Bacteriophage control of bacterial virulence.

In 1930, Felix d’Herelle wrote “. . .the actions and reactions are not solely between these two beings, man and bacterium, for the bacteriophage also intervenes; a third living being and, hence, a third variable is introduced” (19). The contribution of bacteriophages to the pathogenicity of their bacterial hosts began to be uncovered as early as 1927, when Frobisher and Brown discovered that nontoxigenic streptococci exposed to filtered supernatants of toxigenic streptococcal cultures acquired the ability to produce scarlatinal toxin (28). We now know that these supernatants contained a bacteriophage encoding the scarlatinal toxin and that these investigators were describing transduction, i.e., the transfer of genetic material to a bacterial cell via phage infection. Although these early investigators lacked a mechanistic explanation for their observations, they postulated that the bacterium was “within certain limits, perhaps a matter of secondary importance and that toxicogenicity might be a property that could be acquired by different types of organisms.” (28). Their hypothesis that bacteria acquire virulence properties has since gained widespread acceptance, as many virulence genes have been shown to undergo transfer among bacteria by phages (via transduction) and other mobile genetic elements such as plasmids (via conjugation). Over time, a number of toxin genes were found to be phage encoded, and consideration of the role of phages in bacterial pathogenesis emphasized the dissemination of toxin genes among bacterial strains (10). However, it has become increasingly clear that toxin genes are only a subset of the diverse virulence factors encoded by bacteriophages. For example, some phages encode regulatory factors that increase expression of virulence genes not encoded by the phage (84), while others encode enzymes that alter bacterial components related to virulence (31, 58). Furthermore, phages have unique properties that enable them to contribute more directly to bacterial virulence than via transduction. Structural components of virion particles, for example, may be directly pathogenic (5, 6, 99). Additionally, phage-encoded genes frequently undergo replication and transcriptional activation following prophage induction, a process that was speculated to have a role in the production of diphtheria toxin by Corynebacterium diphtheriae as early as 1960 (3). Since d’Herelle’s time, his notion of the phage as a third variable in bacterial pathogenesis has proven correct. However, while d’Herelle emphasized the diminution of bacterial virulence by bacteriophages (19), we have instead come to learn that phages serve as a driving force in bacterial pathogenesis, acting not only in the evolution of bacterial pathogens through gene transfer, but also contributing directly to bacterial pathogenesis at the time of infection. This review provides a discussion of (i) the discovery of phage-encoded virulence factors, (ii) bacterial virulence properties altered by phages, (iii) regulation of phage-encoded virulence factors, and (iv) the role of in situ prophage induction in the control of bacterial virulence.