Wolbachia strains for disease control: ecological and evolutionary considerations

Abstract

Wolbachia are endosymbionts found in many insects with the potential to suppress vectorborne diseases, particularly through interfering with pathogen transmission. Wolbachia strains are highly variable in their effects on hosts, raising the issue of which attributes should be selected to ensure that the best strains are developed for disease control. This depends on their ability to suppress viral transmission, invade host populations, persist without loss of viral suppression and not interfere with other control strategies. The potential to achieve these objectives is likely to involve evolutionary constraints; viral suppression may be limited by the ability of infections to spread due to deleterious host fitness effects. However, there are exceptions to these patterns in both natural infections and in novel associations generated following interspecific transfer, suggesting that pathogen blockage, deleterious fitness effects and changes to reproductive biology might be at least partly decoupled to achieve ideal infection attributes. The stability of introduced Wolbachia and its effects on viral transmission remain unclear, but rapid evolutionary changes seem unlikely. Although deliberate transfers of Wolbachia across species remain particularly challenging, the availability of strains with desirable attributes should be expanded, taking advantage of the diversity available across thousands of strains in natural populations.

Keywords: Aedes; Wolbachia; deployment issues; disease control; fitness; strain attributes; virus.

Figure 1

The first two principal components explaining 80.6% of total variation among Wolbachia strains in natural and transinfected hosts for the levels of: cytoplasmic incompatibility/male killing, maternal transmission, fitness costs and blockage of RNA viruses. Empty symbols denote natural Wolbachia infections, and filled symbols denote transinfections. Each infection attribute is ordered as: 0 (no effect), 1 (low), 2 (medium/partial) or 3 (high/full effect). Fitness cost has an additional value of −1 for infection effects that are somewhat beneficial. Twenty-one data points summarize values extracted from Table 1. Overall effects in natural hosts seem different from those in transinfected hosts, and the effects are also virus-dependent. Colinearity between fitness cost and viral blockage suggest that there is a possible trade-off between these effects, such that strains with strong viral protection might be difficult to spread due to higher deleterious effects on the host. Exceptions to these patterns, ho wever, indicate that it may be possible to achieve a desirable combination of infection attributes, but more strains need to be tested.