Potential community acquisition of NMC-A-producing <i>Enterobacter ludwigii</i>, an underestimated environmental threat?

Carbapenemase-producing Enterobacteriaceae (CPE) have emerged in the two last decades and are a global threat to public health. These bacteria were then classified as critical priority pathogens by the WHO for research and antibiotic development.1 In France, monitoring the digestive carriage of CPE is recommended for patients hospitalized in ICUs or returning from endemic areas. This screening allows the hospital staff to take hygiene measures to avoid serious nosocomial outbreaks. Here, we report two concomitant cases of IMI-producing Enterobacter ludwigii colonization at Montpellier University Hospital, France. IMIs are Ambler class A carbapenemases first described in 1993 and relatively uncommon in mainland France.2,3 E. ludwigii isolates were identified in rectal samples from two patients hospitalized in September 2020. The first patient was admitted to the orthopaedic ward for hip replacement surgery. Because he was sharing his room with a patient carrying an OXA-48-producing Escherichia coli, CPE screening was requested. In the same week, the laboratory received a sample from another patient hospitalized in the cardiology department for coronary angiography. CPE screening was requested because this patient had been in contact with another patient carrying a carbapenem-resistant Acinetobacter baumannii complex organism. Two isolates from these patients grew on chromID CARBA SMART selective medium (bioMérieux, Marcy-l’Étoile, France), and were identified as Enterobacter cloacae complex using MALDI-TOF MS (Bruker Daltonics, Germany). Antibiotic susceptibility testing using the disc diffusion method, interpreted according to the CASFM-EUCAST 2019 clinical breakpoints, showed resistance to carbapenems (https://www.sfm-microbiologie.org/2020/04/07/casfm-eucast-v1-0-avril-2020); see Table S1 (available as Supplementary data at JAC Online). As the NG-Test® CARBA 5 immunochromatographic assay (NG Biotech, France) was negative, the KPC/MBL and OXA-48 Confirm Kit (Rosco Diagnostica, Denmark) was used and a class A carbapenemase was detected for both strains. When confronted with this antimicrobial susceptibility profile, a rapid change towards yellow in the CARBA NP test is a good indicator of the production of these carbapenemase types.4 Unfortunately, a colorimetric hydrolysis assay such as the CARBA NP test was not carried out in 2020. Given the extremely rare occurrence of this resistance phenotype, these isolates were sequenced by MicrobesNG (Birmingham, UK) using the Oxford Nanopore® and Illumina® technologies to investigate the origin of the contamination. Genomic sequences were deposited on NCBI (accession numbers: GCF_031851605.1 and GCF_031851585.1). The genomic data indicated that both isolates belonged to the species E. ludwigii, were ST374, and harboured the same carbapenemase gene (blaNMC-A). This gene encodes NMC-A carbapenemase, very close to IMI-type carbapenemases.5 In the literature, another NMC-A-producing E. ludwigii ST374 strain was isolated in a Canadian hospitalized patient in 2014, but the genome sequence was not deposited in databases.6 In the NCBI database, six other E. ludwigii ST374 genomes were available, isolated from human and aquatic samples from the USA and the UK. Phylogenetic analyses showed that the two isolates under study had only one SNP in their common genome, establishing their clonality (Figure 1a). This common genome (4080 genes) represented 88.2% of the total genome of these two strains (on 4624 genes for GCF_031851605.1 and 4627 for GCF_031851585.1). Characterization of the genetic environment of blaNMC-A showed that it was inserted in the E. ludwigii chromosome in both isolates, on the same EcloIMEX-1 integrative mobile element as previously described (Figure 1b).7 This mobile genetic element is closer to the EludIMEX-1 initially described by Antonelli et al. than the EcloIMEX-1 described by Boyd et al. (2 SNPs versus 241). However, given that the genetic structures of EcloIMEX-1 and EludIMEX-1 are similar, and because most recent studies used this name to characterize the mobile genetic element carrying blaNMC-A,3,6 we preferred the name EcloIMEX-1 to describe the element carried by our isolates. (a) Neighbour-joining tree based on eight E. ludwigii ST374 common genomes [the two clinical strains from Montpellier University Hospital (in red) and genomes available on NCBI], rooted using the E. ludwigii reference genome (GCF 001750725.1). On the right of the tree, a SNP distance matrix between these genomes is represented. The presence (green square) or absence (white square) of the IMI-type carbapenemase gene (blaNMC-A) is indicated. The origin of each genome (human or water) is indicated by symbols on the right. Genomes included in this comparison were annotated with Bakta (v1.6.1, https://github.com/oschwengers/bakta) and aligned with Roary (v3.13.0, https://sanger-pathogens.github.io/Roary). The tree was generated with FastTree (v2.1.10, http://www.microbesonline.org/fasttree). (b) Genetic structure of the integrative mobile element carrying the carbapenemase gene in the two studied isolates, compared with the reference structure of EcloIMEX-1 (GenBank accession number: KY033370) and EludIMEX-1 (GenBank accession number: KR919803). Red, blaNMC-A and its transcriptional regulator; yellow, genes encoding the recombinase/resolvase pair required to mobilize this element; brown, setB and yeiP framing the genetic element; green, other genes. Homology between sequences is indicated by shades of grey. Genetic structures were annotated with Bakta v1.6.1) The figure was drawn using Easyfig (v2.2.5, https://mjsull.github.io/Easyfig). This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. To explore this clonal dissemination, an epidemiological investigation was carried out. As these two patients were hospitalized in two different wards and because no epidemiological link was found, nosocomial cross-contamination was ruled out. We hypothesize a potential common community source of contamination, as already been described in Mayotte during an outbreak of IMI-1-producing E. cloacae.8 Even if IMI-type carbapenemases are rare in clinical practice (0.9% of all the CPE isolated in France in 2020),9 they are more frequently isolated from aquatic environmental samples (rivers, lakes)10 and from seafood (e.g. shrimps, shellfish).11 An aquatic environmental origin seemed likely because the two patients lived in two small towns around Thau’s pond, near the Mediterranean Sea. This place is known for the production of oysters and other shellfish for human consumption. Patients may have been exposed to these pathogens by ingesting contaminated seafood. Alternatively, they could have been directly exposed to IMI-producing isolates during swimming because the isolates were identified during the bathing season. Our laboratory reported this contamination route for a previous patient with IMI-2-producing Enterobacter asburiae bacteraemia, following drowning in a river near Montpellier.12 Another possibility is indirect exposure via food11 or marine avian fauna. A Swedish study recently reported IMI-3-producing E. ludwigii carriage in gulls.13 As these seabirds have synurbic activity (i.e. mostly live in urban areas), they could play a vector role between an aquatic reservoir and humans. It would be interesting to investigate the presence of IMI-producing Enterobacteriaceae in these ecosystems (aquatic environment, seafood, seagulls). Finally, although blaNMC-A was inserted in the chromosome of the two strains described here, other IMI variants (especially IMI-2 and IMI-3) are inserted in plasmids, and have been detected in species other than the E. cloacae complex,14 thus increasing their potential risk of dissemination. Indeed, a hospital outbreak of IMI-17-producing E. ludwigii (gene carried on a plasmid) was previously reported in Israel, likely originating from the community before spreading in the hospital.15 It cannot be excluded that this phenomenon might eventually extend to other IMI variants. This might lead to their emergence in humans. In conclusion, this report reinforces the need to study CPE dissemination using a One Health approach in order to identify the transmission routes and limit the spread of these genes. In addition, through this study we hope to highlight the existence of community-acquired CPE that can be carried by patients without any previous antibiotic exposure. We acknowledge the ISO 9001-certified IRD itrop HPC (member of the South Green Platform) at IRD Montpellier for providing HPC resources that have contributed to the research results reported within this paper; https://bioinfo.ird.fr/; http://www.southgreen.fr. This study was carried out as part of our routine work. All the authors: none to declare. Table S1 is available as Supplementary data at JAC Online.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.