Developing a novel approach to complex evolution of antimicrobial resistance via mobile genetic elements from a persistent environmental biofilm

doi:10.21203/rs.3.rs-4697425/v1

Download PDF

Article

Developing a novel approach to complex evolution of antimicrobial resistance via mobile genetic elements from a persistent environmental biofilm

https://doi.org/10.21203/rs.3.rs-4697425/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Mobile genetic elements are key to the global emergence of antibiotic resistance. We successfully reconstructed the complete bacterial genome and plasmid assemblies of isolates sharing the same bla_KPC carbapenemase gene to understand evolution over time in six confined hospital drain biofilms over five years. From 82 isolates we identified 14 unique strains from 10 species with 113 bla_KPC−carrying plasmids across 16 distinct replicon types. To assess dynamic gene movement, we introduced the 'Composite-Sample Complex', a novel mathematical approach to using probability to capture the directional movement of antimicrobial resistance genes accounting for the co-occurrence of both plasmids and chromosomes within an isolate, and highlighting likely donors and recipients. From the validated model, we demonstrate frequent transposition events of bla_KPC from plasmids to other plasmids, as well as integration into the bacterial chromosome within specific drain biofilms. We present a novel approach to estimate the directional movement of antimicrobial resistance via gene mobilization.

Biological sciences/Computational biology and bioinformatics/Computational models

Biological sciences/Evolution/Molecular evolution

Biological sciences/Microbiology/Microbial genetics/Bacterial genetics

Biological sciences/Evolution/Evolutionary genetics

Antimicrobial resistance has become one of the greatest human health threats. Carbapenem-resistant Enterobacterales are key examples of this threat, having spread rapidly worldwide and across niches, making them difficult to track and monitor, and being associated with multidrug resistance and poor clinical outcomes with invasive infection^1,2. Carbapenem resistance in Enterobacterales is mainly conferred by hydrolyzing enzymes encoded by carbapenemase genes, which are often carried on mobile genetic elements (MGEs), allowing these to be rapidly shared between bacterial strains and species. This horizontal gene transfer, especially between environmental organisms and clinical pathogens, is driving much of the carbapenemase crisis, but genomic analysis tools have been limited in tracking and quantifying the dissemination dynamics of these important genes^3,4.

In recent years, hospital wastewater plumbing has been identified as a major reservoir for antibiotic-resistant pathogens to evolve, with sink drains frequently implicated in the spread of hospital-acquired infections, and disease outbreaks^1,5–7. Water system biofilms in sink drains and traps are a unique ecosystem for nosocomial pathogens, where survival and proliferation are often facilitated by the selective pressure exerted by disposal of antibiotics and use of disinfectants^8–10. The complex microbial diversity of these biofilms provides opportunities for AMR gene movement and sharing between bacteria via horizontal gene transfer of transposons and conjugative plasmids^11,12.

In 2007, with the introduction of Klebsiella pneumoniae carbapenemase producing organisms (KPCO) to our hospital we had previously described complex bla_KPC transmission dynamics within patients through transposon and plasmid sharing across 62 unique bacterial strains from the first 181 patients known to be affected¹³. As wastewater premise plumbing represented a major KPCO reservoir and a plausible contributor to transmission¹⁴, we implemented environmental surveillance in 2013 to detect KPCO in sink drains, sink p-trap water, as well as toilet/hopper lids and water, in intensive care units across our institution.

In this report, we present our findings from all longitudinal KPCO isolates from drain biofilms and p-trap water collected from six sinks in four different medical ICU rooms over five years. Using both long- and short-read whole genome sequencing to generate complete hybrid assemblies with circularized chromosome and plasmid sequences, we successfully reconstructed their full-length sequences. These circularized sequences were annotated using bioinformatics tools to investigate their genetic content and relatedness. The information was then integrated into a mathematical framework called the "Composite-Sample Complex", which captures the co-occurrence of chromosomal and plasmid contigs and enables us to deduce gene mobility and directionality patterns from a collection of isolates (Figure.1).

Drain biofilm and P-trap water were sampled contemporaneously across six sinks in four ICU rooms at least three times annually within the five-year study period (286 sampling instances) with a KPCO-positivity rate of 20–40% for all rooms across all samplings. The longitudinal sampling resulted in 87 KPCO isolates across six sinks (52 drain/35 P-trap) from four rooms (See Supplementary Table 1). Eighty-two of the KPCO isolates (94%) were successfully assembled and included in the analysis (Supplementary Fig. 2).

Distribution of species and strains

Ten unique bla_KPC-positive species were identified in the drains and P-traps. Serratia marcescens was the most frequently isolated species, accounting for 24/82 sequenced isolates (29%), and Citrobacter freundii and Raoultella ornithinolytica the next most common, with 19 isolates each (23%) (Fig. 2, Supplementary Figure). Across all species, we identified 14 distinct clades (based on a threshold to define a clade of > 400 SNVs) including four C. freundii clades and two S. marcescens clades (Fig. 2). Given how genetically distinct clades were from each other, bla_KPC presence across species/clades was most consistent with multiple horizontal transfer events^15,16.

We observed distinct species/clade compositions in the unique sinks across different units. Four of the ten species were exclusively confined to one of the sinks, namely Enterobacter hormaechei in Sink-1PS, Aeromonas hydrophila in Sink-1BS, and Enterobacter asburiae and K. pneumoniae in Sink-2BS (Fig. 2).

The remaining six species were shared across multiple locations, but certain clades were specific to particular sinks (e.g. S. marcescens clade 2 in Sink-4PS, C. freundii clade 4 in Sink-1PS, and C. freundii clade 3 in Sink-2BS). Additionally, species/clade composition within some sinks either changed over time as well as observing several clades persisting for the five-year duration in a single sink (Supplementary Fig. 2). These findings suggest that KPCO dynamics within sinks vary substantially from sink-to-sink with each sink fixture undergoing distinct evolutionary processes with some degree of independence.

Complex, nested variation in the genetic context of bla_KPC

To assess the dynamics of horizontal transfer of bla_KPC within environmental isolates, we examined variation in bla_KPC and its genetic environment. bla_KPC is typically carried in a highly conserved replicative 10 kb transposon, Tn4401¹⁷, which can also be mobilized as part of a complex nested, Russian doll-like system by plasmids and other mobile genetic elements such as insertion sequences and other transposons³. Within Tn4401, ISKpn7 is typically located upstream of bla_KPC with the same orientation, while ISKpn6 is positioned downstream of bla_KPC in the opposite orientation (Fig. 3). Both ISKpn6 and ISKpn7 can function as mobile genetic elements^19,22.

In this study, two bla_KPC alleles were identified in three known Tn4401 variants, with additional variation observed in the insertion sequence, downstream of bla_KPC (i.e. ISKpn6). These composite transposons were seen across 113 different plasmids, some of which were shared across species/clades (Fig. 3). Overall, with the many varied genetic contexts of bla_KPC within this data set demonstrates a complex history of gene mobilization across species and plasmids at several trackable genetic levels; allele, Tn4401 structural variant with and without Tn5403 insertion, replicon typed plasmid and species (Fig. 4).

The distribution of the two bla_KPC alleles identified (bla_KPC−2, bla_KPC−3; differ by one SNV) could not clearly be attributed to any single factor such as unit/sink location, species/clade, or mutation given the different wider genetic contexts (Supplementary Table 2). None of the 82 isolates carried both bla_KPC−2 and bla_KPC−3 simultaneously, as observed previously¹⁸. Different isolates of the same clade did however harbour different bla_KPC alleles. For example, one clade of K. michiganensis had bla_KPC−2 on the pUVA01 and IncFII(pBK30683) plasmids in one isolate sampled from Sink-3PS, and bla_KPC−3 on the pUVA01 and IncN plasmids in two other isolates from the same sink.

The robust association between the bla_KPC gene and ISKpn6/ISKpn7 within Tn4401 is well-documented in the literature^19–21. Among the 82 isolates, we identified two structural isoforms of ISKpn6: ISKpn6a and ISKpn6b. The genomic characterization of these two isoforms revealed more intricate rearrangement structures within our collection ISKpn6a closely aligns with the wild-type ISKpn6 sequence in the ISFinder database. In contrast, ISKpn6b exhibits a novel insertion of Tn5403 at nucleotide position 406 of ISKpn6 (Fig. 3). ISKpn6a was the most prevalent isoform in environmental samples, found in 62 isolates across all 10 bacterial species, 14 clades, and 12 of the plasmid Inc types. It was observed in all 6 units Dec 2013-Sep 2014. In contrast, ISKpn6b was confined mainly to specific species (R. ornithinolytica, K. oxytoca, K. michiganensis), and plasmid Inc types (IncFIA(HI1)__IncFII(K), IncN, IncM1, Col440II), predominantly observed in units Sink-4PS and Sink-3PS, and from Mar 2014 to Sep 2018. The average copy number of bla_KPC when associated with ISKpn6b was noted to be significantly higher than when associated with ISKpn6a (1.39, stdev [0.57] versus 1.82 [stdev: 0.59]; p = 0.003, Wilcoxon rank sum test). This observation is consistent with the hypothesis that Tn5403 enhances the transposition capability of bla_KPC, as has been demonstrated in vitro²³.

As bla_KPC is always contained within Tn4401 in our dataset, we identified structural variation, single-nucleotide variation (SNV), and flanking sequences of Tn4401 across the environmental samples to track and validate the mobilization events of bla_KPC. Among 82 isolates, there were three structural variants of Tn4401, namely, Tn4401b, Tn4401a, and Tn4401h in 79 (96.3 %), 1 (1.2%), 2 (2.4%) isolates, respectively. All structural variants carried bla_KPC, but they exhibited different deletions in the promoter region upstream of bla_KPC as previously described²⁴. Tn4401b was the most prevalent variant, being found in 8/10 species, widely spreading across all sampled sinks.

Using replicon typing as a crude metric of plasmid diversity, there were 113 bla_KPC-carrying plasmids across 15 distinct known replicon types [as well as the non-typeable but characterized plasmid pKPC_UVA01 (repA_CP013325)]^16,25. The top 3 Inc types associated with bla_KPC carriage were pKPC_UVA01 (in 33 samples), IncN (in 21 samples), and hybrid IncU__IncX5 (in 18 samples). These major KPC-carrying Inc types were shared across different units supporting the concept that bla_KPC was acquired by the plasmid outside of a specific study sink, with pKPC_UVA01, IncN, and IncU__IncX5 observed in 5, 4, and 4 sinks, respectively. The IncU/IncX KPC hybrid plasmid was only identified in one clade of S. marcescens (across multiple sinks), suggesting vertical transmission and persistence of a bla_KPC plasmid within this S. marcescens clade over time. In contrast, pKPC_UVA01 was shared across 5 species, and in 25/33 of isolates bla_KPC had more than one genetic context consistent with bla_KPC dissemination via horizontal gene transfer through transposition and plasmid transfer in this genetic context.

Given the diversity of genetic contexts for bla_KPC that were observed, we applied our framework to characterize the putative mobilization events that contributed to this complex diversity.

Characterizing bla_KPC transfers by Building a Composite-Sample Complex Model

To capture the complex dynamics of bla_KPC movement and evolution within a relatively stable biofilm we took a novel approach. We treated each isolate as a collection of co-occurring circularized genomic elements (either chromosomes or replicon type labeled plasmids), and as an evolutionary snapshot where the genomic element could be a source or a target of bla_KPC. We then developed a mathematical framework of plasmid chromosome co-occurrence called the "Composite-Sample Complex" (CSC) (See Method Section and Supplementary Fig. 4). This framework enables us to deduce gene mobility and directionality patterns from a collection of isolates. The primary principles are that only one plasmid replicon type or chromosome can exist in any instance and, bla_KPC can be integrated into or from any genomic context. Each CSC constitutes the union of all isolates in the collection which is being investigated (e.g. within a single clade, or a single sink), the vertices are the individually labeled plasmids/chromosome, vertex weights are determined by the number of isolates that contain that specific element in the collection, and the weight of the number of vertices with bla_KPC indicates origin versus source. By integrating replicon typing information and treating all closed genomic elements within a bacterial cell as a potential target for resistance integration, the Composite-Sample Complex ensures reliable identification of these mobility patterns. Moreover, this framework assists in accurately pinpointing the source and target of gene mobility.

Transposition: Plasmid to chromosome

We observed the integration of bla_KPC into the chromosome in Citrobacter freundii (in 4 samples), Raoultella ornithinolytica (in 9 samples), and Enterobacter asburiae (in 1 sample). Interestingly, while Citrobacter freundii and Raoultella ornithinolytica were shared across different units, the genomic KPC-carrying integration site in the chromosomes was specific to each sink, with Citrobacter freundii found in Sink-1PS and Raoultella ornithinolytica in Sink-4PS, suggesting that these integrations into chromosome took place in sink specific biofilm environment. This is an important biological validation as chromosomes, by definition, will not mobilize into another species or strain like a mobile plasmid. The CSC is then further supported in the finding that the genomic context of chromosomal integration is unique to the specific drain biofilm.

We can successfully hypothesize the source of bla_KPC transposition for each of these integrations, as depicted in Fig. 5. To elucidate our approach, we use the KPC integration into the R. ornithinolytica chromosome in unit Sink-4PS as an illustrative example of the model. Within unit Sink-4PS, 9 out of 12 R. ornithinolytica isolates harbor KPC in their chromosomes, all of which are associated with KPC-carrying IncN plasmids (Fig. 5a, Fig. 5b). Even among the remaining R. ornithinolytica isolates lacking KPC in their chromosomes, they still possess KPC-carrying IncN plasmids. Furthermore, no other KPC-carrying plasmid displays such a high degree of co-occurrence with KPC-carrying R. ornithinolytica chromosomes. Hence, we posit that the IncN type is the origin of KPC transposition into R. ornithinolytica chromosomes (Fig. 5c).

To validate the hypothesized transposition pattern, we examined the variations of bla_KPC and its surrounding genetic elements in both the source and target. We then checked for consistency between them (Table 2). Specifically, the co-occurring IncN plasmids and R. ornithinolytica chromosomes both possess bla_KPC−3, which is flanked by ISKpn6b and further surrounded by Tn4401b. This arrangement is in agreement with the notion that the IncN plasmid type serves as the origin of KPC transposition onto the chromosomes of R. ornithinolytica. Additionally, we examined the 5bp flanking sequence of Tn4401. The variations in the flanking sequence of Tn4401 indicate that the mobilization of the KPC occurred through the transposition of Tn4401, see Table 2 line 1.

Using a methodology analogous to our earlier approach, we analyzed the CSCs and KPC-induced CSCs constructed from isolates of the same bacterial clade from a particular sink across all possible strain-sink combinations. We successfully identified two other KPC plasmid-chromosome transposition pathways: from pKPC_UVA01 to C. freundii in unit Sink-1PS (Fig. 5d), and from pKPC_UVA01 to E. asburiae in unit Sink-2BS (Fig. 5e). Again, these transposition pathways are supported by subsequent validation (Table 2) from the genomic context. Notably, all identified KPC plasmid to chromosome transposition patterns are sink-specific, consistent with transpositions within the specific sink drain/ptrap biofilm, highlighting that this mobilization is potentially occurring while in the environmental biofilm.

Transposition: plasmid to plasmid

In addition to chromosomal transposition events, we identified plasmid-plasmid transposition pathways of bla_KPC. Three distinct pathways were observed in unit Sink-4PS: IncN to Col440II, IncN to IncFIA(HI1)__IncFII(K) in R. ornithinolytica (Fig. 5c), and IncC to Col(pHAD28) in S. marcescens (Fig. 5f). The co-occurring IncN and Col440II plasmids, as well as the IncN and IncFIA(HI1)/IncFII(K) plasmids, consistently carry bla_KPC−3, which is flanked by ISKpn6b in a Tn4401b isoform. Meanwhile, the co-occurring IncC and Col(pHAD28) plasmids all carry bla_KPC−3 with ISKpn6a wild type.

These findings provide insights into the distinct plasmid to plasmid transposition pathways, utilizing the genetic consistency of bla_KPC and associated mutations within each pathway to provide additional validation support to the model. The observed transposition events contribute to the understanding of bla_KPC dynamics within the specific environment of unit Sink-4PS, emphasizing the diversity of plasmid types involved in the dissemination of antimicrobial resistance genes. The results were derived from the evaluation of the CSC and then validated with complete closed genomic data and emphasized the ongoing real-world evolution of strains in the environment over time with a focus on a single gene as a proof of concept.

Table 2

KPC transposition patterns and the corresponding KPC genomic variations.
Species	Source Contig	Target Contig	Source bla_KPC Variant	Source ISKpn6 Variants	Source Tn4401 elements	Target bla_KPC Variant	Target ISKpn6 Variants	Target Tn4401 elements	Agreement
Raoultella ornithinolytica	IncN	chromosome	bla_KPC-3	ISKpn6b	GTTCT-Tn4401b-GTTCT	bla_KPC-3	ISKpn6b	AAACT-Tn4401b-AAACT	Yes
Citrobacter freundii	pKPC_UVA01	chromosome	bla_KPC-2	ISKpn6a	TBD-Tn4401b-TBD	bla_KPC-2	ISKpn6a	TBD-Tn4401b-TBD	Partially*
Enterobacter asburiae	pKPC_UVA01	chromosome	bla_KPC-2	ISKpn6a	GTTCT-Tn4401b-GTTCT	bla_KPC-2	ISKpn6a	ATAAA-Tn4401b-ATAAA	Yes
Raoultella ornithinolytica	IncN	Col440II	bla_KPC-3	ISKpn6b	GTTCT-Tn4401b-GTTCT	bla_KPC-3	ISKpn6b	AAATA-Tn4401b-AAATA	Yes
Raoultella ornithinolytica	IncN	IncFIA(HI1)__IncFII(K)	bla_KPC-3	ISKpn6b	GTTCT-Tn4401b-GTTCT	bla_KPC-3	ISKpn6b	TATTA-Tn4401b-TATTA, ATGTA-Tn4401b-ATGTA	Yes**
Serratia marcescens	IncC	Col(pHAD28)	bla_KPC-3	ISKpn6a	GTTCT-Tn4401b-GTTCT	bla_KPC-3	ISKpn6a	ACATT-Tn4401b-ACATT	Yes
* we could not identify the flanking sequence that is associated to the source as there are multiple choices
* *the result suggests two independent transpositions from IncN to IncFIA(HI1)__IncFII(K)

Conjugation

To understand potential transfer of genetic material between bacterial strains within sinks, we investigate the CSC and the KPC-induced CSC constructed from all the isolates collected from each individual sink (Supplementary Fig. 3). Through our construction, replicon types situated at the junction of the CSC may indicate the presence of conjugative plasmids, given these replicon types are shared among multiple species. We observed that pKPC_UVA01 appears at the junction of the composite sample complex in 4 out of 6 sinks. This observation aligns with prior findings, which suggest that pKPC_UVA01 is highly conjugative and plays a pivotal role in the outbreak of KPC dissemination within our hospital¹⁶. Notably, C. freundii is the only species common to these four rooms, and nearly all C. freundii isolates carry pKPC_UVA01. Consequently, we hypothesize that C. freundii serves as the source of pKPC_UVA01 conjugation.

Another plasmid, Col(pHAD28), could also be considered highly conjugative by this model but also consideration needs to be given for this plasmid being stably carried by many isolates before introduction of bla_KPC. It has been previously observed that a Col(pHAD28) type plasmid p3223 being shared across many species of Gammaproteobacteria in both UVA hospital isolate collection and NCBI's global collection, and could have synergistic relationships with large KPC-carrying plasmids²⁶. Our initial screening suggests that the majority of the Col(pHAD28) plasmids in our environmental isolates have identical or similar length with p3223, supporting the hypothesis that they are related via conjugation. However, we do observe length variation in some of the S. marcescens and K. michiganensis isolates, suggesting they are undergoing frequent rearrangements. Following a similar line of reasoning, we see that S. marcescens is the likely source of Col(pHAD28) conjugation.

Due to frequent rearrangements, plasmids diverge from the conventional tree-like evolution possible with single nucleotide variants accumulating at random in a genome which allow for Bayesian inferences of evolutionary pathways. Consequently, global alignment-based methods and traditional phylogenetic inference are unsuitable for studying plasmid evolution²⁷. Current methods that investigate transposition primarily focus on identifying similar subsequences that exhibit a certain degree of co-occurrence on different plasmid contigs. However, even with the identification of conserved patterns, it remains insufficient to reconstruct the transposition pathway. Here we use a relatively stable environmental niche over a five-year period to carefully observe some of the complex evolution occurring within bacterial strains all sharing a tractable genetic element (i.e. bla_KPC) to describe some of the complexity and as a first step to building a new approach evolution driven by mobile genetic elements. The method uses the co-occurrence of a collection of plasmids with a chromosome in an isolate as an evolutionary snap shot of potential targets and donors of bla_KPC. We then label the elements which have bla_KPC and use probability within which to make inferences about the directionality and movement of a specific gene within or across bacterial strains over time. This approach recognizes that a collection of plasmids and a specific bacterial chromosome is not randomly assembled but maintained as a collection of circularized genomic contents by vertical bacterial replication in which gene mobility can occur or be introduced. Although this model may be imperfect and require validation to be definitive, it could be scalable and used for hypothesis building around the origin and directionality of mobile genetic elements and their associated genes.

As described in the first portion of the manuscript, there is so much diversity in the genomic context of this single gene in a single built environment that manually lining up 82 isolates, and the hundreds of plasmids can feel like a challenging and confusing effort on which little biologic speculation about directionality can be established. Using this hypothesis-building tool, we have been able to organize the plasmids to estimate mobility and then confirm with additional data and metadata (sink location, species). In this effort, we have successfully discerned sink-specific KPC mobility patterns within and across clades, including integrating bla_KPC into chromosomes. Based on the data, this likely occurred within the biofilm environment in more than one case. Integration into the chromosome is a hypothesized method for maintaining important genes within a bacterial cell²⁸, and here we demonstrate this occurrence and the natural environment for the mobility to occur. Finding these real-world instances can provide additional clues to validate the location and setting where this antimicrobial resistance evolution is taking place.

Aside from successfully applying a novel approach to the antimicrobial resistance movement, there are several additional biological findings. Tn5403 was identified in K. pneumoniae as a 'helper element' enabling mobilization of non-conjugative plasmids via cointegrate formation^23,29 and demonstrating real-world validation of increased mobility associated with another transposon which had demonstrated this in a laboratory model²³. This further supports our approach to trying to build a model which can detect enhanced movement and evolutionary history as critical in understanding antimicrobial resistance gene dissemination.

As long-read sequencing becomes widely available and easily accessible in both research and clinical settings, there is an anticipation that the ability to assemble mobile genetic elements within their true genomic context will rapidly expand as the context of mobile elements is often not available with short-read data. However, we still need new approaches to assess and track the movement of those elements within and between bacterial cells to understand the evolutionary processes. Here, we have taken a spatiotemporally confined fully sequenced longitudinal collection of multi-species bacterial isolates that harbor the same carbapenemase gene from a persistently colonized environmental reservoir to look first at bla_KPC gene mobility within and across strains combining scientific and genetic principals to create an informed mathematical model for hypothesis generation and testing. Our CSC model employs simplicial complexes, which extend the concepts of networks and hypergraphs to model higher-order relationships in real-world data sets³⁰. This sophisticated mathematical framework, as the foundation of topological data analysis^31,32, has been successfully applied in fields such as network science³³, computational biology^34,35, medical imaging³⁶ and biomedicine³⁷, demonstrating its utility and effectiveness. Here, we tested and validated this novel approach which we believe could be deployed more widely to understand antimicrobial resistance evolution as it has first been vetted on this smaller well curated data set.

There are several limitations to the approach and will need to assess its future scalability to larger data sets. First, replicon typing serves as an informative means to classify plasmids, as two different plasmids containing the same replicon are incompatible in the same cell. However, relying solely on replicon typing to infer evolutionary closeness can be misleading, as two evolutionarily independent plasmids which have long been carried by unique strains may still share the same type. In the future we could integrate additional plasmid markers to better define distinct plasmids especially when applying the CSC across strains and species. The conjugative nature of plasmids further complicates tracking gene mobility on plasmids through replicon typing, as demonstrated that we were unable to completely work out the mechanisms of movement for Col(pHAD28) (i.e. frequent co-mobilization with other plasmids versus long term carriage by many strains versus source of bla_KPC transmission between cells).

In conclusion, we report a targeted genomic characterization of KPCOs isolated from our hospital’s wastewater plumbing environment over a five-year sampling frame to understand resistance gene sharing using a novel method. The evolutionary tools we currently have to describe mobile genetic elements among bacteria remain limited, we were able to demonstrate plausible KPC gene movement within and across clades with a data set paired with a novel mathematical approach. We believe this type of approach could be more widely used to understand the importance of bacterial gene sharing and evolution driven by mobile genetic elements in many future contexts.

Sampling Methodology and Approach

As a part of an on-going surveillance of premise plumbing sources at our institution, we collected samples from wastewater sites of sink plumbing environment (drain swabs and p-trap water) from six sinks in four medical intensive care rooms across two units between Dec-2013 and Dec-2018. The four rooms were selected based on their persistent colonization over time, room occupancy of clinically similar patients, Medical Intensive Care Unit (MICU) sink status and lack of physical intervention to exchange plumbing within the timeframe. Rooms with Sinks- 1PS, 1BS, 2PS, and 2BS were situated in MICU which opened in 2012 and rooms with Sinks- 3PS and 4PS were in another MICU which opened in 1981 (Extended Data Table 1 and Extended Data Fig. 1). All sinks had similar designs with oval uniform Corian® (DuPont) bowl with an overflow.

All collected samples were processed to capture KPC-positive isolates, as previously described¹⁴. All isolates shown to be phenotypically positive for carbapenemase production³⁸, were subjected to real-time PCR to detect bla_KPC.

Whole-genome sequencing (WGS)

All viable confirmed bla_KPC positive isolates underwent short-read sequencing on Illumina HiSeq2000 instrument as previously described³⁸. Long read sequencing was performed on the MinION platform using the Spot On MK1 R9 version flowcells (Oxford Nanopore, Oxford, UK). Library preparation was done using the native barcoding kit (EXP-NBD104) and ligation sequencing kit (SQK-LSK109), following the manufacturer’s protocol (Native barcoding genomic DNA protocol).

Bioinformatics Analysis

Illumina WGS data analysis. The quality controls steps for raw Illumina short-reads included adapter removal using Trim Galore (v0.6.4; parameters: --length 35 --stringency 5)³⁹, followed by trimming the low-quality tails using Trimmomatic (v0.39; parameters: TRAILING:20 SLIDINGWINDOW: 4:20 MINLEN:70)⁴⁰. Species identification was performed by calculating the genomic distances of the read sets against a collection of complete assemblies in NCBI’s RefSeq database using MASH⁴¹(v2.1). The closest matched reference genome for each species was used to perform alignment using bwa mem⁴² and variant calling using freebayes v1.2.0⁴³, as implemented in SNIPPY⁴⁴ (v4.6.0). Within each species, pairwise whole-genome single-nucleotide variant (wgSNV) distances were calculated, and maximum-likelihood phylogenies for each species were built on the SNV-site alignments using IQ-TREE (v2.1.3; substitution model: GTR + ASC)^45,46, followed by joint ancestral reconstruction with pyjar^47,48 (v1.0). Cluster of isolates within 25 wgSNVs of each other were considered to be clonal based on short read data. The multilocus sequence typing based on (usually) seven house-keeping genes was performed on de-novo assembled contigs generated using SPAdes^49–51 (v3.15.4).

De-novo hybrid assembly. Using the high-quality Illumina read sets as a reference, the raw Nanopore long-reads were quality filtered using Filtlong (v0.2.0) for higher-quality subset of long reads⁵². Fully resolved bacterial assemblies were reconstructed from the high-quality long and short read data using the hybrid-assembly approach implemented in Unicycler (V0.4.9)⁵³. The assemblies were annotated using Prokka⁵⁴ (v1.14.6). Replicon-based incompatibility group typing was performed using PlasmidFinder⁵⁵ (v2.1), against the Enterobacteriaceae database accessed: Feb 2020(parameters: -mincov 0.50 --threshold 0.80). Acquired antimicrobial resistance genes were screened using NCBI’s AMRFinder tool v3.6.7⁵⁶, which relies on a curated AMR protein database and a collection of hidden Markov models (parameters: --ident_min 0.90 --coverage_min 0.50). Identification of insertion sequences was performed using ABRicate v0.8.11⁵⁷, against the ISFinder database (accessed: Feb 2020) (parameters: --mincov 50 --minid 80)⁵⁸.

Targeted Transposon Evaluation

We performed a targeted evaluation of transposons, which play pivotal roles in the mobilization of bla_KPC²⁷. Tn4401 has been reported to have at least six isoforms (designated as a, b, c, d, e, and i). These isoforms are characterized by distinct deletions, predominantly found upstream of bla_KPC^24,59. To delve into the intricacies of Tn4401, we applied the TETyper tool (v1.1) to short-read sequencing data, to detect and characterize single-nucleotide variants (SNVs) and deletions within Tn4401, enabling us to track specific isoforms of the transposon as well as to understand the genetic contexts they inhabit flanking inverted repeats which are the 5-bp TSD generated with Tn4401 insertion into a new genetic context as a marker of movement⁶⁰.

We also consider another transposon, Tn5403, which has been shown to integrate into Tn4401 downstream of bla_KPC on multiple occasions in epidemiologically unique settings^29,60–62. Tn5403 is a 3,663 bp Tn3-family transposon that mobilizes via replicative transposition with 5 bp TSD²³. To identify the genetic variations of Tn5403, we perform the Smith–Waterman local alignment algorithm⁶³ to all assembly sequences, with reference sequences retrieved from the TnCentral database^60,63.

The Composite-Sample Complex

The Composite-Sample Complex represents a mathematical framework for integrating the information obtained from replicon typing and the associated closed plasmids and chromosomal contigs to conceptualize evolutionary events of bla_KPC of transposition events.

Given a collection of isolates \(\:C=\{{S}_{1},\:\dots\:,{S}_{m}\}\), the Composite-Sample Complex \(\:(X\left(C\right),\:{w}_{C})\), where \(\:X\left(C\right)\) is a simplicial complex and \(\:{w}_{C}\) is a collection of simplex weights, is constructed as follows (see Supplementary Fig. 4 for an illustrative example):

Contigs labels: For each isolate, \(\:S\), label each contig with its species name (chromosome) or replicon type (plasmid). Since each isolate contains only 1 chromosome contig and distinct plasmid contigs that cannot share the same replicon type due to incompatibility. This contig labeling is effectively an injective mapping to a collective set of species names and replicon types sorted by hybrid closed genomic data from each isolate. Consequently, an isolate is represented as a set of contig labels of plasmids and chromosome, \(\:S=\{{a}_{1},\:\dots\:,{a}_{n}\}\). In this framework it is possible for labels to be shared across different isolates (e.g. IncN plasmid in multiple isolates) but does not necessarily infer that they are the same plasmid across the collection.

Initialize with 0-Simplices (Vertex Set): the 0-simplices constitute the union of all isolates in the collection \(\:C\), each represented as a set of labels. The weight of a specific vertex, \(\:a\), denoted as \(\:{w}_{C}\left(a\right)\), equals the number of isolates in \(\:C\) that contain a specific contig \(\:a\).

Step up in dimension: higher-dimensional simplices and their weights are constructed analogously. A \(\:k\)-simplex, \(\:\sigma\:=\{{a}_{1},\:\dots\:,{a}_{k+1}\}\), is part of \(\:X\left(C\right)\) if there exists at least one isolate, \(\:S{\prime\:}\), in \(\:C\), where \(\:\sigma\:\) is a subset of \(\:S{\prime\:}\). The weight of \(\:\sigma\:\), \(\:{w}_{C}\left(\sigma\:\right)\) represents the number of isolates where \(\:{a}_{1},\:\dots\:,{a}_{k+1}\) occur together. This expression is used to quantify the co-occurrence of certain elements within a given set of isolates.

The KPC Induced Composite-Sample Complex

Here we construct a model to understand the movement of a gene via transposition within an isolate and a collection of isolates using frequency of the contig with and without the gene to assign a weight \(\:w\). Suppose we are given a Composite-Sample Complex \(\:(X\left(C\right),\:{w}_{C})\) and gene \(\:KPC\), the Induced Composite-Sample Complex \(\:({X}^{KPC}\left(C\right),\:{w}_{C}^{KPC})\) is constructed as follows (see Supplementary Fig. 4 for an illustrative example):

For any simplex (isolate) \(\:\sigma\:=\{{a}_{1},\:\dots\:,{a}_{k+1}\}\) in X(C), the \(\:KPC\)-induced weight, \(\:{w}_{C}^{KPC}\left(\sigma\:\right)\) counts the number of isolates where \(\:{a}_{1},\:\dots\:,{a}_{k+1}\) co-occur, and each \(\:{a}_{i}\) in the isolate carries \(\:KPC\). The induced complex, \(\:{X}^{KPC}\left(C\right)\) contains all simplices \(\:\sigma\:\), such that \(\:{w}_{C}^{KPC}\left(\sigma\:\right)>0\).

This construction allows for the examination of relationships and co-occurrences specific to contigs carrying the \(\:KPC\).

Inferring KPC mobilization

The key idea is to consider both: the Composite-sample complex and the \(\:KPC\)-induced composite-sample complex. This provides the contextual framework for analyzing transposition events associated with g. Given a CS complex \(\:(X\left(C\right),\:{w}_{C})\) and two contig labels \(\:A\) and \(\:B\), we introduce the following criterion (#):

Target deficiency: \(\:{w}_{C}^{KPC}\left(\left[B\right]\right)<\:{w}_{C}\left(\left[B\right]\right)\). Indicates that not all contigs \(\:B\) carry KPC.

High degree of co-occurrence: \(\:{w}_{C}^{KPC}\left(\left[A,B\right]\right)=\:{w}_{C}^{KPC}\left(\left[B\right]\right)\). Every time we observe \(\:B\) carrying KPC, we also observe \(\:A\) carrying KPC in the same isolate.

Source redundancy: \(\:{w}_{C}^{KPC}\left(\left[A,B\right]\right)<\:{w}_{C}^{KPC}\left(\left[A\right]\right)\). KPC-carrying \(\:A\) does not always co-occur with KPC-carrying \(\:B\).

Suppose for a pair \(\:(A,B)\), the criterion (#) holds, then we alert to a potential transposition event mobilizing KPC from \(\:A\) to \(\:B\) in the composite sample \(\:C\). The comparison of these complexes enables us to systematically and efficiently formulate testable hypotheses, which will subsequently undergo validation through the examination of nucleotide and structural variations of bla_KPC genetic environments.

Validation of KPC Mobilization

We applied our method on UVA environmental data to infer KPC mobility patterns. To bolster the credibility of the hypothesized mobilization event, a thorough validation process is essential which was done on this longitudinal environmental data set. This validation involves cross-checking the structural and nucleotide variations of KPC itself and the genomic environment of KPC between different contigs, reinforcing the reliability of the inferred gene mobility patterns.

The validation includes verifying the consistency of bla_KPCand ISKpn6 variants between the source and target contigs. Additionally, because of the strong association between bla_KPC and Tn4401, the TETyper results for Tn4401 can be utilized to validate the KPC mobility patterns that we deduced from the composite-sample complex. Moreover, by combining these two approaches, we can systematically reconstruct the relationship between variant-flanking sequence combinations and plasmids/chromosomes contigs, thereby gaining deeper insights into the transposition events (see supplement for full validation method).

Authors and contributors

AJM: Principal investigator, responsible for sample collection, overseeing the project, and contributing to writing and reviewing the manuscript, TJL: Developed mathematical models, refined and reviewed bioinformatics, and contributed to writing and reviewing the manuscript, KEB: Conducted bacterial sequencing, performed bioinformatics analysis, and contributed to editing the manuscript, HP developed bioinformatics pipelines, performed hybrid assembly and assisted with writing and editing the manuscript, DWE: Provided feedback on the model and reviewed the manuscript, NS: Assisted with sequencing coordination and provided feedback on the model and reviewed the manuscript, DWC: Provided feedback on the model and reviewed the manuscript, ASW: Provided feedback on the model and reviewed the manuscript, SN: Refined bioinformatics methods, checked outputs, and contributed to editing the manuscript, QH: Developed and refined mathematical models, contributed to writing and reviewing the manuscript, SK: Contributed to developing figures, sample collection, reviewing bioinformatics, and writing and reviewing the manuscript, CCB: Provided feedback on the model and reviewed the manuscript, CR: Contributed to the development of the mathematical model, refined outputs, and participated in writing and reviewing the manuscript.

Conflicts of interest

The author(s) declare that there are no conflicts of interest.

Funding information

This study was funded by U.S. Centers for Disease Control and Prevention (CDC) grants Pathogen Genomic Centers of Excellence NU50CK000631-01-00 as well as BAA 200-2017-96194 and BAA 200-2018-02917.

CDC. Antibiotic Resistance Threats in the United States. (Centers for Disease Control and Prevention, Atlanta, GA, 2019).
Antimicrobial Resistance Collaborators, Global burden of bacterial antimicrobial resistance in 2019: a systematic analysis. Lancet (2022). https://doi.org/10.1016/S0140-6736(21)02724-0
Sheppard, A. E. et al. Nested Russian Doll-Like Genetic Mobility Drives Rapid Dissemination of the Carbapenem Resistance Gene blaKPC. Antimicrob Agents Chemother 60, 3767-3778 (2016). https://doi.org/10.1128/aac.00464-16
Harmer, C. J. & Hall, R. M. The A to Z of A/C plasmids. Plasmid 80, 63-82 (2015). https://doi.org/10.1016/j.plasmid.2015.04.003
Gordon, A. 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 64, 1435-1444 (2017). https://doi.org/10.1093/cid/cix132
Kotay, S., Chai, W., Guilford, W., Barry, K. & Mathers, A. J. Spread from the Sink to the Patient: in situ Study Using Green Fluorescent Protein (GFP) Expressing- Escherichia coli to Model Bacterial Dispersion from Hand Washing Sink Trap Reservoirs. Appl Environ Microbiol (2017). https://doi.org/10.1128/AEM.03327-16
WHO. No time to Wait: Securing the future from drug-resistant infections. UNAd Hoc Interagency Coordinating Group on Antimicrobial Resistance (IACG) (2019).
Hocquet, D., Muller, A. & Bertrand, X. What happens in hospitals does not stay in hospitals: antibiotic-resistant bacteria in hospital wastewater systems. J Hosp Infect 93, 395-402 (2016). https://doi.org/10.1016/j.jhin.2016.01.010
Lamba, M., Graham, D. W. & Ahammad, S. Z. Hospital Wastewater Releases of Carbapenem-Resistance Pathogens and Genes in Urban India. Environ Sci Technol 51, 13906-13912 (2017). https://doi.org/10.1021/acs.est.7b03380
Voigt, A. M. et al. The occurrence of antimicrobial substances in toilet, sink and shower drainpipes of clinical units: A neglected source of antibiotic residues. International Journal of Hygiene and Environmental Health 222, 455-467 (2019). https://doi.org/https://doi.org/10.1016/j.ijheh.2018.12.013
Element, S. J. et al. Growth in a biofilm promotes conjugation of a bla (NDM-1)-bearing plasmid between Klebsiella pneumoniae strains. mSphere 8, e0017023 (2023). https://doi.org/10.1128/msphere.00170-23
Røder, H. L. et al. Biofilms can act as plasmid reserves in the absence of plasmid specific selection. NPJ Biofilms Microbiomes 7, 78 (2021). https://doi.org/10.1038/s41522-021-00249-w
Sheppard, A. E. et al. Nested Russian Doll-like Genetic Mobility Drives Rapid Dissemination of the Carbapenem Resistance Gene blaKPC. Antimicrob Agents Chemother (2016). https://doi.org/10.1128/AAC.00464-16
Mathers, A. J. et al. Intensive Care Unit Wastewater Interventions to Prevent Transmission of Multispecies Klebsiella pneumoniae Carbapenemase-Producing Organisms. Clin Infect Dis 67, 171-178 (2018). https://doi.org/10.1093/cid/ciy052
Duchêne, S. et al. Genome-scale rates of evolutionary change in bacteria. Microb Genom 2, e000094 (2016). https://doi.org/10.1099/mgen.0.000094
Mathers, A. J. et al. Klebsiella pneumoniae carbapenemase (KPC)-producing K. pneumoniae at a single institution: insights into endemicity from whole-genome sequencing. Antimicrob Agents Chemother 59, 1656-1663 (2015). https://doi.org/10.1128/aac.04292-14
Naas, T. et al. Genetic structures at the origin of acquisition of the beta-lactamase bla KPC gene. Antimicrob Agents Chemother 52, 1257-1263 (2008). https://doi.org/10.1128/aac.01451-07
Mathers, A. J. et al. Klebsiella quasipneumoniae Provides a Window into Carbapenemase Gene Transfer, Plasmid Rearrangements, and Patient Interactions with the Hospital Environment. Antimicrob Agents Chemother 63 (2019).
Cuzon, G., Naas, T. & Nordmann, P. Functional characterization of Tn4401, a Tn3-based transposon involved in blaKPC gene mobilization. Antimicrob Agents Chemother 55, 5370-5373 (2011). https://doi.org/10.1128/aac.05202-11
Huang, J. et al. Comparative Analysis of bla(KPC) Expression in Tn4401 Transposons and the Tn3-Tn4401 Chimera. Antimicrob Agents Chemother 63 (2019). https://doi.org/10.1128/aac.02434-18
Naas, T., Cuzon, G., Truong, H. V. & Nordmann, P. Role of ISKpn7 and deletions in blaKPC gene expression. Antimicrob Agents Chemother 56, 4753-4759 (2012). https://doi.org/10.1128/aac.00334-12
Sugita, K. et al. Molecular Analysis of bla(KPC-2)-Harboring Plasmids: Tn4401a Interplasmid Transposition and Tn4401a-Carrying ColRNAI Plasmid Mobilization from Klebsiella pneumoniae to Citrobacter europaeus and Morganella morganii in a Single Patient. mSphere 6, e0085021 (2021). https://doi.org/10.1128/mSphere.00850-21
Rinkel, M., Hubert, J. C., Roux, B. & Lett, M. C. Identification of a new transposon Tn5403 in a Klebsiella pneumoniae strain isolated from a polluted aquatic environment. Curr Microbiol 29, 249-254 (1994). https://doi.org/10.1007/bf01577436
Cheruvanky, A. et al. Enhanced Klebsiella pneumoniae Carbapenemase Expression from a Novel Tn4401 Deletion. Antimicrob Agents Chemother 61 (2017). https://doi.org/10.1128/aac.00025-17
Kamruzzaman, M., Mathers, A. J. & Iredell, J. R. A Novel Plasmid Entry Exclusion System in pKPC_UVA01, a Promiscuous Conjugative Plasmid Carrying the bla(KPC) Carbapenemase Gene. Antimicrob Agents Chemother 66, e0232221 (2022). https://doi.org/10.1128/aac.02322-21
Barry, K. E. et al. Don't overlook the little guy: An evaluation of the frequency of small plasmids co-conjugating with larger carbapenemase gene containing plasmids. Plasmid 103, 1-8 (2019). https://doi.org/10.1016/j.plasmid.2019.03.005
Partridge, S. R., Kwong, S. M., Firth, N. & Jensen, S. O. Mobile Genetic Elements Associated with Antimicrobial Resistance. Clin Microbiol Rev 31 (2018). https://doi.org/10.1128/cmr.00088-17
Tazzyman, S. J. & Bonhoeffer, S. Why There Are No Essential Genes on Plasmids. Mol Biol Evol 32, 3079-3088 (2015). https://doi.org/10.1093/molbev/msu293
He, S. et al. Mechanisms of Evolution in High-Consequence Drug Resistance Plasmids. mBio 7 (2016). https://doi.org/10.1128/mBio.01987-16
Carlsson, G. E. Topology and data. Bulletin of the American Mathematical Society 46, 255-308 (2009).
Wasserman, L. Topological Data Analysis. Annual Review of Statistics and Its Application 5, 501-532 (2018). https://doi.org/10.1146/annurev-statistics-031017-100045
Chazal, F. & Michel, B. An Introduction to Topological Data Analysis: Fundamental and Practical Aspects for Data Scientists. Front Artif Intell 4, 667963 (2021). https://doi.org/10.3389/frai.2021.667963
Moore, T. J., Drost, R. J., Basu, P., Ramanathan, R. & Swami, A. in 2012 Proceedings IEEE INFOCOM Workshops. 238-243.
Giusti, C., Ghrist, R. & Bassett, D. S. Two's company, three (or more) is a simplex : Algebraic-topological tools for understanding higher-order structure in neural data. J Comput Neurosci 41, 1-14 (2016). https://doi.org/10.1007/s10827-016-0608-6
Barrett, C. et al. Motifs in SARS-CoV-2 evolution. Rna 30, 1-15 (2023). https://doi.org/10.1261/rna.079557.122
Singh, Y. et al. Topological data analysis in medical imaging: current state of the art. Insights Imaging 14, 58 (2023). https://doi.org/10.1186/s13244-023-01413-w
Skaf, Y. & Laubenbacher, R. Topological data analysis in biomedicine: A review. J Biomed Inform 130, 104082 (2022). https://doi.org/10.1016/j.jbi.2022.104082
Mathers, A. J., Carroll, J., Sifri, C. D. & Hazen, K. C. Modified Hodge Test versus Indirect Carbapenemase Test: Prospective Evaluation of a Phenotypic Assay for Detection of Klebsiella pneumoniae Carbapenemase (KPC) in Enterobacteriaceae. J Clin Microbiol 51, 1291-1293 (2013). https://doi.org/10.1128/JCM.03240-12
TrimGalore- A wrapper tool around Cutadapt and FastQC to consistently apply quality and adapter trimming to FastQ files (2015).
Bolger, A. M., Lohse, M. & Usadel, B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114-2120 (2014). https://doi.org/10.1093/bioinformatics/btu170
Ondov, B. D. et al. Mash: fast genome and metagenome distance estimation using MinHash. Genome Biology 17, 132 (2016). https://doi.org/10.1186/s13059-016-0997-x
Li, H. Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. (2013). <https://arxiv.org/abs/1303.3997>.
Garrison, E. & Marth, G. Haplotype-based variant detection from short-read sequencing. (2012). <https://arxiv.org/abs/1207.3907>.
snippy: fast bacterial variant calling from NGS reads v. version 0.4.3 (2015).
Kalyaanamoorthy, S., Minh, B. Q., Wong, T. K. F., von Haeseler, A. & Jermiin, L. S. ModelFinder: fast model selection for accurate phylogenetic estimates. Nat Methods 14, 587-589 (2017). https://doi.org/10.1038/nmeth.4285
Nguyen, L. T., Schmidt, H. A., von Haeseler, A. & Minh, B. Q. IQ-TREE: a fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies. Mol Biol Evol 32, 268-274 (2015). https://doi.org/10.1093/molbev/msu300
pyjar: A python implementation of the joint ancestral state reconstruction algorithm (GitHub https://github.com/simonrharris/pyjar).
Pupko, T., Pe'er, I., Shamir, R. & Graur, D. A fast algorithm for joint reconstruction of ancestral amino acid sequences. Mol Biol Evol 17, 890-896 (2000). https://doi.org/10.1093/oxfordjournals.molbev.a026369
Jolley, K. A. & Maiden, M. C. BIGSdb: Scalable analysis of bacterial genome variation at the population level. BMC Bioinformatics 11, 595 (2010). https://doi.org/10.1186/1471-2105-11-595
Nurk, S. et al. Assembling single-cell genomes and mini-metagenomes from chimeric MDA products. J Comput Biol 20, 714-737 (2013). https://doi.org/10.1089/cmb.2013.0084
mlst Scan contig files against traditional PubMLST typing schemes (GitHub: https://github.com/tseemann/mlst).
filtlong - quality filtering tool for Nanopore and PacBio reads (GitHub https://github.com/rrwick/Filtlong ).
Wick, R. R., Judd, L. M., Gorrie, C. L. & Holt, K. E. Unicycler: Resolving bacterial genome assemblies from short and long sequencing reads. PLoS Comput Biol 13, e1005595 (2017). https://doi.org/10.1371/journal.pcbi.1005595
Seemann, T. Prokka: rapid prokaryotic genome annotation. Bioinformatics 30, 2068-2069 (2014). https://doi.org/10.1093/bioinformatics/btu153
Carattoli, A. & Hasman, H. PlasmidFinder and In Silico pMLST: Identification and Typing of Plasmid Replicons in Whole-Genome Sequencing (WGS). Methods Mol Biol 2075, 285-294 (2020). https://doi.org/10.1007/978-1-4939-9877-7_20
Feldgarden, M. et al. Validating the AMRFinder Tool and Resistance Gene Database by Using Antimicrobial Resistance Genotype-Phenotype Correlations in a Collection of Isolates. Antimicrob Agents Chemother 63 (2019). https://doi.org/10.1128/aac.00483-19
T, S. ABRicate. <https://github.com/tseemann/abricate>.
Siguier, P., Perochon, J., Lestrade, L., Mahillon, J. & Chandler, M. ISfinder: the reference centre for bacterial insertion sequences. Nucleic Acids Res 34, D32-36 (2006). https://doi.org/10.1093/nar/gkj014
Araújo, B. F. et al. Insights into a novel Tn4401 deletion (Tn4401i) in a multidrug-resistant Klebsiella pneumoniae clinical strain belonging to the high-risk clonal group 258 producing KPC-2. Int J Antimicrob Agents 52, 525-527 (2018). https://doi.org/10.1016/j.ijantimicag.2018.08.011
Sheppard, A. E. et al. TETyper: a bioinformatic pipeline for classifying variation and genetic contexts of transposable elements from short-read whole-genome sequencing data. Microb Genom 4 (2018). https://doi.org/10.1099/mgen.0.000232
Rada, A. M. et al. Dynamics of bla(KPC-2) Dissemination from Non-CG258 Klebsiella pneumoniae to Other Enterobacterales via IncN Plasmids in an Area of High Endemicity. Antimicrob Agents Chemother 64 (2020). https://doi.org/10.1128/aac.01743-20
Salamzade, R. et al. Inter-species geographic signatures for tracing horizontal gene transfer and long-term persistence of carbapenem resistance. Genome Med 14, 37 (2022). https://doi.org/10.1186/s13073-022-01040-y
Smith, T. F. & Waterman, M. S. Identification of common molecular subsequences. J Mol Biol 147, 195-197 (1981). https://doi.org/10.1016/0022-2836(81)90087-5

There is NO Competing Interest.

Download PDF

Version 1

posted

You are reading this latest preprint version

Developing a novel approach to complex evolution of antimicrobial resistance via mobile genetic elements from a persistent environmental biofilm

Status:

Version 1

Abstract

Figures

Introduction

Results

Distribution of species and strains

Transposition: Plasmid to chromosome

Transposition: plasmid to plasmid

Conjugation

Discussion

Methods

Sampling Methodology and Approach

Whole-genome sequencing (WGS)

Bioinformatics Analysis

Targeted Transposon Evaluation

The Composite-Sample Complex

The KPC Induced Composite-Sample Complex

Inferring KPC mobilization

Validation of KPC Mobilization

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1