RNAi-based adaptive immunity in Aedes aegypti against arboviruses

Abstract

Arthropod-borne viruses (arboviruses) represent an ongoing threat to human and animal health. Mosquitoes, including Aedes aegypti, are important vectors of arboviruses from diverse viral families, such as dengue virus (DENV), Rift Valley fever virus (RVFV) and chikungunya virus (CHIKV). During infection, mosquitoes mount antiviral immune responses that can influence viral replication and transmission. Differences in these responses may contribute to variation in vector competence, affecting the ability of arboviruses to establish infection, disseminate to secondary tissues and ultimately be transmitted to vertebrate hosts.

RNA interference (RNAi) represents a major antiviral defence mechanism in mosquitoes, particularly through the exogenous small interfering RNA (exo-siRNA) pathway. Although the general mechanism of this pathway is well characterised, the contributions of individual domains of its key protein, Dicer 2 (Dcr2), remain incompletely understood. To investigate the roles of the helicase, PAZ and RNase III domains, conserved amino acids were identified through multiple sequence alignments and targeted mutations were introduced. Previous investigations demonstrated the importance of the helicase and RNase III domains for antiviral activity, RNA silencing and virus-derived small interfering RNA (vsiRNA) production during Semliki Forest virus (SFV) infection. In this study, PAZ domain mutants expressed in Ae. aegypti Dcr2 knockout cells exhibited reduced antiviral activity against SFV compared with wild-type Dcr2, while retaining intermediate silencing activity. Furthermore, the characteristic 21-nucleotide vsiRNA population broadened to 20–22 nucleotides, highlighting the importance of the PAZ domain in accurate vsiRNA processing.

To determine whether these domain-specific effects extend to arboviruses with different genome organisations, the Dcr2 mutants were further investigated during infection with RVFV and Bunyamwera orthobunyavirus (BUNV), both possessing tripartite negative-sense RNA genomes. All three investigated Dcr2 domains contributed to antiviral activity against both viruses. Notably, one helicase mutant retained antiviral activity against RVFV comparable to wild-type Dcr2 despite producing RVFV-derived small RNAs at levels comparable to Dcr2-deficient controls. This finding suggests that antiviral activity and detectable viral small RNA production may not always correlate directly, warranting further investigation into the underlying mechanism.

A second focus of this thesis was the potential contribution of virus-derived DNA (vDNA) and non-retroviral integrated RNA virus sequences (NIRVS) to antiviral immunity and persistent arboviral infections. To investigate possible effects of previous viral exposure, mosquito cells were primed through arboviral infection followed by viral clearance and subsequently challenged with either the original or a different arbovirus. Antiviral responses varied depending on the priming virus and individual cell clone. In particular, BUNV-primed cells exhibited reduced SFV replication, with the effect becoming more pronounced at higher passage numbers, suggesting a possible time-dependent component of antiviral priming. Clone-specific differences were also observed in small RNA production. However, the contribution of potential NIRVS integration, the priming procedure itself and other virus- or host-specific factors remains unresolved. Further investigations, including whole-genome sequencing and transcriptomic analyses, are required to elucidate these mechanisms.

Together, these findings demonstrate the importance of individual Dcr2 domains in antiviral responses against arboviruses from distinct viral families, particularly highlighting that mutations in the helicase and PAZ domains can impair antiviral activity despite the presence of a functional RNase III domain. Furthermore, they reveal the complexity of antiviral responses following previous arboviral exposure, including virus-, clone- and passage-dependent effects. This work provides new insights into mosquito–arbovirus interactions and identifies avenues for further investigation into antiviral immunity, viral priming and their potential influence on vector competence.