Symbionts commonly provide broad spectrum resistance to viruses in insects: a comparative analysis of Wolbachia strains

Abstract

In the last decade, bacterial symbionts have been shown to play an important role in protecting hosts against pathogens. Wolbachia, a widespread symbiont in arthropods, can protect Drosophila and mosquito species against viral infections. We have investigated antiviral protection in 19 Wolbachia strains originating from 16 Drosophila species after transfer into the same genotype of Drosophila simulans. We found that approximately half of the strains protected against two RNA viruses. Given that 40% of terrestrial arthropod species are estimated to harbour Wolbachia, as many as a fifth of all arthropods species may benefit from Wolbachia-mediated protection. The level of protection against two distantly related RNA viruses--DCV and FHV--was strongly genetically correlated, which suggests that there is a single mechanism of protection with broad specificity. Furthermore, Wolbachia is making flies resistant to viruses, as increases in survival can be largely explained by reductions in viral titer. Variation in the level of antiviral protection provided by different Wolbachia strains is strongly genetically correlated to the density of the bacteria strains in host tissues. We found no support for two previously proposed mechanisms of Wolbachia-mediated protection--activation of the immune system and upregulation of the methyltransferase Dnmt2. The large variation in Wolbachia's antiviral properties highlights the need to carefully select Wolbachia strains introduced into mosquito populations to prevent the transmission of arboviruses.

Figure 1. Phylogeny of Wolbachia strains and respective level of protection and within-host density.

(A) The phylogeny is based on the sequence of the MLST genes 16S rRNA, aspC, atpD, ftsZ, sucB, groEL, coxA and fbpA. Branch labels represent posterior support values. Nodes with less than 50% support were collapsed. The scale bar indicates time in coalescent units. (B–C) Flies were either infected with (B) DCV or (C) FHV. Survival is expressed as the negative natural log of the hazard ratio compared to Wolbachia-free flies, as estimated from a Cox's mixed-effect model. Error bars are standard errors. Symbols above the bars give the significance relative to the Wolbachia-free controls (*: P<0.05; **: P<0.01; ***: P<0.001). (D) Wolbachia density is expressed as the ratio of Wolbachia genomic DNA to Drosophila genomic DNA, as estimated by quantitative PCR. Different letters indicate significant differences based on a Tukey's honest significance test on ln-transformed data.

Figure 2. Survival of flies carrying different Wolbachia strains or being Wolbachia-free.

Flies were either infected with (A) DCV, (B) FHV or (C) mock-infected with Ringer's solution.

Figure 3. Correlation between protection, viral titers and Wolbachia density.

Dots indicate mean value of the traits for each Wolbachia strain. Error bars are standard errors. Solid lines show predicted values from linear regressions for illustrative purposes. r_g is the genetic correlation between traits. (A) Correlation of survival between DCV- and FHV-infected flies (negative natural log of hazard ratios). (B) Correlation between DCV and FHV titers. (C–D) Correlation between viral titer and survival following (C) DCV infection or (D) FHV infection. Viral titers were estimated as viral RNA concentrations relative to the Drosophila gene EF1α100E. (E–F) Relationship between Wolbachia density and survival in (E) DCV- and (F) FHV-infected flies. Wolbachia density was estimated as the ratio between copy numbers ofthe Wolbachia gene atpD and the Drosophila gene Actin 5C.

Figure 4. Effect of Wolbachia strains on viral titers.

(A–B) Relative viral titer in (A) DCV- and (B) FHV-infected flies. Relative titers are normalised by the mean titer of Wolbachia-free controls (uninfected). Error bars are standard errors. Symbols above the bars give the significance relative to the Wolbachia-free controls based on a Dunnett's test (*: P<0.05; **: P<0.01; ***: P<0.001).