Hostname: page-component-7479d7b7d-t6hkb Total loading time: 0 Render date: 2024-07-16T01:32:38.980Z Has data issue: false hasContentIssue false

Genomic investigation of multispecies and multivariant blaNDM outbreak reveals key role of horizontal plasmid transmission

Published online by Cambridge University Press:  12 February 2024

Nenad Macesic
Affiliation:
Department of Infectious Diseases, The Alfred Hospital and School of Translational Medicine, Monash University, Melbourne, Australia Centre to Impact AMR, Monash University, Clayton, Australia
Adelaide Dennis
Affiliation:
Department of Infectious Diseases, The Alfred Hospital and School of Translational Medicine, Monash University, Melbourne, Australia
Jane Hawkey
Affiliation:
Department of Infectious Diseases, The Alfred Hospital and School of Translational Medicine, Monash University, Melbourne, Australia
Ben Vezina
Affiliation:
Department of Infectious Diseases, The Alfred Hospital and School of Translational Medicine, Monash University, Melbourne, Australia
Jessica A. Wisniewski
Affiliation:
Department of Infectious Diseases, The Alfred Hospital and School of Translational Medicine, Monash University, Melbourne, Australia
Hugh Cottingham
Affiliation:
Department of Infectious Diseases, The Alfred Hospital and School of Translational Medicine, Monash University, Melbourne, Australia
Luke V. Blakeway
Affiliation:
Department of Infectious Diseases, The Alfred Hospital and School of Translational Medicine, Monash University, Melbourne, Australia
Taylor Harshegyi
Affiliation:
Department of Infectious Diseases, The Alfred Hospital and School of Translational Medicine, Monash University, Melbourne, Australia
Katherine Pragastis
Affiliation:
Department of Infectious Diseases, The Alfred Hospital and School of Translational Medicine, Monash University, Melbourne, Australia
Gnei Zweena Badoordeen
Affiliation:
Department of Infectious Diseases, The Alfred Hospital and School of Translational Medicine, Monash University, Melbourne, Australia
Pauline Bass
Affiliation:
Department of Infectious Diseases, The Alfred Hospital and School of Translational Medicine, Monash University, Melbourne, Australia
Andrew J. Stewardson
Affiliation:
Department of Infectious Diseases, The Alfred Hospital and School of Translational Medicine, Monash University, Melbourne, Australia
Amanda Dennison
Affiliation:
Microbiology Unit, Alfred Hospital, Melbourne, Australia
Denis W. Spelman
Affiliation:
Department of Infectious Diseases, The Alfred Hospital and School of Translational Medicine, Monash University, Melbourne, Australia Microbiology Unit, Alfred Hospital, Melbourne, Australia
Adam W.J. Jenney
Affiliation:
Department of Infectious Diseases, The Alfred Hospital and School of Translational Medicine, Monash University, Melbourne, Australia Microbiology Unit, Alfred Hospital, Melbourne, Australia
Anton Y. Peleg*
Affiliation:
Department of Infectious Diseases, The Alfred Hospital and School of Translational Medicine, Monash University, Melbourne, Australia Centre to Impact AMR, Monash University, Clayton, Australia Infection Program, Monash Biomedicine Discovery Institute, Department of Microbiology, Monash University, Clayton, Australia
*
Corresponding author: Anton Y. Peleg; Email: anton.peleg@monash.edu
Rights & Permissions [Opens in a new window]

Abstract

Objectives:

New Delhi metallo-β-lactamases (NDMs) are major contributors to the spread of carbapenem resistance globally. In Australia, NDMs were previously associated with international travel, but from 2019 we noted increasing incidence of NDM-positive clinical isolates. We investigated the clinical and genomic epidemiology of NDM carriage at a tertiary-care Australian hospital from 2016 to 2021.

Methods:

We identified 49 patients with 84 NDM-carrying isolates in an institutional database, and we collected clinical data from electronic medical record. Short- and long-read whole genome sequencing was performed on all isolates. Completed genome assemblies were used to assess the genetic setting of blaNDM genes and to compare NDM plasmids.

Results:

Of 49 patients, 38 (78%) were identified in 2019–2021 and only 11 (29%) of 38 reported prior travel, compared with 9 (82%) of 11 in 2016–2018 (P = .037). In patients with NDM infection, the crude 7-day mortality rate was 0% and the 30-day mortality rate was 14% (2 of 14 patients). NDMs were noted in 41 bacterial strains (ie, species and sequence type combinations). Across 13 plasmid groups, 4 NDM variants were detected: blaNDM-1, blaNDM-4, blaNDM-5, and blaNDM-7. We noted a change from a diverse NDM plasmid repertoire in 2016–2018 to the emergence of conserved blaNDM-1 IncN and blaNDM-7 IncX3 epidemic plasmids, with interstrain spread in 2019–2021. These plasmids were noted in 19 (50%) of 38 patients and 35 (51%) of 68 genomes in 2019–2021.

Conclusions:

Increased NDM case numbers were due to local circulation of 2 epidemic plasmids with extensive interstrain transfer. Our findings underscore the challenges of outbreak detection when horizontal transmission of plasmids is the primary mode of spread.

Type
Original Article
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
© The Author(s), 2024. Published by Cambridge University Press on behalf of The Society for Healthcare Epidemiology of America

Carbapenem-resistant Gram-negative bacteria cause significant morbidity and mortality globally and were declared a critical priority by the World Health Organization. Reference Tacconelli, Carrara and Savoldi1,2 New Delhi metallo-β-lactamases (NDMs) are significant contributors to the spread of carbapenem resistance. Reference Wu, Feng, Tang, Qiao, McNally and Zong3 Although endemic in South and Southeast Asia, NDMs are increasingly noted in other settings, including North America and Europe, where they have caused substantial clonal outbreaks. Reference Lapp, Crawford and Miles-Jay4Reference Nordmann and Poirel7 The spread of NDMs is particularly concerning due to the paucity of treatment options for metallo-β-lactamase infections. Reference Paul, Carrara and Retamar8,Reference Tamma, Aitken, Bonomo, Mathers, van Duin and Clancy9 Clinical data regarding new treatments, such as cefiderocol and aztreonam-avibactam combinations, are limited, whereas older agents, such as polymyxins, have significant toxicities. Reference Paul, Carrara and Retamar8,Reference Tamma, Aitken, Bonomo, Mathers, van Duin and Clancy9 Stopping NDM spread is an urgent priority.

In Australia, NDMs have traditionally been associated with travel from endemic countries. Reference Lane, Brett and Schultz10 From 2019 onward, we have noted an increasing number of NDM infections and colonization episodes at our institution, which persisted despite coronavirus disease 2019 (COVID-19) travel restrictions. These episodes involved multiple bacterial host species, leading us to hypothesize that local NDM spread may be occurring through horizontal transfer of NDM plasmids. We therefore sought to determine the clinical and genomic epidemiology of NDM carriage at our institution from 2016 to 2021. More specifically, we sought to apply long-read sequencing approaches to characterize NDM genetic settings accurately and to understand within-patient diversity and transfer dynamics of NDM plasmids.

Methods

Study setting and population

The study was approved by the Alfred Hospital Ethics Committee. We reviewed an institutional database of carbapenem-resistant Gram-negative isolates spanning 2002–2021 at a healthcare system comprising a tertiary-care hospital, a community hospital, and a geriatric and rehabilitation hospital. We identified NDM carriage in 49 patients (84 isolates) from September 2016 to June 2021, with 81 (96%) of 84 isolates from the tertiary-care hospital and 3 (3.6%) of 84 isolates from the geriatric and rehabilitation hospital. Clinical data were extracted from the electronic medical record (including patient movement data) and a Charlson comorbidity index score calculated. Reference Charlson, Pompei, Ales and MacKenzie11 NDM isolates were classified as associated with colonization or infection. Reference Horan, Andrus and Dudeck12 Crude mortality rates at 7 and 30 days were recorded. Given the study included patients from September 2016 to June 2021, we divided the study into 2 equal periods from first to final NDM patient (2016–2018 and 2019–2021) to allow comparison over time. Patients were screened for carbapenemase-producing organisms following discharge from the intensive care unit (ICU) if they had a history of international hospitalization or were contacts of patients with carbapenemase-producing organisms, according to state guidelines. 13 Due to the absence of systematic NDM surveillance, we considered that patients may have been colonized in the 30 days prior to first isolation of NDM-carrying organism. Reference Ruppé, Armand-Lefèvre and Estellat14,Reference Macesic, Gomez-Simmonds and Sullivan15 We identified overlaps on the same ward at the same time as potential between-patient transmission events.

Isolate selection and genomic analyses

Routine antimicrobial susceptibility testing was performed using Vitek2 (BioMérieux) to presumptively identify NDM-carrying isolates. Isolates were screened for NDM presence using the Xpert Carba-R multiplex PCR assay (Cepheid) if meropenem MIC was >0.25 mg/L. All 84 NDM-carrying isolates identified were included in the study and underwent short-read (Illumina) and long-read (Oxford Nanopore Technologies) whole-genome sequencing (WGS). One genome was excluded from further analysis due to species mismatch.

Genomic analyses included resistance gene and plasmid replicon detection with Abricate version 1.0.0 software and in silico multilocus sequence typing (MLST) using ‘mlst’ version 2.19.0 software (Supplementary Methods online). 16,17 We performed core genome-based phylogenetic analyses on key STs (≥2 genomes available from ≥2 patients) to create a core genome alignment and to calculate pairwise single-nucleotide variant (SNV) distances. We identified NDM-carrying contigs that were putative plasmids in long-read assemblies using Abricate. MOB-typer version 1.4.9 software was used to determine plasmid replicons present and to identify plasmid groups according to its clustering algorithm. Reference Robertson, Bessonov, Schonfeld and Nash18 We compared individual plasmids using progressiveMauve version 2.4.0.r4736 software, Reference Darling, Mau and Perna19 and we assessed NDM flanking regions using Flanker version 0.1.5 software. Reference Matlock, Lipworth and Constantinides20

Statistical analysis

Categorical variables were compared using the χ2 test or the Fisher exact test, and continuous variables were compared using the Student t test or the Mann–Whitney–Wilcoxon test, as appropriate using R version 4.1.1 software.

Results

Study population and clinical characteristics

Of 49 patients with NDM carriage at our institution, 38 (78%) were identified from 2019 to 2021. Clinical characteristics of patients are shown in Table 1. A history of international travel in the prior year was noted in 20 (41%) of 49 patients. Significantly fewer patients reported travel in 2019–2021 than in 2016–2018: 11 (29%) of 38 versus 9 (81%) of 11; P = .037). NDM was detected a median of 5 days (IQR, 0–17) after admission. Most patients (36 of 49, 73%) were admitted from the community. Of these 49 patients, 33 (67%) had a history of hospitalization in the 3 months prior, including admission to ICU in 15 (45%) of those 33 patients. The first NDM isolate detected was associated with infection in 7 (14%) of 49 patients. NDM infections developed in 7 (17%) of 42 patients with initial NDM colonization (Table 2). The crude 7- and 30-day mortality rates in patients with NDM infections were 0% and 14% (2 of 14 patients), respectively.

Table 1. Clinical Characteristics of Study Cohort

Note. IQR, interquartile range; NDM, New Delhi metallo-β-lactamase.

a Units unless otherwise specified.

Table 2. Details of New Delhi Metallo-β-Lactamase (NDM) Infections, 2016–2021

Isolate characteristics and genomic epidemiology

NDM-carrying bacteria were highly diverse, with 41 bacterial strains (defined as unique species/MLST combinations) noted (Fig. 1, Supplementary Table 1 online, and Supplementary Fig. 1 online). No dominant bacterial strain was identified; a maximum of 7 genomes was noted in Escherichia coli ST405 (5 patients), Klebsiella pneumoniae ST16 (4 patients), and Klebsiella quasipneumoniae ST5551 (3 patients). Of 12 strains found in ≥2 patients, 6 strains had between-patient pairwise SNV distances <20 SNVs, suggestive of possible clonal spread (Supplementary Table 2 online). Also, 4 NDM variants were detected: bla NDM-1 (48 of 83 genomes, 58%), bla NDM-5 (20 of 83, 24%), bla NDM-7 (12 of 83, 14%), and bla NDM-4 (3 of 83, 4%). Non-NDM carbapenemase genes were detected in 11 (13%) of 83 genomes (bla IMP-4, 4 genomes; bla IMP-62, 3 genomes; bla OXA-23-like, 3 genomes; and bla OXA-48-like, 2 genomes). Also, 10 (12%) of 83 genomes carried mcr-9.1, a novel determinant of colistin resistance. We noted changes in epidemiology from E. coli in 2016–2018 (13 of 15 genomes, 86%) to non–E. coli species in 2019–2021 (54 of 68 genomes, 79%; P < .001), and from bla NDM-5 (11 of 15 genomes, 73%) to non-bla NDM-5 variants (59 of 68 genomes, 87%; P < .001).

Figure 1. New Delhi metallo-β-lactamase variants and bacterial host strains, 2016–2021. NDM variants and bacterial host strains shown over 2 study periods. There was a diversity in both the bacterial hosts and NDM variants, with 4 NDM variants being noted across 41 bacterial host strains, and a shift from NDM-5 and Escherichia coli in 2016–2018 to NDM-1/NDM-7 and non–E. coli species in 2019–2021. Note. NDM, New Delhi metallo-β-lactamase; ST, sequence type.

Plasmid analyses

To assess the hypothesis of plasmid transmission and to identify temporal shifts, we performed plasmid clustering to identify plasmid groups; we quantified SNVs across the plasmid backbone; and we assessed large-scale rearrangements. We detected 13 distinct plasmid groups, of which 3 of 13 had multiple NDM variants present (Fig. 2A), and 8 of 13 were noted in multiple bacterial host strains (Fig. 2B). In 2016–2018, there was a diversity of plasmid groups (8 groups across 11 patients). In 2019–2021, 3 dominant epidemic plasmid groups (bla NDM-1 IncN, bla NDM-5/NDM-7 IncX3 and bla NDM-1 IncC) were noted in 30 (79%) of 38 patients and 49 (71%) of 68 genomes in that period (Fig. 2A).

Figure 2. Genetic settings of bla NDM, 2016–2021. (A) Genetic settings of bla NDM and corresponding NDM variants over study, as defined by 2 study periods. We detected NDM variants in 13 distinct plasmid groups as well as integration into the bacterial chromosome. (B) Genetic settings of bla NDM and corresponding bacterial host strains over the study, as defined by 2 study periods. Note. NDM, New Delhi metallo-β-lactamase; ST, sequence type.

Overall, bla NDM-1 IncN plasmids were the largest group (25 genomes in 14 patients) and were highly homogenous (1 SNV across the backbone and minor structural variation near bla NDM-1) (Fig. 3 and Supplementary Table 3 online). IncX3 plasmids were similarly homogenous with near identical structures regardless of NDM variant (Fig. 3). Epidemic bla NDM-7 IncX3 plasmids carried only 2 SNVs across the backbone. Despite being clustered to the same plasmid group, IncC plasmids were more diverse with 23 SNVs across the backbone with an obvious subcluster of 6 near-identical plasmids from 4 patients (median pairwise SNV distance, 1). This subcluster corresponded to a plasmid initially found in E. coli ST176 in 2019, then in clonal Enterobacter hormaechei ST1015 in 3 other patients in 2020–2021. We compared these epidemic plasmids to plasmids circulating globally (Supplementary Fig. 2 online and Supplementary Table 4 online). IncX3 plasmids were almost identical to global NDM IncX3 plasmids, with 100% coverage and 99.96% identity. IncN and IncC plasmids were divergent to NDM global plasmids. IncN had 95% coverage and 99.93% identity, and IncC had 93% coverage and 99.99% identity.

Figure 3. Comparative analyses of NDM IncX3, IncN and IncC plasmids. We aligned representative plasmids from each of the 3 epidemic NDM plasmid groups from our study. Each colored field represents a locally collinear block, a homologous region of sequence shared by multiple plasmids without any rearrangement of that region. The same colors indicate the same regions present in different plasmids. IncX3 plasmids were structurally nearly identical regardless of NDM variant. bla NDM-1 IncN plasmids had minor structural variation near bla NDM-1 due to a palindromic sequence. Although they belonged to the same plasmid group, IncC plasmids were more diverse with three plasmid subtypes, as shown. Note. NDM, New Delhi metallo-β-lactamase; ST, sequence type.

Other plasmid groups with multiple plasmids available for comparison showed larger-scale structural differences despite sharing a similar backbone (Supplementary Fig. 3 online). Each NDM flanking region was associated with a single plasmid group, except for a transposon capable of carrying 2 NDM variants (bla NDM-1 and bla NDM-5) that was identified across 7 plasmid groups (Fig. 4). All patients but 1 with this flanking region had a history of travel to South or Southeast Asia. In 1 patient, we noted both the bla NDM-5 IncX3 and IncFIA-type (AA170 AH820) plasmids were carrying this flanking region (Fig. 4, plasmids 3 and 6, and Supplementary Fig. 4 online, patient 4). This transposon was not found in any other IncX3 plasmids in the study, which raises the possibility of movement of a transposable element between different plasmids within the patient.

Figure 4. Promiscuous NDM transposon found in 7 plasmid groups. The top of the figure shows the genes contained within the transposon, including bla NDM. The bottom of the figure shows the transposon (in color) inserting into diverse plasmids from 7 different plasmid groups. This finding was suggestive of movement of the transposon between different plasmid backbones. Note. NDM, New Delhi metallo-β-lactamase.

Within-patient plasmid analysis

Multiple genomes were available for analysis in 18 (37%) of 49 patients. Moreover, 6 patients had persistent strain colonization (multiple isolates of the same bacterial strain with same NDM plasmid) and 4 patients had multiple colonization events (presence of different plasmid groups), including 2 patients with plasmids carrying different NDM variants (Supplementary Fig. 4 online). In the remaining 8 of 18 patients, we detected the same plasmid group across different bacterial hosts. We compared plasmids within patients to distinguish potential interstrain plasmid transfer from multiple colonization events. The 7 of 8 patients colonized with bla NDM-1 IncN and bla NDM-7 IncX3, epidemic plasmids had highly similar plasmids suggestive of interspecies transfer (Fig. 5A and 5B). For the patient colonized with nonepidemic plasmids (Fig. 5C), there were significant structural rearrangements of an IncFII-type plasmid, making it less likely that interstrain plasmid transfer had occurred. In patients with multiple plasmids of the same plasmid group there was a median SNV distance of 0 (range, 0–267), which is shown by plasmid group in Supplementary Table 5 (online).

Figure 5. Analysis of potential within-patient plasmid transfer between bacterial host strains. Overall, 8 patients had the same NDM variant/plasmid group combinations across multiple bacterial host strains. We aligned plasmids from each of the bacterial host strains. (A) NDM-1 IncN plasmids. (B) NDM-7 IncX3 plasmids. (C) NDM-4 IncFII-type (AA450 AI539) plasmids. The bacterial host strains are shown on the left of the plasmids. Each colored field represents a locally collinear block, a homologous region of sequence shared by multiple plasmids without any rearrangement of that region. The same colors indicate the same regions present in different plasmids. Note. NDM, New Delhi metallo-β-lactamase; ST, sequence type.

Transmission analysis

We analyzed potential in-hospital transmission events by combining detailed patient movement data with plasmid-level genomic data. We defined a potential transmission event when there was spatiotemporal overlap between patients that shared closely related NDM plasmids (ie, the same plasmid group and NDM variant with ≤5 SNVs difference in the backbone). Despite multiple patients having highly similar plasmids, only 8 patients were linked in 5 potential transmission events, all involving the IncN plasmid (Supplementary Fig. 5 online). Despite using criteria aimed at increasing sensitivity, we were unable to directly link the cases of 41 (84%) of 49 patients.

Discussion

In this study, we combined clinical data with insights from long-read genomic sequencing to analyze the epidemiology of NDM-carrying bacteria at our institution. Although reports of previous NDM outbreaks noted clonal spread of specific strains such as K. pneumoniae ST147 and ST307, Reference Martin, Corey and Sannio5,Reference Biedrzycka, Izdebski and Urbanowicz6,Reference Haller, Kramer and Becker21 increased case numbers in our institution were due to circulation of several epidemic plasmids. For most patients, we were unable to document clear healthcare-associated NDM acquisition, highlighting the possibility of acquisition in the community or through undetected sources in the hospital. Ultimately, 29% patients developed NDM infection and the 30-day crude mortality rate among these patients was 14%, consistent with previously reported mortality rates of ∼10%–20%. Reference Seo, Kim and Kim22,Reference Zhao, Kennedy and Perry23 Our findings also underscore the challenges of outbreak detection when horizontal transmission of plasmids is the primary mode of spread, particularly due to the remarkable plasticity of plasmids, which can undergo substantial changes even within a single patient. Reference Gomez-Simmonds, Annavajhala and Tang24

We noted significant shifts in NDM epidemiology during the study. From 2016 to 2018, most colonized patients had a history of travel to endemic regions such as South and Southeast Asia, as noted previously in our setting. Reference Lane, Brett and Schultz10 Correspondingly, both the bacterial hosts and plasmids carrying NDMs were diverse with a predominance of E. coli and bla NDM-5, and no transmission events between patients were detected. This finding contrasted with 2019–2021, when we noted increased NDM case numbers driven by patients with no travel history, which persisted despite strict travel restrictions imposed in Australia due to the COVID-19 pandemic.

Rather than being due to a single bacterial clone or plasmid, this spread was due to IncN and IncX3 epidemic plasmids. These plasmids were capable of extensive interstrain transfer, found in 14 and 11 bacterial strains, respectively. IncX3 plasmids have been recognized as drivers of NDM spread globally Reference Wu, Feng, Tang, Qiao, McNally and Zong3,Reference Campos, Silva and Santos25,Reference Paskova, Medvecky and Skalova26 and plasmids from our study closely matched internationally circulating plasmids. NDM IncN plasmids have previously been less recognized. Reference Hirabayashi, Yahara and Mitsuhashi27Reference Acman, Wang and van Dorp30 In our study, both plasmids were highly conserved across different bacterial hosts and different patients, suggesting circulation of these plasmids and acquisition by multiple patients. We also noted multiple NDM IncC plasmids; however, these were more heterogenous, comprising a subcluster of 6 highly related plasmids across 4 patients that suggested horizontal transfer of the IncC plasmid then clonal spread of E. hormaechei ST1015.

Given these genomic findings, we attempted to detect healthcare-associated plasmid transmission. Despite applying liberal thresholds to increase sensitivity, we were only able to detect transmission events with direct epidemiological and genomic links in 8 (16%) of 49 patients. The absence of spatiotemporal overlap but high genomic relatedness of plasmids may be explained by undetected acquisition in the hospital such as from asymptomatic NDM colonized patients. We hypothesize that this is the most likely explanation for our findings because we did not conduct routine prospective surveillance. Prior work has shown that relying on clinical disease for detection substantially underestimates the number of patients colonized with carbapenemase-producing organisms. Reference Macesic, Gomez-Simmonds and Sullivan15,Reference Macesic, Gomez-Simmonds and Sullivan31 In addition, environmental sources (eg, hospital sinks and drains) are increasingly recognized as reservoirs of carbapenemase-producing organisms. Reference Kizny Gordon, Mathers and Cheong32 We assessed for sink colonization in the ICU in 2019 and did not detect NDM-carrying organisms, Reference Macesic, Hawkey and Vezina33,Reference Macesic, Hawkey and Vezina34 but we have not repeated systematic testing subsequently. Acquisition outside our institution (eg, in the community or in other hospital networks) is also possible. Increased local NDM acquisition in the Australian state of Victoria has been highlighted as a growing problem in recent state guidelines. 13 Indeed, in the 3 months prior, 67% of patients had a history of admission to hospital and 31% had a history of admission to the ICU. Our findings raise concern about a potential widespread NDM outbreak with subsequent establishment of endemicity, as noted due to healthcare-associated transmission across multiple institutions in Tuscany in 2018–2019. Reference Martin, Corey and Sannio5,Reference Coppi, Antonelli and Niccolai35Reference Tavoschi, Forni and Porretta37

Through our comprehensive long- and short-read sequencing of all study isolates, we were able to unravel several aspects of NDM plasmid spread. First, interstrain transfer of NDM plasmids occurred commonly and was a key driver for the increased NDM case numbers in 2019–2021. In addition, we demonstrated that interstrain transfer likely occurred within patients, highlighting the potential for diversification of genetic settings of epidemic plasmids within patients’ microbiomes. Reference Macesic, Hawkey and Vezina33,Reference Leon-Sampedro, DelaFuente and Diaz-Agero38 Second, although sophisticated plasmid clustering techniques provided much higher resolution than approaches such as rep typing, NDM plasmid plasticity rendered even these clusters imperfect. Reference Gomez-Simmonds, Annavajhala and Tang24 We noted closely related plasmid backbones that carried different NDM variants (eg, IncX3 plasmids) or NDM flanking regions. Finally, in addition to conducting plasmid-level analyses, we showed that NDM spread may result from mobile genetic elements smaller than plasmids, such as the NDM transposon we identified in multiple plasmid backbones, including within the same patient. This transposon was also noted as one of the most common NDM flanking regions in a recent comprehensive analysis of global NDM plasmids. Reference Acman, Wang and van Dorp30

These findings have important implications for use of WGS for outbreak detection where horizontal spread is the major contributor. Although determining whether plasmids are likely identical is feasible (eg, our analyses of IncN and IncX3 epidemic plasmids), defining differences between plasmids is a more complex endeavor. An objective threshold of plasmid relatedness suggestive of transmission is needed and will likely require use of multiple metrics including measures of identity (eg, SNV analyses, average nucleotide identity), gene content, and large-scale structural rearrangements and may vary between different plasmids. Reference Hawkey, Wyres and Judd39 Circular, closed plasmid assemblies enabled by long-read sequencing will be central to these efforts. Reference Harris and Alexander40

This study had several limitations. First, the research was observational and was based at a single center. Although we noted a diverse range of NDM plasmids and variants, these findings may not be generalizable to all NDMs. Second, we only conducted active surveillance in selected patients, likely limiting our detection of NDM colonization. However, we did have systematic collection of all clinical NDM isolates with resulting accurate detection of NDM infection episodes. Finally, we used assembly approaches that result in the highest-quality plasmid assemblies, but plasmid assembly remains challenging and necessitates manual review and curation.

In summary, our study determined the changing dynamics of NDM spread in our institution as the epidemiology shifted from an association with travel to likely local acquisition. Using long-read sequencing, we were able to pinpoint this to the arrival of successful epidemic NDM IncN and IncX3 plasmids that circulated concurrently and were noted in many bacterial hosts, suggesting interstrain transfer. These epidemic plasmids remained remarkably stable across multiple bacterial strains within and between patients. We also noted that plasmids were capable of substantial plasticity, complicating efforts at determining plasmid transmission. These findings have important implications for the future use of long-read sequencing in detection and control of outbreaks where horizontal transmission plays a significant role.

Supplementary material

To view supplementary material for this article, please visit https://doi.org/10.1017/ice.2024.8

Acknowledgments

Financial support

This work was supported by the National Health and Medical Research Council of Australia (Emerging Leader 1 Fellowship APP1176324 to N.M. and Practitioner Fellowship APP1117940 to A.Y.P). The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Competing interests

N.M. has received research support from GlaxoSmithKline, unrelated to the current study. A.Y.P. and A.J.S. have received research funding from MSD through an investigator-initiated research project. All other authors declare no conflict of interest.

Footnotes

PREVIOUS PRESENTATION. Part of these data were presented at the 33rd European Congress of Clinical Microbiology & Infectious Diseases on April 15–18, 2023, in Copenhagen, Denmark.

References

Tacconelli, E, Carrara, E, Savoldi, A, et al. Discovery, research, and development of new antibiotics: the WHO priority list of antibiotic-resistant bacteria and tuberculosis. Lancet Infect Dis 2018;18:318327.CrossRefGoogle ScholarPubMed
Antimicrobial Resistance C. Global burden of bacterial antimicrobial resistance in 2019: a systematic analysis. Lancet 2022;399:629655.CrossRefGoogle Scholar
Wu, W, Feng, Y, Tang, G, Qiao, F, McNally, A, Zong, Z. NDM Metallo-beta-lactamases and their bacterial producers in healthcare settings. Clin Microbiol Rev 2019;32:e0011518.CrossRefGoogle Scholar
Lapp, Z, Crawford, R, Miles-Jay, A, et al. Regional spread of blaNDM-1–containing Klebsiella pneumoniae ST147 in postacute care facilities. Clin Infect Dis 2021;73:14311439.CrossRefGoogle Scholar
Martin, MJ, Corey, BW, Sannio, F, et al. Anatomy of an extensively drug-resistant Klebsiella pneumoniae outbreak in Tuscany, Italy. Proc Natl Acad Sci U S A 2021;118:e2110227118.CrossRefGoogle ScholarPubMed
Biedrzycka, M, Izdebski, R, Urbanowicz, P, et al. MDR carbapenemase-producing Klebsiella pneumoniae of the hypervirulence-associated ST23 clone in Poland, 2009–19. J Antimicrob Chemother 2022;77:33673375.CrossRefGoogle ScholarPubMed
Nordmann, P, Poirel, L. Epidemiology and diagnostics of carbapenem resistance in gram-negative bacteria. Clin Infect Dis 2019;69:S521S528.CrossRefGoogle ScholarPubMed
Paul, M, Carrara, E, Retamar, P, et al. European Society of Clinical Microbiology and Infectious Diseases (ESCMID) guidelines for the treatment of infections caused by multidrug-resistant gram-negative bacilli (endorsed by European society of intensive care medicine). Clin Microbiol Infect 2022;28:521547.CrossRefGoogle ScholarPubMed
Tamma, PD, Aitken, SL, Bonomo, RA, Mathers, AJ, van Duin, D, Clancy, CJ. Infectious Diseases Society of America 2023 guidance on the treatment of antimicrobial resistant gram-negative infections. Clin Infect Dis 2023:ciad428.CrossRefGoogle Scholar
Lane, CR, Brett, J, Schultz, M, et al. Search and contain: impact of an integrated genomic and epidemiological surveillance and response program for control of carbapenemase-producing enterobacterales. Clin Infect Dis 2021;73:e3912e3920.CrossRefGoogle ScholarPubMed
Charlson, ME, Pompei, P, Ales, KL, MacKenzie, CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373383.CrossRefGoogle ScholarPubMed
Horan, TC, Andrus, M, Dudeck, MA. CDC/NHSN surveillance definition of healthcare-associated infection and criteria for specific types of infections in the acute care setting. Am J Infect Control 2008;36:309332.CrossRefGoogle ScholarPubMed
Victorian guideline on carbapenemase-producing organisms for health services, version 1.1. 2023. Victorian Department of Health website. https://www.health.vic.gov.au/infectious-diseases/victorian-guideline-on-cpo-for-health-services. Accessed December 20, 2023.Google Scholar
Ruppé, E, Armand-Lefèvre, L, Estellat, C, et al. Acquisition of carbapenemase-producing Enterobacteriaceae by healthy travellers to India, France, February 2012 to March 2013. Eurosurveillance 2014;19:20768.CrossRefGoogle ScholarPubMed
Macesic, N, Gomez-Simmonds, A, Sullivan, SB, et al. Genomic surveillance reveals diversity of multidrug-resistant organism colonization and infection: a prospective cohort study in liver transplant recipients. Clin Infect Dis 2018;67:905912.CrossRefGoogle ScholarPubMed
Abricate: Mass screening of contigs for antimicrobial resistance or virulence genes [computer program]. 2020.Google Scholar
mlst: Scan contig files against traditional PubMLST typing schemes [computer program]. 2020.Google Scholar
Robertson, J, Bessonov, K, Schonfeld, J, Nash, JHE. Universal whole-sequence-based plasmid typing and its utility to prediction of host range and epidemiological surveillance. Microb Genom 2020;6.Google ScholarPubMed
Darling, AE, Mau, B, Perna, NT. progressiveMauve: multiple genome alignment with gene gain, loss and rearrangement. Plos One 2010;5:e11147.CrossRefGoogle ScholarPubMed
Matlock, W, Lipworth, S, Constantinides, B, et al. Flanker: a tool for comparative genomics of gene flanking regions. Microb Genom 2021;7.Google ScholarPubMed
Haller, S, Kramer, R, Becker, K, et al. Extensively drug-resistant Klebsiella pneumoniae ST307 outbreak, northeastern Germany, June to October 2019. Euro Surveill 2019;24.CrossRefGoogle ScholarPubMed
Seo, H, Kim, HJ, Kim, MJ, et al. Comparison of clinical outcomes of patients infected with KPC- and NDM-producing Enterobacterales: a retrospective cohort study. Clin Microbiol Infect 2021;27:1167.CrossRefGoogle ScholarPubMed
Zhao, S, Kennedy, S, Perry, MR, et al. Epidemiology of and risk factors for mortality due to carbapenemase-producing organisms (CPO) in healthcare facilities. J Hosp Infect 2021;110:184193.CrossRefGoogle ScholarPubMed
Gomez-Simmonds, A, Annavajhala, MK, Tang, N, et al. Population structure of blaKPC-harboring IncN plasmids at a New York City medical center and evidence for multi-species horizontal transmission. J Antimicrob Chemother 2022;77:18731882.CrossRefGoogle Scholar
Campos, JC, Silva, MJFd, Santos, PRNd, et al. Characterization of Tn 3000, a transposon responsible for blaNDM-1 dissemination among Enterobacteriaceae in Brazil, Nepal, Morocco, and India. Antimicrob Agents Chemother 2015;59:73877395.CrossRefGoogle Scholar
Paskova, V, Medvecky, M, Skalova, A, et al. Characterization of NDM-encoding plasmids from Enterobacteriaceae recovered from Czech hospitals. Front Microbiol 2018;9:1549.CrossRefGoogle ScholarPubMed
Hirabayashi, A, Yahara, K, Mitsuhashi, S, et al. Plasmid analysis of NDM metallo-beta-lactamase–producing Enterobacterales isolated in Vietnam. Plos One 2021;16:e0231119.CrossRefGoogle ScholarPubMed
Netikul, T, Sidjabat, HE, Paterson, DL, et al. Characterization of an IncN2-type blaNDM-(1)–carrying plasmid in Escherichia coli ST131 and Klebsiella pneumoniae ST11 and ST15 isolates in Thailand. J Antimicrob Chemother 2014;69:31613163.CrossRefGoogle ScholarPubMed
Wailan, AM, Sartor, AL, Zowawi, HM, Perry, JD, Paterson, DL, Sidjabat, HE. Genetic contexts of blaNDM-1 in patients carrying multiple NDM-producing strains. Antimicrob Agents Chemother 2015;59:74057410.CrossRefGoogle ScholarPubMed
Acman, M, Wang, R, van Dorp, L, et al. Role of mobile genetic elements in the global dissemination of the carbapenem resistance gene bla(NDM). Nat Commun 2022;13:1131.CrossRefGoogle ScholarPubMed
Macesic, N, Gomez-Simmonds, A, Sullivan, SB, et al. Genomic surveillance reveals diversity of multidrug-resistant organism colonization and infection: a prospective cohort study in liver transplant recipients. Clin Infect Dis 2018;67:905912.CrossRefGoogle ScholarPubMed
Kizny Gordon, AE, Mathers, AJ, Cheong, EYL, et al. The hospital water environment as a reservoir for carbapenem-resistant organisms causing hospital-acquired infections—a systematic review of the literature. Clinical Infectious Diseases 2017;64:14351444.CrossRefGoogle ScholarPubMed
Macesic, N, Hawkey, J, Vezina, B, et al. Genomic dissection of endemic carbapenem resistance reveals metallo-beta-lactamase dissemination through clonal, plasmid and integron transfer. Nat Commun 2023;14:4764.CrossRefGoogle ScholarPubMed
Macesic, N, Hawkey, J, Vezina, B, et al. Genomic dissection of endemic carbapenem resistance reveals metallo-β-lactamase dissemination through clonal, plasmid and integron transfer. Nat Commun 2023;14:4764.CrossRefGoogle ScholarPubMed
Coppi, M, Antonelli, A, Niccolai, C, et al. Nosocomial outbreak by NDM-1–producing Klebsiella pneumoniae highly resistant to cefiderocol, Florence, Italy, August 2021 to June 2022. Euro Surveill 2022;27:2200795.CrossRefGoogle ScholarPubMed
Falcone, M, Giordano, C, Barnini, S, et al. Extremely drug-resistant NDM-9–producing ST147 Klebsiella pneumoniae causing infections in Italy, May 2020. Euro Surveill 2020;25.CrossRefGoogle ScholarPubMed
Tavoschi, L, Forni, S, Porretta, A, et al. Prolonged outbreak of New Delhi metallo-beta-lactamase–producing carbapenem-resistant Enterobacterales (NDM-CRE), Tuscany, Italy, 2018 to 2019. Euro Surveill 2020;25:2001779.CrossRefGoogle Scholar
Leon-Sampedro, R, DelaFuente, J, Diaz-Agero, C, et al. Pervasive transmission of a carbapenem resistance plasmid in the gut microbiota of hospitalized patients. Nat Microbiol 2021;6:606616.CrossRefGoogle ScholarPubMed
Hawkey, J, Wyres, KL, Judd, LM, et al. ESBL plasmids in Klebsiella pneumoniae: diversity, transmission and contribution to infection burden in the hospital setting. Genome Med 2022;14:97.CrossRefGoogle ScholarPubMed
Harris, PNA, Alexander, MW. Beyond the core genome: tracking plasmids in outbreaks of multidrug-resistant bacteria. Clin Infect Dis 2021;72:421422.CrossRefGoogle ScholarPubMed
Figure 0

Table 1. Clinical Characteristics of Study Cohort

Figure 1

Table 2. Details of New Delhi Metallo-β-Lactamase (NDM) Infections, 2016–2021

Figure 2

Figure 1. New Delhi metallo-β-lactamase variants and bacterial host strains, 2016–2021. NDM variants and bacterial host strains shown over 2 study periods. There was a diversity in both the bacterial hosts and NDM variants, with 4 NDM variants being noted across 41 bacterial host strains, and a shift from NDM-5 and Escherichia coli in 2016–2018 to NDM-1/NDM-7 and non–E. coli species in 2019–2021. Note. NDM, New Delhi metallo-β-lactamase; ST, sequence type.

Figure 3

Figure 2. Genetic settings of blaNDM, 2016–2021. (A) Genetic settings of blaNDM and corresponding NDM variants over study, as defined by 2 study periods. We detected NDM variants in 13 distinct plasmid groups as well as integration into the bacterial chromosome. (B) Genetic settings of blaNDM and corresponding bacterial host strains over the study, as defined by 2 study periods. Note. NDM, New Delhi metallo-β-lactamase; ST, sequence type.

Figure 4

Figure 3. Comparative analyses of NDM IncX3, IncN and IncC plasmids. We aligned representative plasmids from each of the 3 epidemic NDM plasmid groups from our study. Each colored field represents a locally collinear block, a homologous region of sequence shared by multiple plasmids without any rearrangement of that region. The same colors indicate the same regions present in different plasmids. IncX3 plasmids were structurally nearly identical regardless of NDM variant. blaNDM-1 IncN plasmids had minor structural variation near blaNDM-1 due to a palindromic sequence. Although they belonged to the same plasmid group, IncC plasmids were more diverse with three plasmid subtypes, as shown. Note. NDM, New Delhi metallo-β-lactamase; ST, sequence type.

Figure 5

Figure 4. Promiscuous NDM transposon found in 7 plasmid groups. The top of the figure shows the genes contained within the transposon, including blaNDM. The bottom of the figure shows the transposon (in color) inserting into diverse plasmids from 7 different plasmid groups. This finding was suggestive of movement of the transposon between different plasmid backbones. Note. NDM, New Delhi metallo-β-lactamase.

Figure 6

Figure 5. Analysis of potential within-patient plasmid transfer between bacterial host strains. Overall, 8 patients had the same NDM variant/plasmid group combinations across multiple bacterial host strains. We aligned plasmids from each of the bacterial host strains. (A) NDM-1 IncN plasmids. (B) NDM-7 IncX3 plasmids. (C) NDM-4 IncFII-type (AA450 AI539) plasmids. The bacterial host strains are shown on the left of the plasmids. Each colored field represents a locally collinear block, a homologous region of sequence shared by multiple plasmids without any rearrangement of that region. The same colors indicate the same regions present in different plasmids. Note. NDM, New Delhi metallo-β-lactamase; ST, sequence type.

Supplementary material: File

Macesic et al. supplementary material

Macesic et al. supplementary material
Download Macesic et al. supplementary material(File)
File 1.3 MB