Competition between mobile genetic elements drives optimization of a phage-encoded CRISPR-Cas system: insights from a natural arms race

Abstract

CRISPR-Cas systems function as adaptive immune systems by acquiring nucleotide sequences called spacers that mediate sequence-specific defence against competitors. Uniquely, the phage ICP1 encodes a Type I-F CRISPR-Cas system that is deployed to target and overcome PLE, a mobile genetic element with anti-phage activity in Vibrio cholerae . Here, we exploit the arms race between ICP1 and PLE to examine spacer acquisition and interference under laboratory conditions to reconcile findings from wild populations. Natural ICP1 isolates encode multiple spacers directed against PLE, but we find that single spacers do not interfere equally with PLE mobilization. High-throughput sequencing to assay spacer acquisition reveals that ICP1 can also acquire spacers that target the V. cholerae chromosome. We find that targeting the V. cholerae chromosome proximal to PLE is sufficient to block PLE and is dependent on Cas2-3 helicase activity. We propose a model in which indirect chromosomal spacers are able to circumvent PLE by Cas2-3-mediated processive degradation of the V. cholerae chromosome before PLE mobilization. Generally, laboratory-acquired spacers are much more diverse than the subset of spacers maintained by ICP1 in nature, showing how evolutionary pressures can constrain CRISPR-Cas targeting in ways that are often not appreciated through in vitro analyses. This article is part of a discussion meeting issue ‘The ecology and evolution of prokaryotic CRISPR-Cas adaptive immune systems’.

DominantVibrio choleraephage exhibits lysis inhibition sensitive to disruption by a defensive phage satellite

“…With these evolutionary hypotheses in mind, ICP1 acquiring the CRISPR-Cas system has changed the game -single spacers encoded in ICP1 and targeting PLE can allow ICP1 to replicate while simultaneously allowing for transduction of PLE (24). If things stopped here and all spacers were singular and created equal, selection could drive PLE to act more like a typical satellite phage, embracing horizontal transfer and allowing ICP1 to slide by producing at least some progeny phage.…”

“…Thus far, all characterization of ICP1 LIN was completed in PLE (-) V. cholerae. In clinical samples of V. cholerae, an increasing percentage of collected isolates from Bangladesh harbor PLE (21). PLE is a phage satellite, and by definition, a parasite of ICP1; although the mechanisms that PLE deploys to inhibit and hijack ICP1 are not completely understood, the lysis kinetics for high MOI ICP1 infections of V. cholerae vary depending on the presence of PLE.…”

“…While exposure to the CRISPR-Cas system was meant to probe evolution at the level of individual random mutations and proclivity to overcome targeting, phages also readily use homologous recombination to evolve during co-infection. To test the impact of LidI PLE 1 on this aspect of evolvability, two variants of ICP1, one devoid of spacers against ICP1 with an inactive Cas1 preventing spacer acquisition (CRISPR*-Cas ICP1) (21) and the other lacking Cas2-3 (CRISPR-Cas* ICP1), were used to coinfect V. cholerae strains with and without LidI PLE 1 . After lysis, progeny phage were tested for their ability to plaque on PLE 1 V. cholerae, which is only possible when homologous recombination between the two variants of ICP1 occurred restoring a functional CRISPR-Cas system able to target PLE 1 (Fig.…”

“…The predominant phage in cholera patient samples is ICP1, a lytic myovirus (19). This phage and its host have been locked in a dynamic arms race with no clear winner as both V. cholerae and ICP1 continue to be isolated from patients in the cholera endemic region of Bangladesh (19)(20)(21)(22). Added into the evolutionary fray is a phage satellite called the PLE (phage-inducible chromosomal island-like element), which is found in the chromosome of V. cholerae clinical isolates (23,24).…”

“…As no infectious ICP1 are produced, PLE activity ultimately defends populations of V. cholerae from ICP1 attack, functionally acting as an abortive infection system. In the face of this assault, ICP1 has acquired a functional Type I-F CRISPR-Cas adaptive immune system to target PLE in a sequence-specific manner, restoring ICP1 progeny phage production and overcoming PLE (24,27). While the full extent of ICP1, PLE, and V. cholerae interactions are unknown and continuing to evolve, efforts to understand how the PLE completely restricts ICP1 production have yet to identify any single PLE-encoded gene product that is necessary for inhibition, suggesting that multiple inhibitory strategies act synergistically to fully eliminate phage production (22,23,26).…”