Abstract
Colistin represents one of the few available drugs for treating infections caused by carbapenem-resistant Enterobacteriaceae. As such, the recent plasmid-mediated spread of the colistin resistance gene mcr-1 poses a significant public health threat, requiring global monitoring and surveillance. Here, we characterize the global distribution of mcr-1 using a data set of 457 mcr-1-positive sequenced isolates. We find mcr-1 in various plasmid types but identify an immediate background common to all mcr-1 sequences. Our analyses establish that all mcr-1 elements in circulation descend from the same initial mobilization of mcr-1 by an ISApl1 transposon in the mid 2000s (2002–2008; 95% highest posterior density), followed by a marked demographic expansion, which led to its current global distribution. Our results provide the first systematic phylogenetic analysis of the origin and spread of mcr-1, and emphasize the importance of understanding the movement of antibiotic resistance genes across multiple levels of genomic organization.
Generated Summary
This research article presents a comprehensive investigation into the global distribution and spread of the mobilized colistin resistance gene, mcr-1. The study utilizes a dataset of 457 mcr-1-positive sequenced isolates to characterize the gene’s prevalence across different plasmid types and to analyze its evolutionary history. The research employs whole-genome sequencing (WGS) data, including 110 new WGS from China, alongside publicly available sequence data from the NCBI repository and the Short Read Archive (SRA). The primary objective is to elucidate the origin, acquisition, emergence, and spread of mcr-1, which poses a significant public health threat due to its association with carbapenem-resistant Enterobacteriaceae. The methodology includes phylogenetic analysis to trace the origins and spread of mcr-1, emphasizing the importance of understanding the movement of antibiotic resistance genes across multiple levels of genomic organization. The study uses Bayesian dating to infer a time phylogeny of the mcr-1 carrying transposon and explores the broader genomic context of mcr-1, including plasmid types and the distribution of transposons.
Key Findings & Statistics
- The study analyzed a dataset of 457 mcr-1-positive sequenced isolates.
- Mcr-1 was identified in various plasmid types.
- The analysis established that all mcr-1 elements originated from the same initial mobilization event in the mid-2000s (2002-2008; 95% highest posterior density).
- The countries with the largest numbers of mcr-1-positive samples are China (212), Vietnam (58), and Germany (25).
- In China, nearly half (45%) of positive isolates stem from the Shandong province.
- The majority of mcr-1-positive isolates belong to E. coli (411).
- The data set also includes mcr-1-positive isolates from seven other bacterial species: Salmonella enterica (29), Klebsiella pneumoniae (8), Escherichia fergusonii (2), Kluyvera ascorbata (2), Citrobacter braakii (2), Cronobacter sakazakii (1), and Klebsiella aerogenes (1).
- The majority of isolates for which sampling dates were available (80%), were collected between 2012 and 2016, with the oldest available isolates dating back to 2008.
- Isolates with metadata on the sample source (n = 360) came from a range of animal (n = 222), human (n = 108), and environmental (n = 30) hosts.
- The large number of mcr-1-positive isolates from China, and the high incidence in the Shandong province can be largely ascribed to the inclusion of our 110 newly sequenced isolates including 49 from Shandong and to another 37 isolates from a previous large sequencing effort.
- A shared region or remnants of it was identified in all 457 sequences surrounding mcr-1.
- 42 sequences contained indication of the presence of ISApll both upstream and downstream.
- The 186 bp region between the 3′-end of the upstream ISApll and mcr-1 contained IR-like sequences similar to the IRR and IRL of ISApll (respectively: 93-142 bp, 23/50 identity; and 125-175 bp, 21/50 identity).
- A small number of sequences (3%) had SNPs present in mcr-1 itself.
- 28 sequences from Vietnam included a secondary 1.7 kb insertion downstream of mcr-1.
- The study included 110 new WGS from China, of which 107 were sequenced with Illumina short reads and three with PacBio long-read technology.
- One hundred and ninety-five isolates were sourced from publicly available assemblies in the NCBI GenBank repository (73 completed plasmids, 1 complete chromosome, 121 assemblies).
- A further 152 sequences were sourced from the NCBI SRA, after being identified as mcr-1-positive using a k-mer index of a snapshot of the SRA as of December 2016.
- The whole data set consists of 256 short-read data sets, 6 long-read PacBio WGS, 121 draft assemblies, and 74 completed assemblies.
- Isolates carrying mcr-1 were identified from 31 countries.
- The 186 bp region between the 3′-end of the upstream ISApll and mcr-1 contained IR-like sequences similar to the IRR and IRL of ISApl1 (respectively: 93-142 bp, 23/50 identity; and 125-175 bp, 21/50 identity).
- The most variable positions in this 186 bp region were at 177 bp and 142 bp, approximately coinciding with the end of the alignment with the IRs and were more variable in sequences lacking ISApll, suggesting possible loss of function of the transposition pathway associated with ISApll.
- A small number of sequences (3%) had SNPs present in mcr-1 itself.
- A subset of the sequences from Vietnam (n = 28) included a secondary 1.7 kb insertion downstream of mcr-1.
- The distribution of transposons carrying one or two copies of ISApll was highly heterogeneous across these plasmid types.
- The first alignment encompasses a shared sequence of 7161 bp between 108 plasmid backgrounds.
- The second alignment is 34,761 bp long and is common to nine IncX4 plasmids.
- The IncI2 background we infer that a common ancestor to all 108 isolates existed in 2006.
- For the IncX4 backgrounds we dated the common ancestor of the eight isolates to 2011.
- The inferred mutation rates obtained for the IncI2 and IncX4 backgrounds consistently lie around 5–10 × 10−5 substitutions per site per year.
- Comparing human (n = 108) and non-human (n = 252) isolates, there were significantly more sequences with some trace of the insertion sequence ISApll both upstream and downstream in non-human isolates (32/220 vs. 5/108, x²-test, p = 0.033).
- Of the 42 isolates that had ISApll fragments both upstream and downstream, the majority were from Asia (n = 30) with only a quarter from Europe (n = 10) (x² test, p = 0.12).
Other Important Findings
- The study identifies a common immediate background for mcr-1 in all sequenced elements, supporting a single origin for mcr-1 elements.
- The current distribution of mcr-1 suggests a possible origin in Chinese livestock, with the spread potentially driven by plasmid transfer and horizontal gene transfer.
- The study highlights the complex dynamics of antibiotic resistance genes across multiple embedded genetic levels (transposons, plasmids, bacterial lineages, and bacterial species).
- The study emphasizes the role of the transposon in the mobilization of mcr-1 and the subsequent stabilization of mcr-1 in various plasmid backgrounds.
- The study notes the presence of ISApll flanking elements and the role of these elements in the transposition mechanism.
- The analysis suggests that agricultural settings may be acting as the source of mcr-1 within bacteria isolated from humans.
- The study found that IncI2 and IncX4 are the dominant plasmid types.
- The study indicates the complex interplay of the transposon, plasmids, and host bacteria in driving the spread of mcr-1.
- The study highlights that the mcr-1 transposon does not evolve differently in these different backgrounds.
- The study suggests that the spread of mcr-1 will not be contained by these bans.
- The presence of colistin resistance on mobile genetic elements poses a significant public health risk.
- The study identified that the ancestral mobilizable state is represented by the transposon carrying both its ISApll elements.
- A single integration event of mcr-1 into an ISApll composite transposon followed by its subsequent spread between multiple genomic backgrounds.
- The study infers that the origin of mcr-1 prior to its mobilization remains elusive.
Limitations Noted in the Document
- The data may be affected by sampling biases, with an overrepresentation of samples from regions with active surveillance and well-funded research.
- The study acknowledges limitations in reaching strong inferences on some important aspects of mcr-1 spread due to the complex dynamics of the transposon, plasmids, and bacterial host.
- The number of WGSs for samples before 2012 is limited, which may impact the accuracy of certain estimations.
- The lack of detailed information on the origin of mcr-1 outside the ISApl1 sequence background.
Conclusion
The study concludes that the global distribution and spread of the mcr-1 gene is characterized by a single integration event into an ISApl1 composite transposon, followed by diversification and dissemination across various genomic backgrounds. The findings point to a common origin in the mid-2000s, with a potential link to Chinese livestock. The research underscores the importance of understanding the mechanisms driving the spread of antibiotic resistance genes, especially within complex genomic structures. The study also emphasizes the need for surveillance and the development of strategies to mitigate the spread of mcr-1 and other antibiotic resistance genes. This study provides insights into the dynamic interplay of mobile genetic elements, plasmid types, and host bacteria. The authors highlight that the spread of the mcr-1 element is not constrained by the agricultural bans implemented in China and Brazil and that the global distribution is achieved through multiple translocations. The study also points to the potential origin of the mcr-1 gene outside the ISApll sequence background and emphasizes the need for further research to pinpoint the source and understand the evolutionary dynamics of these resistance genes. The presence of similar sequences downstream of other mcr genes implies some functional role, either in the formation of the mobile element and/or in its continued mobilization. The study suggests agricultural settings may be acting as a source for the mcr-1 gene and that trade and global movement of colonized humans may be contributing to the current distribution.