Explorando o viroma de mosquitos vetores: uma abordagem metagenômica utilizando pequenos RNAs da resposta imune do hospedeiro

Abstract

Mosquitoes are the primary vectors of arboviruses worldwide. The recent application of high-throughput nucleic acid sequencing techniques coupled with metagenomics has enabled the discovery of hundreds of viruses in mosquitoes. Most of these viruses are classified as insect-specific, incapable of infecting vertebrate hosts. Several studies have demonstrated the ability of these viruses to impact the vector competence of mosquitoes for the transmission of medically important arboviruses. Despite the advances provided by metagenomics in uncovering new viruses in eukaryotes, three major computational challenges are associated with this approach when applied in large-scale: associating viral sequences with segmented genomes of the same virus; distinguishing between exogenous viral sequences and endogenous viral elements; and the lack of sensitivity in detecting highly divergent viral sequences due to the dependence on sequence similarity-based techniques. Our group proposes a metagenomic strategy using small RNAs from host immune responses to address these challenges in viral metagenomics. In this study, we analyzed 122 novel small RNA libraries generated from samples of vector mosquitoes from 10 species collected on four continents. We identified 28 viruses belonging to more than nine viral families and 1726 endogenous viral sequences derived from 115 viral species across 24 viral families. We identified and characterized a viral segment not detected by sequence similarity methods, belonging to a Narnavirus, providing a clear example of the dark matter of viral metagenomics. We developed the Small RNA Metavir tool, an automated pipeline for metagenomic analysis with small RNAs that enables the distinction between exogenous and endogenous viral sequences using machine learning classifiers trained in this study. In addition to the precise detection of viral sequences, we demonstrated that our strategy can detect RNAs from Wolbachia, a crucial endosymbiont that affects the vector competence of mosquitoes. We showed that the load of Wolbachia small RNAs has a negative correlation with the load of insect-specific viruses in mosquitoes from Niterói-RJ, where this bacterium is used for the control of arbovirus transmission by Aedes aegypti. In this thesis, we obtained results and developed bioinformatics approaches that allow for the precise characterization of the virome of vector mosquitoes and its analysis in light of biological components relevant to mosquito vector competence. The results contribute to advancements in understanding the intricate relationships between mosquitoes and viruses, essential for the development of new strategies for the biological control of these pathogen vectors.