Insights into RNAi-based antiviral immunity in Lepidoptera: acute and persistent infections in Bombyx mori and Trichoplusia ni cell lines

Abstract

The control of viral infections in insects is a current issue of major concern and RNA interference (RNAi) is considered the main antiviral immune response in this group of animals. Here we demonstrate that overexpression of key RNAi factors can help to protect insect cells against viral infections. In particular, we show that overexpression of Dicer2 and Argonaute2 in lepidopteran cells leads to improved defense against the acute infection of the Cricket Paralysis Virus (CrPV). We also demonstrate an important role of RNAi in the control of persistent viral infections, as the one caused by the Macula-like Latent Virus (MLV). Specifically, a direct interaction between Argonaute2 and virus-specific small RNAs is shown. Yet, while knocking down Dicer2 and Argonaute2 resulted in higher transcript levels of the persistently infecting MLV in the lepidopteran cells under investigation, overexpression of these proteins could not further reduce these levels. Taken together, our data provide deep insight into the RNAi-based interactions between insects and their viruses. In addition, our results suggest the potential use of an RNAi gain-of-function approach as an alternative strategy to obtain reduced viral-induced mortality in Lepidoptera, an insect order that encompasses multiple species of relevant economic value.

Figure 1

Overexpression of Dcr2 and Ago2 results in improved viability upon CrPV infection in High Five cells. Cells were transfected with overexpression constructs containing the entire ORF of Bm-dcr2 and Bm-ago2. A group transfected with the pEA-pac control vector was used as control. The transfected cells were infected with CrPV (MOI25), or treated with PBS, and the viability was assessed after 48 h. The graph depicts box-plots of the normalized cell viability, i.e. the viability of the CrPV-infected cells normalized to the viability of the PBS-treated cells, in percentage. Statistical analysis (Shapiro-Wilk normality test and Unpaired T-tests) was performed in GraphPad Prism 7 (****P < 0.0001; **P < 0.01; n = 6).

Figure 2

Differential regulation of dcr2 and ago2 upon CrPV infection in High Five cells. Cells were infected with CrPV (MOI10) or treated with PBS. The transcript levels of Tn-dcr2 (A) and Tn-ago2 (B) were measured 0, 2, 8 and 24 h post-infection. The graphs depict box-plots of the normalized relative transcript levels, i.e. the relative transcript levels of the CrPV-infected cells normalized to the relative transcript levels of the PBS-treated cells. Statistical analysis (Shapiro-Wilk normality test and Unpaired T-tests) was performed in GraphPad Prism 7 (*P < 0.05; **P < 0.01; ***P < 0.005; ****P < 0.0001; n = 4).

Figure 3

Overexpression of Dcr2 and Ago2 does not induce changes in the mlv transcript levels in High Five cells. Cells were transfected with overexpression constructs containing the entire ORF of Bm-dcr2 and Bm-ago2. A group transfected with the pEA-pac control vector was used as control. The mlv relative transcript levels were assessed 1, 2 and 3 days after the transfection, respectively. The graphs depict box-plots of mlv relative transcript levels. Statistical analysis (Shapiro-Wilk normality test and Unpaired T-tests) was performed in GraphPad Prism 7 (n = 4).

Figure 4

Virus-specific siRNAs are bound to Ago2 in BmN cells. (A) Size distribution of the Ago2-bound viral reads (Sf-RV, Bm-MLV and Bm-IV), (B) Ago2-bound siRNAs map to the genome of Sf-RV, Bm-MLV and Bm-IV. (black: sense reads; grey: anti-sense reads).

Figure 5

Knockdown of dcr2 and ago2 results in increased mlv transcript levels in lepidopteran cells. High Five cells were transfected with dsdcr2 (dcr2 Knockdown), dsago2 (ago2 Knockdown) or dsluc (Control). The graphs depict box-plots of mlv relative transcript levels. Statistical analysis (Shapiro-Wilk normality test and Mann–Whitney test) was performed in GraphPad Prism 7 (*P < 0.05; n = 4).