Converging lines of evidence fail to detect nanovirus replication in its insect vector

Nanoviruses have a multipartite circular ssDNA genome and are transmitted by aphids in a circulative manner. Although generally considered non-propagative within their vectors, direct experimental evidence supporting the absence of viral replication remains limited. This question is particularly relevant for multipartite viruses because replication within vectors could theoretically help maintain genome integrity during transmission, decreasing dilution effects and the probability of genome segment loss. Here, we investigated whether faba bean necrotic stunt virus (FBNSV, genus Nanovirus, family Nanoviridae) replicates in aphids using several independent approaches. Viral load dynamics were quantitative PCR-monitored after virus acquisition, the replication-associated protein M-Rep was searched for within gut and salivary gland cells by immunofluorescence, viral DNA localization and strand specificity were analysed using HCR-DNA-FISH (hybridization chain reaction fluorescence in situ hybridization) and confocal microscopy, viral transcription was investigated by Illumina mRNA sequencing, and viral DNA replication was analysed by Nanopore sequencing. All approaches converged towards the same conclusion. Viral loads did not increase after acquisition, M-Rep was undetectable in aphid tissues, viral DNA accumulated outside nuclei, no negative-strand viral DNA was detected, and only negligible quantities of viral transcripts and dsDNA molecules were detected in RNA-seq and Nanopore-seq datasets from viruliferous aphids, although these molecules were readily detected in FBNSV-infected plants used as the acquisition source. While none of these approaches could alone suffice to definitely exclude a weak, localized and/or transient replication, they together strongly support the conclusion that FBNSV is transmitted in a genuinely circulative non-propagative manner.