1 Introduction
The therapeutic success of antibiotics is challenged by the spread of antimicrobial resistance genes (ARGs) among bacteria that cause healthcare-associated and community-acquired infections, an escalating global health threat (Arias & Murray, 2009; Davies & Davies, 2010; WHO, 2014). In 2017, the World Health Organization released a list of pathogens against which healthcare systems urgently need new antimicrobial alternatives due to their commonly extensive antibiotic resistance profiles (WHO, 2017). These include Klebsiella pneumoniae, Escherichia coliEnterobacter spp., Serratia spp. and several other taxa commonly resistant to important beta-lactams, such as carbapenems and third-generation cephalosporins. Other critical pathogens on this list are fluoroquinolone-resistant Salmonella spp. and Shigella spp. These pathogens are closely related and belong to the family of Enterobacteriaceae . It is noteworthy that the medically-relevant ARGs within this family have been primarily found encoded by conjugative plasmids; mobile genetic elements that frequently mediate their dissemination among bacteria through horizontal gene transfer (HGT) (Carattoli, 2009).
The currently known variety of plasmids found within Enterobacteriaceae have been divided into 28 distinct groups according to their incompatibility (Inc); i.e. their inability to coexist over time in the same cell-line (Couturier et al., 1988; Novick, 1987; Rozwandowicz et al., 2018). In order to understand the role of specific plasmid groups in the context of ARG dissemination, it is important to understand their ecology, including the diversity of their hosts (host range). Generally, plasmid host range denotes the phylogenetic or taxonomic breadth of organisms that can carry a plasmid; however, it can be further broken down into i) which recipients the plasmid can be transferred into; ii) which recipients the plasmid can replicate within; and iii) whether the plasmid is stably maintained over time in the cell-line. The two initial stages are known as the transfer - and replication host range, respectively, and the latter is termed evolutionary host range. Since these stages occur sequentially, the host range narrows from one stage to the next (Suzuki et al., 2010). The evolutionary host range and the classification of plasmids are closely related (Redondo-Salvo et al., 2020), thus understanding the host range of archetypes representing a plasmid group reveals crucial knowledge about plasmid ecology. Recently, Redondo-Salvo and colleagues defined a host range scale for plasmids, which grades host range according to the highest taxonomic rank they distribute in; from I at the species level being very narrow, e.g. IncFIB, to VI at the phylum level being very broad, e.g. IncP1 (P1), which was shown to cross phylum level (Redondo-Salvo et al., 2020). Nonetheless, it is important to bear in mind that for specific strains or taxa within the plasmid host range, barriers may exist that prevent plasmid establishment. For example, host-encoded defense systems can form effective barriers against HGT by blocking the entry of foreign nucleic acids. These include restriction/modification (R/M) systems (Oliveira et al., 2014), CRISPR-Cas systems (Makarova et al., 2019), and Wadjet systems (Doron et al., 2018). In response, plasmids have developed various systems to evade host defense systems (anti-defense systems) e.g. Anti-R/M and Anti-CRISPR proteins, which directly interact with and inhibit R/M and CRISPR-Cas systems, respectively (Mahendra et al., 2020; Roy et al., 2020). Thus measurements of distribution breath, i.e. the taxonomic distance between the hosts in which a given plasmid is found (Redondo-Salvo et al., 2020), only describe the potential host range of a plasmid.
One of the Enterobacteriaceae plasmid groups, the IncHI1A group (HI1A), comprises conjugative plasmids typically in the size range of 75 to 400 kb (Rozwandowicz et al., 2018). HI1A plasmids are associated with the dissemination of ARGs such as the bla NDM-1 gene, which confers resistance towards carbapenems, a last resort drug used when treating extended spectrum beta-lactamase (ESBL) producing Enterobacteriaceae spp. (Carattoli, 2013; Dolejska et al., 2013). The transfer of some plasmids in the HI1A group, such as R27, is thermo-sensitive and primarily conjugate below 30° C, a feature that is speculated to promote the transmission of ARGs in the environment (Maher & Taylor, 1993; Sherburne et al., 2000). Additionally, plasmids in the HI1A group carry genes encoding thick flexible pili, which enable high conjugative transfer efficiency within both planktonic and surface-associated bacterial communities, thus emphasizing their potential to disseminate ARGs across a variety of environments (Bradley et al., 1980). Cultivation-based transfer experiments have found that HI1A plasmids can also be transferred to bacteria belonging to genera outside Enterobacteriaceae, such as Vibrio spp. and Aeromonas spp. (Maher & Taylor, 1993). Based on such experiments, and supported by bioinformatic predictions (Suzuki et al., 2010), HI1A plasmids are believed to have a wide or intermediate host range. However, an understanding of the initial host range stages, stage i) and ii), of HI1A is crucial for an improved understanding of the frequency and phylogenetic extent of plasmid-mediated HGT in the environment. Some transfer events may never extend beyond these initial host range stages, and therefore not result in stable plasmid-host association, however, still result in transfer of the plasmid from this host. Thus, a short-term host may function as a crucial stepping-stone for plasmids reaching into stable hosts.
Wastewater treatment plants (WWTPs) have been suggested to facilitate HGT of ARGs among bacteria (Guo et al., 2017; Li et al., 2018). Indeed, all known antibiotic resistance mechanisms have been found represented in WWTPs reservoirs (Rizzo et al., 2013), along with multiple mobile genetic elements, including plasmids (Rizzo et al., 2013; Zhang et al., 2011). Furthermore, studies have shown that the microbial communities of WWTPs are highly permissive towards broad host range plasmids (i.e. highly capable of taking up a given plasmid) (Jacquiod et al., 2017; Li et al., 2018). Moreover, it has been revealed that a core-permissive fraction of bacteria, capable of receiving several types of plasmids, are abundant across WWTPs (Li et al., 2018). In addition, certain bacterial taxa have repeatedly been identified as a part of the core-permissive fraction of bacteria across diverse environments (Klümper et al., 2015; Li et al., 2018; Musovic et al., 2014; Pinilla-Redondo et al., n.d.).
In this study, we investigated the host range, including the initial host range stages, and transfer efficiency of the HI1A plasmid R27 in the microbial community of urban residential sewage collected at three WWTPs located in urban areas of southern Sweden. We utilized a dual fluorescent reporter gene platform, which previously has been used to examine transfer of various plasmid groups, including IncI1 (Anjum et al., 2018, 2019) and P1 (Jacquiod et al., 2017; Klümper et al., 2015; Li et al., 2018; Musovic et al., 2014; Pinilla-Redondo et al., n.d.). This platform is based on a plasmid-encoded green fluorescent protein (gfp ) gene under the control of a lacIq repressible promoter. Donor cells, which introduce the plasmid to the community, carry a chromosomalmCherry along with lacIq and hence, although carrying the plasmid, do not express GFP. However, since indigenous bacteria in the sewage community likely do not encodelacIq , which is a modified version oflacI , transconjugants express GFP constitutively. We used fluorescence-activated cell sorting (FACS) for high throughput identification of transconjugants in order to reveal the plasmids transfer host ranges (henceforth referred to as host range). Specifically, host range was investigated through post-mating sorting and 16S rRNA gene amplicon sequencing analysis of taxonomic and phylogenetic relationships of the transconjugant communities. Although the permissiveness towards broad host range plasmids of the microbial community in WWTPs have been assessed utilizing a similar strategy (Jacquiod et al., 2017; Li et al., 2018), the capability of the sewage transconjugant community to further disseminate plasmids to potential pathogens have not been investigated. Thus, subsequently, we assessed the donor potential of sorted transconjugants by further enriching this fraction in plasmid selective media to further track plasmid transfer to a model Enterobacteriaceae pathogen. In parallel, we performed the same experiments with pB10, a representative of the well-studied P1 group (Jacquiod et al., 2017; Klümper et al., 2015; Li et al., 2018; Musovic et al., 2014; Pinilla-Redondo et al., n.d.), which we hypothesized to represent the maximal host range grade VI (Redondo-Salvo et al., 2020).