The study provides a novel insight into the prevalence of blaNDM−4 gene, which was mediated by rare plasmid types, classified as IncFII(K)-repB(R1701). Additionally, it highlights the coexistence of OXA-9 and NDM-4 on individual plasmids, specifically pL2388-NDM-OXA and pL3835-NDM-OXA in K.pneumoniae. Furthermore, we report the first instance where a single plasmid harbors two inverted tandem repeats of the blaNDM−4 gene. These findings underscore the crucial need for robust surveillance methods to hinder the proliferation of antibiotic-resistant bacterial strains. Furthermore, they raise concerns regarding the potential escalation of resistance and dissemination of antibiotic-resistant bacteria.
NDM-4 was first detected in 2012, Indian [1] and 2014 in China [18].Possessing a high ability to hydrolyze carbapenems and several bulky cephalosporins [1]. OXA-9, exhibits heightened resistance to ceftazidime and various β-lactam antibiotics, first reported in 1993 [19], with a relatively low incidence rate observed in China. In this study, we present the genome sequences of two carbapenem-resistant K. pneumoniae strains, L2388 and L3835, which were isolated from distinct hospitals in Hangzhou, China. These strains exhibited resistance to a broad spectrum of tested antibiotics (Table 2).
Analysis of the genetic context of NDM-4 revealed that the predominant sequence was a 5,380 bp fragment designated as blaNDM−4-ble-trpF-dsbD-cutA-GroES-groL, consistent with previous findings and exhibiting relative conservation. Although the surrounding MGEs varied, most plasmids from GenBank were flanked by IS91 and a combination of IS30 and IS26, aligning with prior reports of ISAba125, IS5, IS26, or IS91 associated with blaNDM−4 [8, 11, 20]. Notably, in two K. pneumoniae isolates from Hangzhou, within plasmid pL2388-NDM-OXA, the NDM-4 gene was bordered by IS15 at both ends. Of particular interest, in pL3835-NDM-OXA, we identified dual copies of blaNDM−4 arranged in inverted tandem, embedded within a Tn3-hypothetical protein-IS26-blaNDM-4-ble-trpF-dsbD-cutA-GroES-groL-hypothetical protein structure (the red box region in Fig. 2B, Fig. 3). Each upstream region of blaNDM-4 featured a Tn3-hypothetical protein-IS26, also linked to IS15. This architecture suggests heightened mobility of these regions, implying that MGEs may exert profound effects on genomic dynamics, a phenomenon warranting further attention.
IS26 was found to be closely associated with many other MGEs, such as Tn3 and class 1 integron, and with numerous different ARGs [21], such as deletions, inversions, and duplications through homologous recombination or non-homologous end joining mechanisms [22]. It plays a pivotal role in disseminating ARGs. While Tn3, located at the termini, does not directly participate in the inverse repetition of NDM-4, may presence as a major transposable element can augment genomic instability [23].
After separately delete the ‘red box’ region, BLASTn showed with 99.5% identity and 97% coverage to each of following plasmids: pKP14ND1-1(CP098376), pKP14ND2-1(CP098377), pKP15ND22-1(CP098379), with no reported 100% coverage plasmid. Consequently, we hypothesized the precursor of pL3835-NDM-OXA, proposed two models to elucidate the novel mechanism behind IS26-mediated tandem duplication of blaNDM−4: (1) Homologous recombination model: the plasmid initially harbors a single ‘red box’ region. Introduction of the second ‘red box’ region, either through replication errors or acquisition of external DNA, brings about a stretch of sufficient homology between the two Tn3-hypothetical protein-IS26 regions. This homologous segment triggers a homologous recombination event, which, under imprecise action of recombinases, integrating in reverse orientation to form the inverted tandem repeat [24]; (2) Post-acquisition blaNDM−4 insertion model: we posit a plasmid initially lacking the blaNDM−4 gene, with both ‘red box’ region acquired externally. The transposition of one IS26 moiety involves recognition and binding to another IS26 located on a separate exogenous DNA sequences. IS26 facilitates the formation of a DNA ring, which then inserts back into the genome, resulting in the generation of an inverted tandem duplication [22, 24] (Fig. 2C-2D).
Studies have indicated that the multiplication of ARGs can augment resistance levels [25]. However, there is currently a paucity of direct evidence specifically addressing how an increased copy number of blaNDM−4 impacts carbapenem resistance. Further research is warranted to elucidate this relationship comprehensively. The integration of novel MGEs in proximity to NDM-4 enriches its genetic context, emphasizing the critical function of MGEs in the lateral dissemination of ARGs. Successful conjugation assays have illustrated the plasmid-mediated transfer of blaNDM−4 to recipient bacteria. Our hypothesis posits that the emergence of novel MGEs in conjunction with NDM-4 expedite the dissemination of NDM-4 resistance, consequently exacerbating concerns over the escalating trend of antimicrobial resistance.
Plasmid analysis showed pL2388-NDM-OXA and similar plasmid types, except for pL3835-NDM-OXA, were all IncFII(K)-repB(R1701) (Table 3). Previous studies have reported that NDM-4 plasmids were mainly IncX3 [5, 9, 10] or IncF [1, 11, 12]. To our knowledge, this study is the first to report the NDM-4 plasmid type as IncFII(K)-repB(R1701)/repB(R1701). This suggests that these plasmids share a common replication initiation protein, repB(R1701), and fall within the IncF family, which is typically associated with efficient conjugative transfer capabilities and broad host range among diverse bacterial species. The plasmid harboring dual NDM-4 segments is labeled only as repB(R1701), without clear assignment to a specific Inc group, possibly indicating less clarity in its classification or highlighting the unique feature of its repB(R1701) protein. To our knowledge, this study represents the first report characterizing the plasmid types carrying blaNDM−4.
Further analysis of NDM-4-positive plasmids (Table 3) reveals a novel conjugative plasmid type, IncF and repB(R1701), mostly IncFII(K)-repB(R1701), which is geographically concentrated in Vietnam and China, predominantly in Vietnam and southern China. Isolates originate from various human clinical samples including stool, sputum, blood et al., collected between 2014 and 2020, but primarily deposited in the NCBI database in 2022, hence the recent availability of data for analysis. The outcomes of this study suggest that both L2388 and L3835 isolates, sourced from patients without foreign residency history and originating from separate hospitals, indicate a broader distribution of these strains than previously anticipated. This delay potentially signals a belated public health alert regarding the severity of NDM-4 resistance and its potentially underestimated transmission dynamics. It is imperative for public health authorities to promptly review and analyze these data to implement measures aimed at curtailing the further spread of resistance.