CRISPR-Cas systems in Enterococcus faecalis and their application in probiotics for the removal of antibiotic resistance

The gut microbiome is composed of diverse bacterial, fungal and archaeal species which function dynamically and reside commensally within a host, proving health benefits such as immune stimulation, nutrient supplementation, and resistance to pathogen colonization. Enterococcus faecalis is a native, commensal inhabitant of the GI tract of most terrestrial animals. It is typically an underrepresented member of the healthy microbiome with its prevalence impeded by competing organisms. However, during antibiotic-mediated dysbiosis, its population blooms. They are then able to reach the bloodstream and cause infections such as endocarditis and bacteremia. E. faecalis infections are challenging to treat due to their intrinsic resistance to certain antibiotics and propensity for acquiring antibiotic resistance genes through horizontal gene transfer. Sequence analysis of clinical isolates has revealed that these strains possess expanded genomes of which >25% is derived from mobile genetic elements. It has previously been observed that MDR strains of E. faecalis lack complete CRISPR-Cas systems, an adaptive immune system which protects bacteria from invading DNA. These systems are able to recognize and cleave specific sequences of DNA by using RNA guides and have found many applications as genetic tools in the manipulation of DNA or the rational design of sequence-specific antimicrobials.
In this dissertation, I inserted the interference machinery of a CRISPR-Cas system in E. faecalis into an MDR strain and restored activity for genome defense using conjugation assays. I noted that CRISPR-mediated defense was not entirely effective in this species and a significant number of transconjugants were obtained even when the plasmid was targeted. Further examination of these transconjugants showed that they were unstable, and depending on the presence or absence of selection for the plasmid, the cells were able to either compromise their CRISPR system or lose the targeted plasmid. More importantly, this instability conferred a growth defect which could then be exploited in composite populations to selectively eliminate undesirable traits. Using this, we were able to target an antibiotic resistance gene and abolish resistance from heterogeneous populations of E. faecalis. Following this discovery, I improved the system by incorporating the entire CRISPR-Cas targeting system on a pheromone-responsive plasmid (PRP) encoding a bacteriocin which enforced its selection. PRPs have notoriously high conjugation frequencies and are known to efficiently disseminate in E. faecalis populations in both in vivo and in vitro conditions. Using these plasmids, I was able to significantly decrease antibiotic resistance from in vitro populations and in an in vivo model of mouse gut colonization. The work presented here provides evidence supporting the use of CRISPR-targeting constructs in probiotics to reduce the circulation of undesirable traits among E. faecalis strains colonizing patients in hospitals with the aim of mitigating the occurrence of MDR infections.