± 95% confidence intervals. Student’s t-test was performed for analyzing the results. Statistical significance was accepted at P < 0.05. ^∗P < 0.05, ^∗∗P < 0.01, and ^∗∗∗P < 0.001.

ETHICS APPROVAL

Animal infection studies were carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the ICMR-National Institute for Research in Bacterial Infections (NIRBI) Kolkata. Protocols were approved by the Institutional Animal Care and Use Committees (IACUC) at the NIRBI Kolkata with the animal ethics approval number PRO/2025/-Jan/2025-2028. All efforts were made to minimize the suffering of animals employed in this study.

DATA AVAILABILITY

Whole genome sequences are available in the National Centre for Biotechnology Information (https://www.ncbi.nlm.nih.gov/) under Bioproject ID PRJEB90998 and Indian Biological Data Centre (https://ibdc.dbtindia.gov.in) under accession number INRP000349.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/jb.00133-25.

Supplemental tables and figure legends. jb.00133-25-s0001.pdf.

Tables S1 to S5 and legends for Fig. S1 and S2.

Figure S1. jb.00133-25-s0002.tiff.

Heat map showing antibiotic susceptibility profiles of V. cholerae.

Figure S2. jb.00133-25-s0003.tiff.

Transformation efficiency of N16961 and C6709 in different antibiotic pressures.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.