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

Abstract

<jats:title>Abstract</jats:title><jats:p>Bacteriophages and their bacterial hosts are locked in a dynamic evolutionary arms race. Phage satellites, selfish genomic islands which exploit both host bacterium and target phage, further complicate the evolutionary fray. One such tripartite system involves the etiological agent of the diarrheal disease cholera – <jats:italic>Vibrio cholerae</jats:italic>, the predominant phage isolated from cholera patients – ICP1, and a phage satellite – PLE. When ICP1 infects <jats:italic>V. cholerae</jats:italic> harboring the integrated PLE genome, PLE accelerates host lysis, spreading the PLE while completely blocking phage production protecting <jats:italic>V. cholerae</jats:italic> at the population level. Here we identify a single PLE gene, <jats:italic>lidI</jats:italic>, sufficient to mediate accelerated lysis during ICP1 infection and demonstrate that LidI functions through disrupting lysis inhibition – an understudied outcome of phage infection when phages vastly outnumber their hosts. This work identifies ICP1-encoded holin and antiholin genes <jats:italic>teaA</jats:italic> and <jats:italic>arrA</jats:italic> respectively, that mediate this first example of lysis inhibition outside the T-even coliphages. Through <jats:underline>l</jats:underline>ysis <jats:underline>i</jats:underline>nhibition <jats:underline>d</jats:underline>isruption, LidI is sufficient to limit the number of progeny phage produced from an infection. Consequently, this disruption bottlenecks ICP1 evolution as probed by recombination and CRISPR-Cas targeting assays. These studies link novel characterization of the classic phenomenon of lysis inhibition with a conserved protein in a dominant phage satellite, highlighting the importance of lysis timing during infection and parasitization, as well as providing insight into the populations, relationships, and evolution of bacteria, phages, and phage satellites in nature.</jats:p><jats:sec><jats:title>Importance</jats:title><jats:p>With increasing awareness of microbiota impacting human health comes intensified examination of, not only bacteria and the bacteriophages that prey upon them, but also the mobile genetic elements (MGEs) that mediate interactions between them. Research is unveiling evolutionary strategies dependent on sensing the milieu: quorum sensing impacts phage infection, phage teamwork overcomes bacterial defenses, and abortive infections sacrifice single cells protecting populations. Yet, the first discovered environmental sensing by phages, known as lysis inhibition (LIN), has only been studied in the limited context of T-even coliphages. Here we characterize LIN in the etiological agent of the diarrheal disease cholera, <jats:italic>Vibrio cholerae</jats:italic>, infected by a phage ubiquitous in clinical samples. Further, we show that a specific MGE, the phage satellite PLE, collapses LIN with a conserved protein during its anti-phage program. The insights gleaned from this work add to our expanding understanding of microbial fitness in natural contexts beyond the canonical bacterial genome and into the realm of antagonistic evolution driven by phages and satellites.</jats:p></jats:sec>

A family of viral satellites manipulates invading virus gene expression and affects cholera toxin mobilization

“…We suspect that the timing of PLE gene cluster expression has evolved to take advantage of ICP1's own transcriptional program. Coordination between PLE gene expression and ICP1's gene expression would be consistent with PLEs' reliance on ICP1 gene products for key steps of the PLE lifecycle [22][23][24][25].…”

“…Few genes hit peak expression at 12 minutes, though an exception are a subset of nucleotide metabolism genes (within the range gp176-gp211). Finally, late genes with peak expression at 16 minutes post infection are comprised primarily of putative structural genes and genes known to be involved in lysis (Fig 2 and S3 Table), [25]. The lysis, capsid, and tail genes occur in three separate clusters, all encoded on the (-) strand (Fig 2 and S3 Table).…”

“…Surprisingly, the most abundant transcripts in the ICP1 and PLE transcriptomes are both predicted to be non-coding. The most abundantly expressed ICP1 transcript is encoded in an approximately 1kb orf-less space between gp139, the start of ICP1's lysis cluster [25], and gp140 (Fig 4). The length of this transcript (~800bp based on the RNA-seq coverage), is comparable to the Giant-Ornate-Lake-and Lactobacialles-Derived (GOLLD) and the Rumen-Originating, Ornate, Large (ROOL) RNAs that have been found in many phage genomes [47].…”

“…Successful PLE parasitism does not abrogate cell lysis, but results in the complete restriction of ICP1 and the release of PLE transducing particles [14] (S1 Fig). The tractable genetics of V. cholerae facilitates mechanistic studies of PLE gene products, and insights have been gained regarding chromosomal excision of PLE [22], PLE DNA replication [23,24], and PLE manipulation of lysis kinetics during infection [25]. Notably, ICP1 genome editing is also possible [26], allowing manipulation of both sides of this host-parasite relationship.…”

“…Our data show that PLEs exhibit conserved manipulation of their host-phage's gene expression, but divergent effects on CTXφ, revealing that PLEs can influence both their hosts' resistance to phage and the mobility of virulence encoding elements.So far, few insights have been gained into the gene expression programs of ICP1 and PLEs. PLE1 expresses its integrase in uninfected cells [22], expression of PLE1's replication initiator, RepA, is induced following infection of ICP1 [23], and the PLE's lysis modulator, LidI, is detectable by Western blot late during ICP1 infection [25]. The PLE integrase's recombination directionality factor (necessary for directing integrase excision activity) is PexA, an ICP1 protein whose native function is unknown, but whose expression can be detected by 5 minutes post infection [22].…”