Broad RNA interference-mediated antiviral immunity and virus-specific inducible responses in Drosophila

Abstract

The fruit fly Drosophila melanogaster is a good model to unravel the molecular mechanisms of innate immunity and has led to some important discoveries about the sensing and signaling of microbial infections. The response of Drosophila to virus infections remains poorly characterized and appears to involve two facets. On the one hand, RNA interference involves the recognition and processing of dsRNA into small interfering RNAs by the host RNase Dicer-2 (Dcr-2), whereas, on the other hand, an inducible response controlled by the evolutionarily conserved JAK-STAT pathway contributes to the antiviral host defense. To clarify the contribution of the small interfering RNA and JAK-STAT pathways to the control of viral infections, we have compared the resistance of flies wild-type and mutant for Dcr-2 or the JAK kinase Hopscotch to infections by seven RNA or DNA viruses belonging to different families. Our results reveal a unique susceptibility of hop mutant flies to infection by Drosophila C virus and cricket paralysis virus, two members of the Dicistroviridae family, which contrasts with the susceptibility of Dcr-2 mutant flies to many viruses, including the DNA virus invertebrate iridescent virus 6. Genome-wide microarray analysis confirmed that different sets of genes were induced following infection by Drosophila C virus or by two unrelated RNA viruses, Flock House virus and Sindbis virus. Overall, our data reveal that RNA interference is an efficient antiviral mechanism, operating against a large range of viruses, including a DNA virus. By contrast, the antiviral contribution of the JAK-STAT pathway appears to be virus specific.

FIGURE 1. Dcr-2 is involved in host-defense against the DNA virus IIV6

(A) Upon injection of IIV6 (5000PFU) in wild-type (yw) and Dcr-2^R416X mutant flies, typical blue paracrystalline structures appeared earlier in the abdomen (arrow head) of the mutant flies. Representative individuals 10 dpi are shown. (B) Groups of 20 wild-type (yw) or Dcr 2^R416X mutant flies were injected with IIV6 or Tris and survival was monitored daily. The difference between the wild-type and Dcr-2 mutant flies is statistically significant. (C) Viral titer in groups of 5 wild-type (yw) or Dcr-2^R416X mutant flies was monitored 10 dpi. (D) Rescue of the hemizygous Dcr-2^L811fsx for the IIV6 susceptibility phenotype by a transposon expressing a wild-type Dcr-2 transgene. Dcr-2^L811fsx hemizygous flies (Dcr-2^L811fsx/Df) are significantly more susceptible than Dcr-2^L811fsx hemizygous flies complemented by a wild-type Dcr-2 transgene (Dcr-2^L811fsx/Df rescue). Df is Df(2R)BSC45, a deficiency that fully uncovers the Dcr-2 locus. All control and genomic rescued flies are in CantonS background. (E) Survival rate of wild-type (yw), R2D2¹ and AGO2⁴¹⁴ mutant flies upon IIV6 or Tris injection. (F) IIV6 DNA load was determined by qPCR in 4 groups of 6 flies of the indicated genotype at 10 dpi. For all panels, the data represent the mean and standard deviation of at least three independent experiments and the difference between controls and mutant flies is statistically significant (* p<0.05, ** p<0.01, *** p<0.001). All experiments are performed at 22°C (A, C, F) or 25°C (B, D and E).

FIGURE 2. Virus-derived siRNAs in S2 cells and Drosophila adult flies infected by the DNA virus IIV6

RNA was extracted 5 days post-infection (dpi) from S2 cells infected by IIV6 (MOI 0.01) and adult wild-type (yw) or mutant (Dcr-2^R416X) flies injected with IIV6 (5000 PFU/fly). (A) Size distribution of the small RNAs matching the viral genome in S2 cells and adult flies of the indicated genotype. (B, C) Distribution of the 21nt siRNAs from the S2 cells (B) and yw adult flies (C) libraries along the IIV6 genome. Each IIV6-derived small RNA (vsiRNA) is represented by the position of its first nucleotide. The vsiRNAs matching the upper and lower strand of the DNA genome are respectively shown above (positive reads number) and below (negative reads number) the horizontal axis, which represents the genomic coordinates. In panel B, the number of reads for 4 peaks going off-scale is indicated next to them in italics. (D) Strand-specific RT-PCR with primers corresponding to the annotated viral genes 206R, 224L, 244L and 261R. The experiment was performed in the presence (+) or absence (−) of reverse transcriptase (RT). NI: non infected; F and R: forward and reverse strand primer used for reverse transcription.

FIGURE 3. The JAK kinase Hopscotch is involved in host-defense against DCV and CrPV

(A, B) Groups of 20 wild-type (OR) or hopscotch (hop^M38/msv1) mutant flies were injected with DCV (500 PFU) (A) or CrPV (5 PFU) (B) and survival was monitored daily. The experiment was repeated three times and the data represent the mean and standard deviation. The difference between wild-type and hop mutant flies is statistically significant (*** p<0.001). In the right panel, viral titer was determined in groups of 5 flies 2 dpi for DCV (A) and 1dpi for CrPV (B). The data represent the mean and standard deviation of three independent experiments and the difference between wild-type and hop mutant flies is statistically significant (* p<0.05, ** p<0.01). (C) DCV and CrPV infection trigger induction of the genes upd2 and upd3, which encode cytokines activating the JAK/STAT pathway. Flies were infected with DCV or CrPV and expression of upd, upd2 and -3 was monitored in groups of 10 flies at the indicated time points by Taqman Q-PCR. The results of at least two independent experiments are shown.

FIGURE 4. Susceptibility of fly mutants for the JAK kinase Hopscotch to infection by SINV, VSV, FHV, DXV and IIV6

Groups of 20 wild-type (OR) or hop mutant flies were injected with SINV (A), VSV (B), FHV (C), DXV (D) or IIV6 (E), and survival was monitored. For VSV and SINV the Tris buffer control injection is also shown, since hop mutant flies showed decreased survival at 29°C after day 16 upon both buffer and virus injection. Kaplan-Meier analysis of the results of at least two independent experiments reveal a statistically significant difference in survival between wild-type and hop mutant flies only in the case of DXV (* p<0.05). ns: not significant.

FIGURE 5. Microarray analysis of gene induction following infection by DCV, FHV or SINV

(A) Venn diagram showing the number of upregulated genes (by a factor of at least 2) following infection by the three viruses. The total number of genes regulated by each virus is indicated in parenthesis. (B) FHV and SINV induce members of the same gene families, but FHV triggers a stronger response. The number of genes belonging to seven gene ontology (GO) functional categories induced by both FHV and SINV or by FHV only are shown. (C) Expression of vir-1 and TotM by Q-PCR normalized for the expression of the housekeeping gene RpL32. Groups of 10 wild-type (OR) flies were injected with Tris buffer or the viruses DCV, CrPV, FHV, SINV, VSV, DXV, IIV6 or pricked with a needle dipped in a concentrated pellet of the Gram-positive bacteria M. luteus and the Gram-negative bacteria E. coli. RNA was extracted at 6h, 1d, 2d, 3d and 4d after challenge. The data represent the mean and the standard errors of at least two independent experiments. p-values were calculated for each time point individually versus the Tris injected control. *p<0.05, **p<0.01, ***p<0.001.