Differential Small RNA Responses against Co-Infecting Insect-Specific Viruses in Aedes albopictus Mosquitoes

Abstract

The mosquito antiviral response has mainly been studied in the context of arthropod-borne virus (arbovirus) infection in female mosquitoes. However, in nature, both female and male mosquitoes are frequently infected with insect-specific viruses (ISVs). ISVs are capable of infecting the reproductive organs of both sexes and are primarily maintained by vertical transmission. Since the RNA interference (RNAi)-mediated antiviral response plays an important antiviral role in mosquitoes, ISVs constitute a relevant model to study sex-dependent antiviral responses. Using a naturally generated viral stock containing three distinct ISVs, Aedes flavivirus (AEFV), Menghai rhabdovirus (MERV), and Shinobi tetra virus (SHTV), we infected adult Aedes albopictus females and males and generated small RNA libraries from ovaries, testes, and the remainder of the body. Overall, both female and male mosquitoes showed unique small RNA profiles to each co-infecting ISV regardless of the sex or tissue tested. While all three ISVs generated virus-derived siRNAs, only MERV generated virus-derived piRNAs. We also studied the expression of PIWI genes in reproductive tissues and carcasses. In contrast to Piwi5-9, Piwi1-4 were abundantly expressed in ovaries and testes, suggesting that Piwi5-9 are involved in exogenous viral piRNA production. Together, our results show that ISV-infected Aedes albopictus produce viral small RNAs in a virus-specific manner and that male mosquitoes mount a similar small RNA-mediated antiviral response to that of females.

Keywords: Aedes albopictus; PIWI-interacting RNA; co-infection; insect-specific viruses; reproductive tissues; sex difference; small interfering RNA.

Figure 1

AEFV, MERV and SHTV infections in reproductive tissues (ovaries and testes) and carcasses. Ae. albopictus adult females and males were intrathoracically injected with an AEFV/MERV/SHTV mixture and collected at 8 days post-injection. RNA levels for each ISV in the co-infected mosquito ovaries, testes or carcasses were quantified by RT-qPCR normalized with actin (A) or RPL18 RNA (B). Each dot represents a pool of 20 reproductive tissues or carcasses.

Figure 2

Comparison of vsiRNA and putative vpiRNA production from ISVs in reproductive tissues and carcasses of female or male Ae. albopictus mosquitoes. The size distribution of total small RNA reads from each sample is shown in (A). Normalized siRNA reads per one million (RPM) mapped reads to the AEFV, MERV or SHTV genome are shown in (B). Proportion of 26-30 nt AEFV-, MERV- or SHTV-derived small RNAs per 100 vsiRNAs is represented in (C). Yellow and blue bars represent positive- and negative-stranded reads, respectively.

Figure 3

Small RNA responses to AEFV in female and male Ae. albopictus reproductive tissues and carcasses. The distribution of siRNA (21 nt) (A) and piRNA-like small RNAs (26-30 nt) (upper panel, B) mapped to the AEFV genome. A schematic illustration of the AEFV genome is shown on top of small RNAs mapping to the viral sequence. Yellow and blue bars represent positive- and negative-stranded reads, respectively. Regions with no coverage are indicated by gray bars. For AEFV-derived 26-30 nt piRNA-like small RNAs, relative nucleotide frequency at each position is shown as a heat map (lower panel in B) in which the color intensity denotes the frequency. The black and red arrowheads point to the 1U and 10A positions, respectively.

Figure 4

Small RNA responses to SHTV in female and male Ae. albopictus reproductive tissues and carcasses. The distribution of siRNA (21 nt) (A) and piRNA-like small RNAs (26-30 nt) (upper panel, B) mapped to the SHTV genome. A schematic illustration of the SHTV genome is shown on top of small RNAs mapping to the viral sequence. Yellow and blue bars represent positive- and negative-stranded reads, respectively. Regions with no coverage are indicated by gray lines. For 26-30 nt SHTV-derived piRNA-like small RNAs, relative nucleotide frequency at each position is shown as a heat map (lower panel in B) in which the color intensity denotes the frequency. 26-30 nt small RNA reads were not detected in ovary samples (B). The black and red arrowheads point to the 1U and 10A positions, respectively.

Figure 5

Small RNA responses to MERV in female and male Ae. albopictus reproductive tissues and carcasses. The distribution of siRNA (21 nt) (A) and piRNA-like small RNAs (26-30 nt) (upper panel, B) mapped to the MERV genome. A schematic illustration of the MERV genome is shown on top of small RNAs mapping to the viral sequence. Yellow and blue bars represent positive- and negative-stranded reads, respectively. Regions with no coverage are indicated by gray lines. For 26-30 nt MERV-derived piRNA-like small RNAs, relative nucleotide frequency at each position is shown as a heat map (lower panel in B), in which the color intensity denotes the frequency. The black and red arrowheads point to the 1U and 10A positions, respectively.

Figure 6

Relative gene expression of siRNA and piRNA pathway components in reproductive tissues and carcasses of Ae. albopictus females and males co-infected with AEFV, MERV, and SHTV. Relative gene expression of the PIWI genes Piwi1-9 and Ago3, together with DCR2 and Ago2, key components of the siRNA pathway, was examined by RT-qPCR. Relative gene expression was calculated by normalization with actin (A) or RPL18 RNA (B). Multiple unpaired t-test with Holm-Sidak correction was applied to calculate the statistical significance. The P values are summarized in Tables S2 and S3.