Dissemination of a IncHI2A plasmid co-harboring the mcr-9 and bla_NDM-1 genes in Israeli hospitals

Abstract

The mcr-9 allele is typically located on IncHI2 type plasmids, but there are only few reports describing the co-occurrence of the bla_NDM gene on the same plasmid. Our aims were to describe the spread of an IncHI2A plasmid harboring both the mcr-9 and the bla_NDM-1 genes in a multicenter study in Israel. All New-Delhi Metallo-β-lactamase-producing Enterobacterales (NDME) isolated from three medical centres in Israel between January 2018 and July 2019 were included. The mcr-9.1 gene was identified in 37/212 (17.4%) of the isolates, mostly in Enterobacter cloacae (34/37, 92%). The mcr-9.1 gene was also identified in Klebsiella pneumoniae sequence types (ST)-76 (n = 2) and Escherichia coli ST-69 (n = 1). In one hospital, out of 32 E. cloacae isolates, 19 (51.35%) were clustered into five transmission clusters of the ST-511, ST-1261 and ST-1750. Four subtypes of a ~ 290 kb IncHI2A type plasmid were identified in all isolates that co-harbored the mcr-9.1 and the bla_NDM-1 genes. This plasmid was identified in all isolates, with four sub-communities (sc), with sc-4 identified in all three species. The resistance genes were surrounded by the IS26 (mcr-9.1) or by the ISAba125 and the IS300 (bla_NDM-1) mobile elements. The dissemination of the mcr-9.1 and the bla_NDM-1 genes was accelerated via clonal spread and the dual carriage on a single plasmid.

Introduction

Since first identified in the early 2010s [1], the New Delhi Metallo-β-lactamase (NDM) has become widespread in Israeli hospitals, and it is now the most common carbapenemase type among carbapenemase-producing Enterobacterales (CPE) [2, 3]. Since NDM CPE are not susceptible to most of the novel β-lactam/β-lactamase inhibitors agents, colistin remains the last therapeutic resort in many cases. In 2016, a novel mobile colistin resistance (mcr) gene was detected in Escherichia coli isolated in China [4]. Ten different mcr alleles (mcr-1 through 10) have been described so far globally [5].

The mcr-9 gene allele was the first reported in 2019 from the USA [6], but since then it has been reported in 40 countries, followed by the mcr-1 allele [6]. The mcr-9 allele is typically harbored by IncHI2 type plasmids, which were reported to carry additional antimicrobial resistance (AMR) genes, but only few reports described the co-occurrence of the bla_NDM gene on the same plasmid [7]. Due to the transmissibility of the mcr-9 harboring plasmid, it presents the potential for dual spread of these AMR genes. In this report, we describe the spread of an IncHI2A plasmid harboring both the mcr-9 and the bla_NDM-1 genes in a multicenter study in Israel, analyzing the molecular and microbiological characteristics of these isolates.

Methods

NDME isolates were collected as part of a multicenter survey, conducted in three medical center in Israel between January 2018 and July 2019. The isolates were sequenced by using the Illumina MiSeq platform and searched for antibiotic resistance genes using Abricate against the database Resfinder [8]. Antimicrobial susceptibility testing to colistin was performed using the Sensititre™ Gram Negative GNX3F AST plates (Thermo Fisher Scientific).

The phylogenetic analyses included multi-locus sequence typing (MLST) and single-nucleotide polymorphisms (SNPs) analysis as described previously [3]. To identify putative transmission chains, a threshold of five or less SNPs between isolates was used.

To identify the genetic environment of the mcr-9.1 gene, the long-read Oxford Nanopore Technology was used to sequence 35 of the isolates. Hybrid assemblies from short and long read sequences were used to assemble the plasmids with Hybracter (https://github.com/gbouras13/hybracter), from which 32 were circularized and used for further analyses. The plasmid contigs were identified using Abricate by the PlasmidFinder database [9] and the location of the resistance genes was determined using the NCBI database with Abricate. Plasmid similarity was compared using Pling (https://github.com/iqbal-lab-org/pling/tree/main?tab=readme-ov-file) [10] and the genomic structure was compared between the isolates as well as with global public datasets.

Results

A total of 212 NDME isolates from surveillance or clinical cultures of hospitalized patients were analyzed [3]. The mcr-9.1 gene was identified in 37/212 (17.4%) of the NDME isolates (Table 1). The most common species harboring the mcr-9.1 gene (mcr-NDME) was Enterobacter cloacae complex (33/37, 89%), followed by Klebsiella pneumoniae (n = 3), and Escherichia coli (n = 1). Almost all mcr-NDME isolates (n = 35) were detected at the Tel-Aviv Sourasky Medical Center (TASMC), with one isolate detected at the Rambam Medical Center (RMC) and the Sha’are-Zedek Medical Center (SZMC) each. All mcr-NDME isolates harbored the bla_NDM-1 gene allele, while the single E. cloacae isolate from SZMC also harbored the bla_KPC-2 gene. The colistin MIC in 35/37 of mcr-NDME isolates was ≤ 2 mg/L. In two E. cloacae isolates (ST-1750), the MIC was 16 and 128 mg/L, respectively. Of note, the elevated MIC was detected also by the VITEK-2 AST system.

Table 1.

Molecular features of enterobacterales isolates co-harboring the bla_NDM-1 and the mcr-9.1 genes

Center1,2	Species1	Sequence Type1	Putative transmission clusters1,3
TASMC (35)	E. cloacae complex (32)	511 (13/3,2), 1261 (10/4), 106 (2/4), 1750 (2/4), 104, 269, 1748, 1749, 1751	511_1 (2), 511_2 (4), 1261_1 (4), 1261_2 (6), 1750_1 (2).
	K. pneumoniae (2)	76 (2/4)	76_1 (2)
	E. coli (1)	69 (1/4)	NA4
RMC (1)	E. cloacae complex	93	NA
SZMC (1)	E. cloacae complex	1752	NA

¹The number of isolates and the plasmid sub-type (if n ≥ 2) is given in parenthesis (n/sub-type); ²Centers: Tel Aviv Sourasky Medical Center (TASMC), Rambam Medical Center (RMC), Shaa’re Zedek Medical Center (SZMC); ³ST_cluster number [3]; ⁴NA not applicable

The most common E. cloacae sequence types (ST) were ST-511 (n = 13) and ST-1261 (n = 10). From a total of 32 E. cloacae from TASMC, 19 isolates (59%) were part of five putative transmission clusters, as were the two ST-76 K. pneumoniae isolates from TASMC.

The bla_NDM-1 and mcr-9.1 genes were located on a ~ 290 kb IncHI2A plasmid [7] (Fig. 1) across all the plasmids with four separate sub-types (sc1-4). The distribution of the sub-types in the different species and ST’s is presented in Table 1. The sc-4 was the most common and was present in different E. cloacae ST’s as well as in the E. coli and K. pneumoniae isolates. The sc-2 and sc-3 were present in E. cloacae ST-511 and the sc-1 was identified in one E. cloacae ST-1752 isolate. The resistance islands were surrounded by the IS26 (mcr-9.1 gene) or by the ISAba125 and the IS300 (bla_NDM-1 gene) mobile elements. The qseB- and qseC-like genes that are potentially involved in the regulation of the mcr-9.1 gene [11] were not detected. Comparison with global public data (https://microreact.org/project/cBSj4t5L5PZPY14WCuNiCN-mcrproject) showed that only four plasmids carrying both the mcr-9.1 and the bla_NDM-1 genes were identified: three reported from China (NZ_CP068288.1, NZ_MH909331.1 and pK710429_2) and one from Nepal (NZ_LC542971.1).

Fig. 1.

The bla_NDM-1 and mcr-9.1 genes harboring IncHI2A plasmid. The four plasmid sub-types are presented on the backbone of the CP073658 plasmid [7]

Discussion

In this study, we analyzed the dissemination of a ~ 290 kb IncHI2A plasmid, co-harboring the mcr-9.1 and the bla_NDM-1 genes. This plasmid was identified in 17% of the NDME isolates, a much higher rate compared with the 1% reported in another study [12]. The high rate of mcr-9 identified in TASMC (35/104, 33%) was due to the spread of several E. cloacae ST’s (ST-511, -1261), which differed from those previously reported by Macesic et al. [13]. In addition to clonal spread, the presence of the same plasmid in different clones and species, suggests that at least part of its dissemination was via horizontal gene transfer.

Enterobacter species were reported as a common species harboring the mcr-9 gene in other studies [6, 12, 14, 15]. The relative predominance of E. cloacae versus other NDME species in TASMC [3], is thus the likely explanation for the high rate of mcr-9.1-harboring NDME in comparison with the other centers. Likewise, the IncHI2A plasmid in our study was reported as the predominant mcr-9 gene harboring type [14] and was similar to a plasmid identified in China [7]. The plasmids in our study were divided into four sub-types (Fig. 1; Table 1) that had similar regions flanking the bla_NDM-1 and the mcr-9.1 genes, the later surrounded by IS26 as previously reported [12–14]. The presence of the plasmid sc-4 in different ST’s and species, suggests that this sub-type facilitated the spread of the bla_NDM-1/mcr-9.1-harboring plasmid.

Except from two resistant isolates, the remaining isolates were susceptible to colistin, a phenomenon commonly found in mcr-9 harboring isolates [12, 13]. Since screening of this gene is seldom done in clinical settings, it is likely that some patients with mcr-NDME infections may be treated with colistin. Thus, the outcome of these patients can be studied to determine the clinical relevance of this gene.

Author contributions

AA—conception and study design, data interpretation, manuscript drafting; SAM—data analysis and interpretation, manuscript drafting; AG—data analysis and interpretation; AR—study design, data analysis and interpretation; MVA—conception, data acquisition; YG—conception, data acquisition; SR—conception and study design, data interpretation, manuscript drafting.

Funding

The study was funded by the German Israeli Foundation Grant I-56-416.6-2016.

Data availability

The NGS datasets generated during the current study are available in the European Nucleotide Archive project number PRJEB53419.

Declarations

Ethics approval and consent to participate

The study was approved by the Ethics Committees of all three centers (0306-21-SZMC, 0508-19-RMB, 04590-19-TLV) with waiver of informed consent.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.