The endosymbiotic bacterium Wolbachia induces resistance to dengue virus in Aedes aegypti

Abstract

Genetic strategies that reduce or block pathogen transmission by mosquitoes have been proposed as a means of augmenting current control measures to reduce the growing burden of vector-borne diseases. The endosymbiotic bacterium Wolbachia has long been promoted as a potential vehicle for introducing disease-resistance genes into mosquitoes, thereby making them refractory to the human pathogens they transmit. Given the large overlap in tissue distribution and intracellular localization between Wolbachia and dengue virus in mosquitoes, we conducted experiments to characterize their interactions. Our results show that Wolbachia inhibits viral replication and dissemination in the main dengue vector, Aedes aegypti. Moreover, the virus transmission potential of Wolbachia-infected Ae. aegypti was significantly diminished when compared to wild-type mosquitoes that did not harbor Wolbachia. At 14 days post-infection, Wolbachia completely blocked dengue transmission in at least 37.5% of Ae. aegypti mosquitoes. We also observed that this Wolbachia-mediated viral interference was associated with an elevated basal immunity and increased longevity in the mosquitoes. These results underscore the potential usefulness of Wolbachia-based control strategies for population replacement.

Figure 1. Inhibition of dengue infection in the mosquito midgut by Wolbachia.

At 3, 6, 9, 12, 15 and 18 days after a blood meal containing DENV-2, mosquito midguts were collected, and the number of genome copies of the DENV genome was determined by qRT-PCR using primers for the NS5 gene; the results were normalized to the Ae. aegypti ribosomal protein S7 (RPS6). Lines indicate the median of the five biological replicates. Significance was determined using a Mann-Whitney U test (P<0.05).

Figure 2. Inhibition of dengue dissemination to the mosquito thorax and head by Wolbachia.

A. At 3, 6, 9, 12, 15 and 18 days after a blood meal containing DENV-2, mosquito thoraces were collected, and genome copies of dengue virus were measured by qRT-PCR using primers for the NS5 gene and normalized with Ae. aegypti RPS6. Lines indicate the median of the five biological replicates. Significance was determined using a Mann-Whitney U test (P<0.05). B. The DENV-2 infection rate was determined by an indirect fluorescent antibody assay performed on head squashes of individual Waco and WB1 females of Ae. aegypti at 14 and 21 days (n = 36) or of Houston and HT1 females of Ae. albopictus at 14 days post-infection (n = 30). Data shown are means of three replicates for Waco and WB1 and two replicates for Houston and HT1, and the bars indicate standard error. There was a significant difference in the infection rate between the Waco and WB1 strains at both 14 and 21 days post-infection (Fisher's exact test, p<0.05).

Figure 3. In vitro assay of DENV-2 transmission by Waco and WB1 mosquitoes at 14 days post-infection.

After the wings and legs had been removed, the proboscis of each mosquito was inserted into a feeding solution for 90 min. Solutions from eight mosquitoes were pooled as one group and analyzed for infectious DENV-2 by plaque assays. The viral titers of each group of mosquitoes after feeding on the solutions were also analyzed in parallel. Lines indicate the log of the median values from five or eight biological replicates. Viral titers were significantly higher in the Waco strain than the WB1 strain for both the feeding solution and whole body (Mann-Whitney U test, P<0.05).

Figure 4. Wolbachia density in the ovaries, midguts and salivary glands of WB1 mosquitoes.

Total RNA was extracted from one ovary or midgut or a pool of three salivary glands using the RNAeasy kit. q-PCR was conducted using primers targeting the wAlbB-wsp gene. The Wolbachia genome copy number was normalized with Ae. aegypti RPS6. Two recombinant plasmids containing the targeted fragments were used to generate separate standard curves for wAlbB-wsp and RPS6. Ten biological replicates were used for each tissue. Error bars represent the standard error; different letters are significantly different (ANOVA, P<0.05).

Figure 5. Expression of selected immune genes induced by Wolbachia in Waco and WB1 mosquitoes.

qRT-PCR was performed using whole bodies of 4- or 5-day-old non-blood-fed females, with three biological replicates for each gene. Gene expression data were normalized with RPS7. The primer sequences have been reported previously . The bars indicate standard error.

Figure 6. Longevity of Waco and WB1 mosquitoes fed with blood mixed with or without DENV.

After mosquitoes were fed a blood meal, either containing DENV (A) or not (B), they were maintained in an incubator at 27°C and 85% humidity. The data shown are means of three replicates (25 for each), and the bars indicate standard error. Results from two independently reared cohorts are shown (cohort 1 [A]; cohort 2 [B]). The survival curves were significantly different between the Waco and WB1 mosquitoes fed with dengue-infected blood (logrank test, p<0.0001).