N6-methyladenosine modification of the Aedes aegypti transcriptome and its alteration upon dengue virus infection in Aag2 cell line

Abstract

The N⁶-methyladenosine (m⁶A) modification of RNA has been reported to affect viral infections. Studies have confirmed the role of m⁶A in replication of several vector-borne flaviviruses, including dengue virus (DENV), in mammalian cells. Here, we explored the role of m⁶A in DENV replication in the mosquito Aedes aegypti Aag2 cell line. We first determined the presence of m⁶A on the RNAs from mosquito cells and using methylated RNA immunoprecipitation and sequencing (MeRIP-Seq) identified m⁶A modification of the mosquito transcriptome and those that changed upon DENV infection. Depletion of m⁶A methyltransferases and the m⁶A binding protein YTHDF3 RNAs decreased the replication of DENV. In particular, we found that the Ae. aegypti ubiquitin carrier protein 9 (Ubc9) is m⁶A modified and its expression increases after DENV infection. Silencing of the gene and ectopic expression of Ubc9 led to reduced and increased DENV replication, respectively. The abundance of Ubc9 mRNA and its stability were reduced with the inhibition of m⁶A modification, implying that m⁶A modification of Ubc9 might enhance expression of the gene. We also show that the genome of DENV is m⁶A modified at five sites in mosquito cells. Altogether, this work reveals the involvement of m⁶A modification in Ae. aegypti-DENV interaction.

Fig. 1. m⁶A modification of Ae. aegypti transcripts analysed by dot blot and methylated RNA immunoprecipitation and sequencing (MeRIP-Seq).

a A diagram showing the MeRIP-Seq data analysis procedure. b Confirmation of RNA N⁶-methyladenosine (m⁶A) methylation in Ae. aegypti. Total RNA from Aag2 and Vero cells were extracted and subjected to a dot blot assay using a specific anti-m⁶A antibody. EGFP transcripts were synthetised in vitro and used as negative control. The input RNAs were directly stained with ethidium bromide. c Normalized density of m⁶A peaks between immunoprecipitated (black lines) and input (red lines) samples following MeRIP-Seq indicating enrichment of m⁶A in the IP samples as part of the quality control of the data. d Pie chart showing distribution of m⁶A peaks in Ae. aegypti transcripts regions. e The consensus m⁶A motif DRA*CH (D = G/A/U, R = G/A, * modified A, H = U/A/C) was enriched in the identified m⁶A peaks.

Fig. 2. Gene ontology scatterplot of Ae. aegypti transcripts with m⁶A modification.

The scatterplot shows the cluster representatives of GO terms identified from Ae. aegypti transcripts with m⁶A modification generated by Revigo. These GO terms remained after the redundancy reduction, in a two-dimensional space derived by applying multi-dimensional scaling to a matrix of the GO terms’ semantic similarities. a Biological process, b molecular function and c cellular component.

Fig. 3. Dengue virus infection alters m⁶A modification of Ae. aegypti transcripts.

a Confirmation of DENV infection of Aag2 cells. RT-PCR analysis of RNA extracted from Aag2 (Mock) and Aag2 cells infected with 1 MOI DENV 5 days after infection using DENV-specific and RPS17 (control) primers. The PCR products were analysed on an agarose gel. b Venn diagram showing summarized numbers of differentially expressed genes, with and without m⁶A peak changes. c Volcano plot showing differentially expressed genes in Aag2 cells in response to DENV infection. Green and yellow dots represent up-regulated and down-regulated genes with fold change >2 and p value <0.05, and those in red and blue are DEGs with m⁶A peaks, with the blue ones being DEGs with changed m⁶A peaks upon DENV infection. d Number of significantly differentially expressed genes between mock and DENV-infected Aag2 cells at five days post infection among which there were 26 with m⁶A peaks, of which four showed modified m⁶A upon DENV infection. e Heatmap of Ae. aegypti genes with m⁶A sites, which were differentially expressed in response to DENV infection (see Table 1 for the list of genes). The four genes that showed differential m⁶A and gene expression upon DENV infection are indicated by asterisks.

Fig. 4. 3-Deazaadenosine, an inhibitor of m⁶A, reduces DENV replication in Ae. aegypti cells.

a Confirmation of the effect of 3-DAA on inhibition of m⁶A modification in Aag2 cells. Total RNA from 3-DAA treated Aag2 and control cells (DMSO treated) were extracted and subjected to a dot blot assay using a specific anti-m⁶A antibody. The input RNAs were directly stained with ethidium bromide. b Inhibition of DENV replication by 3-DAA. Aag2 cells treated with 100 μM 3-DAA or DMSO were infected with DENV 72 h after treatment with 3-DAA. Cells were collected at 72 hpi for quantification of DENV RNA by RT-qPCR. c Viability of Aag2 cells treated with 3-DAA (100 μM) measured at 72 hpi. Error bars represent mean ± S.D. ns, not significant; *p < 0.05, t-test.

Fig. 5. Silencing m⁶A methyltransferase genes reduces DENV replication.

Expression levels of a METTL3 and b METTL14 in Aag2 cells transfected with dsRNA to METTL3 (dsMETTL3), and double transfection with dsRNAs to METTL3 and METTL14. dsRNA to GFP (dsGFP) was used as negative control in addition to the Cellfectin transfection reagent only. c, d Reductions in DENV replication in Aag2 cells by silencing METTL3/METTL14 or both. Cells were collected at 72 hpi for RNA extraction and quantification of DENV RNA by RT-qPCR (c), and the supernatants were harvested for quantification of DENV virions by plaque assay (d). Error bars represent mean ± S.D. ns, not significant, **p < 0.01; ***p < 0.001; ****p < 0.0001. ANOVA test with post hoc comparisons.

Fig. 6. m⁶A YTHDF3 reader modulates DENV replication in Ae. aegypti cells.

a Confirmation of silencing of YTHDF3 homologue by dsRNA (dsYTHDF3) in Aag2 cells. dsRNA to GFP (dsGFP) was used as a negative control in addition to Cellfectin transfection reagent only. b, c Inhibition of DENV replication by YTHDF3 silencing. Aag2 cells transfected with dsRNA targeting GFP or dsRNA targeting YTHDF3 were infected with DENV 72 h after transfection. Cells were collected at 72 hpi for RNA extraction and quantification of DENV RNA by RT-qPCR (b), and the supernatants were harvested for quantification of DENV virions by plaque assay (c). Error bars represent mean ± S.D. *p < 0.05; **p < 0.01. ANOVA test with post hoc comparisons.

Fig. 7. Ubc9 facilitates DENV replication in Ae. aegypti cells.

a Confirmation of silencing of four genes selected from differentially expressed genes by RNAi in Aag2 cells. b Inhibition of DENV replication by Ubc9 (AAEL025779) silencing. Aag2 cells transfected with dsRNA targeting GFP or dsRNA targeting the four genes were infected with DENV-2 72 h after transfection. Cells were collected at 72 hpi for quantification of DENV RNA by RT-qPCR. Control GFP expression levels in a and b were adjusted to 1. ***p < 0.001 (t-test). c Location of the m⁶A peak in Ubc9. d RT-qPCR analysis of RNA from Aag2 cells infected with DENV for 72 h showed consistency in up-regulation of Ubc9 transcript levels with the RNA-Seq data. FC, fold change. e Aag2 cells transfected with pSLfa-GFP or pSLfa-GFP-Ubc9 were infected with DENV 72 h after transfection. Cells were collected at 72 hpi for RNA extraction and quantification of DENV genomic RNA by RT-qPCR. Error bars represent mean ± S.D. *p < 0.05; **p < 0.01. One-way ANOVA with post hoc comparisons.

Fig. 8. Suppression of m⁶A modification reduces Ubc9 transcript levels.

a Confirmation of silencing of METTL3 and METTL14 by RNAi in Aag2 cells. Control GFP was adjusted to one. *p < 0.05; **p < 0.01 (one-way ANOVA with post hoc comparisons). b Decrease of Ubc9 expression by silencing m⁶A writers. Aag2 cells were transfected with dsRNA targeting GFP or dsRNA targeting METTL3 or METTL14. Cells were collected at 72 days post-transfection for quantification of Ubc9 RNA levels by RT-qPCR. ****p < 0.0001; ns, not significant, one-way ANOVA with post hoc comparisons. c RT-qPCR was used to evaluate the transcript levels of Ubc9 after using m⁶A inhibitor, 3-DAA. *p < 0.05 (t-test). Error bars represent mean ± S.D.

Fig. 9. DENV infection promotes Ubc9 RNA stability.

a Measurement of Ubc9 RNA in mock and DENV-infected Aag2 cells. At 72 hpi, cell culture medium was replaced with medium containing 1 μM ActD. RNA was collected at the indicated times post-treatment and subjected to RT-qPCR to determine remaining relative RNA levels. **p < 0.01; ***p < 0.001 (t-test). b Measurement of Ubc9 RNA in m⁶A DENV-infected m⁶A-inhibited Aag2 cells with 3-DAA. At 72 hpi, cell culture medium was replaced with medium containing 3-DAA. After 24 h post treatment, cell culture medium was replaced with medium containing ActD. RT-qPCR was used to measure the remaining relative RNA levels *p < 0.05; **p < 0.01 (t-test).

Fig. 10. DENV RNA is modified by m⁶A in infected mosquito cells.

a MeRIP-RT-qPCR analysis of RNA harvested from DENV-infected Aag2 cells (72 hpi) and immunoprecipitated with anti-m⁶A or IgG using primers to DENV genome. For the extracellular virions, the supernatant collected from DENV-infected cells was treated at room temperature with 17 μg of RNase A per ml for 20 min to remove any naked RNA. Eluted RNA was quantified as a percentage of input. m⁶A Control, positive control RNA with m⁶A modification; unmodified, negative control RNA without m⁶A modification. Error bars represent mean ± S.D. *p ≤ 0.01, ***p ≤ 0.001 (t-test). b Map of m⁶A-binding sites in the DENV RNA genome by MeRIP-Seq of RNA isolated from DENV-infected Aag2 cells based on three biological replicates. Read coverage, normalized to the total number of reads mapping to the viral genome for each experiment, is in red for MeRIP-Seq and in blue for input RNA-Seq. The five identified peaks (P1-P5) are indicated on the graph.