Variation in bacterial pathotype is consistent with the sit-and-wait hypothesis

Abstract

The sit-and-wait hypothesis predicts that bacteria can become more virulent when they survive and transmit outside of their hosts due to circumventing the costs of host mortality. While this hypothesis is largely supported theoretically and through comparative analysis, experimental validation is limited. Here we test this hypothesis in Streptococcus suis, an opportunistic zoonotic pig pathogen, where a pathogenic ecotype proliferated during the change to intensive pig farming that amplifies opportunities for fomite transmission. We show in an in vitro environmental survival experiment that pathogenic ecotypes survive for longer than commensal ecotypes, despite similar rates of decline. The presence of a polysaccharide capsule has no consistent effect on survival. Our findings suggest that extended survival in the food chain may augment the zoonotic capability of S. suis. Moreover, eliminating the long-term environmental survival of bacteria could be a strategy that will both enhance infection control and curtail the evolution of virulence.

Keywords: Pharaoh’s curse; Streptococcus suis; environmental survival; evolution of virulence; indirect transmission; resource allocation hypothesis; zoonosis.

Fig. 1.. Prediction of the sit-and-wait hypothesis. (a) Direct transmission of a host-associated microorganism relies on a relatively healthy host to transmit, selecting against very high virulence. (b) Indirect transmission could allow for higher virulence, because the fitness costs associated with host ill-health and immobility are relaxed. (c) This leads to a predicted positive correlation between virulence and environmental survival.

Fig. 2.. Pathogens have longer survival times than commensals but no consistent difference in rates of population decline. Box plots showing (a) survival time (last day alive) and (b) rate of population decline for a panel of commensal and pathogenic ecotypes of S. suis. Boxes and whiskers show first and third quartile and points show outliers. Rate of decline was estimated by Poisson regression, which did not converge for two commensal isolates (see Table S2).