Anopheles Rin activity is proviral for ONNV and influences multiple immune pathways in mosquitoes

Rin/G3BP is proviral for infection of many alphaviruses in mammalian cells and for infection of CHIKV in Aedes albopictus [12,19,25] but the role of Anopheles Rin on viral infection has not been examined. Therefore, we first tested whether Rin is proviral for ONNV in Anopheles coluzzii mosquitoes by silencing Rin transcript using RNAi-mediated gene silencing. Double-stranded RNA (dsRNA) specific for Rin (dsRin), or irrelevant GFP control (dsGFP) were injected into mosquitoes followed by bloodmeal infection with ONNV. Unfed females were removed, and ONNV infection in abdomens was measured by viral titration 3 days (d) post-bloodmeal. We found that Rin silencing significantly decreased ONNV infection prevalence (defined as the proportion of fully fed mosquitoes positive for ONNV) (Fig 1A) and infection intensity or titer (defined as the viral load measured only in the ONNV-positive mosquitoes) (Fig 1B). This result indicates that Rin displays a proviral activity for ONNV in An. coluzzii.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Rin is a proviral factor for ONNV infection in Anopheles coluzzii mosquitoes.

Female mosquitoes were treated with dsRin or dsGFP two days prior to an infectious bloodmeal. ONNV infection was measured 3 d post-bloodmeal by viral titration. (A) The effect of Rin silencing on infection prevalence among bloodfed mosquitoes as a proportion of the total mosquitoes analyzed. y-axis, percentage of mosquitoes positive for ONNV in dsRin or dsGFP-treated mosquitoes. (B) The effect of Rin silencing on infection intensity in the positive mosquito samples. y-axis, load of viral particles, each point represents the viral titer in one mosquito. Error bars indicate the SEM. Results are from three independent replicates, n = 72 per condition. Analysis of infection prevalence was performed using a chi-squared analysis test. A two-tailed non-parametric unpaired Mann-Whitney test was performed to assess the statistical significance of the difference in viral titer in mosquitoes. The P-values were considered significant if p < 0.05 (** P < 0.01, *** P < 0.005, ns non-significant).

https://doi.org/10.1371/journal.ppat.1014423.g001

We next determined whether Rin influences basal immunity in uninfected mosquitoes by measuring the expression of specific components of Imd, Jak/STAT, RNAi, and Toll pathways.

Silencing of Rin in naïve mosquitoes displayed a striking effect upon the Imd and JAK/STAT pathways (Fig 2A), which were both shown to be efficient antiviral pathways against ONNV in Anopheles [24]. In the Imd pathway, Rin silencing caused decreased transcript abundance of the Imd positive regulator transcription factor Rel2 and two Imd pathway effector genes, the leucine-rich repeat protein APL1A and the thioester-containing protein TEP4. Both APL1A and TEP4 are transcriptionally regulated by Rel2 but not by the Toll pathway positive regulator Rel1 [26–28]. Therefore, APL1A and TEP4 are informative specific reporters for Imd pathway activity, and at least APL1A is antiviral for ONNV in Anopheles (TEP4 was not tested) [24]. Rel2, APL1A and TEP4 displayed a gradual tendency of increased transcript levels from 24 h to 72 h post-Rin silencing, but which was not significant for any except for TEP4 at 72 h, for an unknown reason. In the JAK/STAT pathway, Rin silencing caused increased transcript abundance of the JAK/STAT positive regulator, Stat-A, and therefore Rin is required to limit Stat-A transcript to basal levels. Thus, silencing of Rin causes widespread dysregulation of the Imd and JAK/STAT pathways, and normal Rin function is required for controlling basal activity of these immune pathways.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Rin function is required for transcriptional basal immunity in uninfected Anopheles mosquitoes.

Female mosquitoes were treated with dsRin or control dsGFP. The influence of Rin silencing at 24, 48 and 72h post-injection on gene transcript abundance is shown for (A) the Imd and JAK/STAT pathways, (B) the siRNA pathway, and (C) the Toll pathway. For all panels, x-axis indicates dsRNA treatments. Each bar represents the level of transcript abundance of each immune gene relative to dsGFP-treated controls (defined as 1.0), and error bars indicate the SEM. Results are from three independent replicates, n = 18 per condition. Student t-test analysis, which compares the mean of two independent groups (control and treatment) were used to assess the statistical significance of the difference in transcript abundance. The P-values were considered significant if p < 0.05 (*P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001).

https://doi.org/10.1371/journal.ppat.1014423.g002

The siRNA pathway has been studied for its role in counteracting mosquito infection by numerous RNA viruses. The endonucleases Argonaute-2 (Ago2) and Dicer-2 (Dcr2) are essential catalytic components of the RNA-induced silencing complex [29], and Ago2 was previously shown to be antiviral for ONNV infection in Anopheles [24]. The transcript abundance of Ago2 and Dicer2 both increased 48h after Rin silencing, and therefore Rin acts to limit the expression of the siRNA pathway in uninfected mosquitoes (Fig 2B). Finally, for the Toll pathway, we tested the expression of the transcription factor Rel1, the Toll negative regulator Cactus as well as effector genes regulated, at least in part, by the Toll pathway, APL1C and TEP3 [26] (Fig 2A). Silencing of Rin does not cause a significant effect on Toll pathway-related transcript abundance, except for causing a late increase of APL1C transcript (Fig 2C).

Taken together, these results from uninfected An. coluzzii indicate that, among these four major arms of mosquito innate immunity, Rin significantly influences the basal activity of key components mainly in the Imd and JAK/STAT pathways. Thus, Rin is required to maintain the overall integrity and homeostasis of the basal immune readiness of mosquitoes in the unstimulated resting state, including of known antiviral factors for ONNV in Anopheles, consistent with the hypothesis that Rin proviral function could be linked to viral manipulation of host immunity.

Anopheles Rin mediates inhibition of the Imd pathway during ONNV infection

We next examined the influence of Rin on Anopheles immunity to ONNV after challenge with an ONNV infective bloodmeal. Transcript abundance of the immune gene panel was measured at 2 d and 3 d after the infective bloodmeal.

Rin silencing during the primary midgut infection of ONNV increases the transcript levels of genes in three of the four major immune signaling pathways, as compared to control mosquitoes treated with dsGFP (Fig 3A-3C). The major influence was observed in the Imd pathway, where the Imd positive regulators Rel2 and Rel2F, and the anti-ONNV factor APL1A [24] are significantly induced by Rin silencing (Fig 3A). Thus, unlike in uninfected mosquitoes, during ONNV infection Rin function acts to decrease transcript levels of these antiviral genes. Interestingly, although the transcript level of the JAK/STAT positive regulator, Stat-A, is regulated by Rin in uninfected mosquitoes (Fig 2A), ONNV infection alters Stat-A regulation by abolishing Rin control over Stat-A. In addition, similarly to naïve mosquitoes, expression of the major regulator of the RNAi pathway, Ago2, remains significantly induced by Rin silencing during infection, although the effect is more significant during infection (Fig 3B). Finally, neither the positive (Rel1) nor negative (Cactus) regulators of the Toll pathway were influenced by absence of Rin activity during infection (Fig 3C). Two immune effectors thought to be regulated by Toll, APL1C and TEP3, were influenced by silencing of Rin during infection. The reason for this is not clear given the absence of a Rin effect on the Toll pathway regulators, and may indicate a previously unrecognized co-regulation of these effectors by pathways other than Toll.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 3. Rin is required for ONNV-dependent inhibition of the Anopheles Imd pathway.

Female mosquitoes were treated with dsRin or control dsGFP and two days later were fed an ONNV infectious blood meal. The influence of Rin silencing is shown for (A) the Imd and JAK/STAT pathways, (B) the siRNA pathway, and (C) the Toll pathway. For all panels, x-axis indicates dsRNA treatments. Each bar represents the level of transcript abundance of each immune gene relative to dsGFP-treated control (defined as 1.0), and error bars indicate the SEM. Results are from three independent replicates, n = 48 per condition. Student t-test analysis, which compares the mean of two independent groups (control and treatment) were used to assess the statistical significance of the difference in transcript abundance. The P-values were considered significant if *P < 0.05, ** P < 0.01, *** P < 0.001, ns non-significant.

https://doi.org/10.1371/journal.ppat.1014423.g003

Overall, and unlike in uninfected mosquitoes, the most important effect of Rin silencing during ONNV infection appeared to be Rel2 overexpression, indicating that Rin activity inhibits the induction of Rel2 only in the presence of virus infection. This suggests that Rin could be serving as a target for viral manipulation, in order to inhibit the potent antiviral Imd pathway.

Rin proviral activity is mediated by its inhibition of Imd function during ONNV infection

We showed above that Rin activity during ONNV infection displays an inhibitory effect upon the Imd pathway, which has potent antiviral activity against ONNV [24]. Therefore, we tested whether the control by Rin of Imd pathway activation could explain Rin proviral activity. Silencing of either Rin or Rel2 alone, or co-silencing of Rin and Rel2, were carried out in mosquitoes followed by bloodmeal challenge with ONNV. Virus levels were measured in individual mosquitoes at 3 d post-bloodmeal by viral titration of abdomens and detection of focus-forming units (FFU). Silencing of Rin significantly decreased the proportion of ONNV-infected mosquitoes (infection prevalence; Fig 4A) as well as the number of ONNV particles (infection intensity or titer; Fig 4B) as compared to the dsGFP-treated control group. Uninfected mosquitoes were removed from the measurement of infection intensity to avoid conflating prevalence and intensity measures. Silencing of Rel2 alone did not alter infection prevalence (Fig 4A), but did cause a significant increase in mosquito viral titers (Fig 4B), consistent with our previous observation for Rel2 and ONNV infection [24]. Thus, ONNV infectivity to Anopheles is compromised by silencing of Rin activity, indicating that Rin activity is required for full ONNV infection levels, and is enhanced by silencing of Rel2.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 4. Rin proviral activity is abolished by co-silencing of the Imd pathway during ONNV infection.

Mosquitoes were treated with the indicated dsRNAs 2 d prior to an ONNV infectious bloodmeal. Fully engorged mosquitoes were sampled at 3 d post-bloodmeal to measure ONNV infection by viral titration. (A) ONNV infection prevalence, the proportion of blood-fed mosquitoes positive for infection was measured. Bars indicate the percentage of mosquitoes positive for ONNV. (B) ONNV infection intensity was measured as the viral titer of only positive mosquitoes (uninfected mosquitoes were removed from the determination of intensity). Points represent the viral titer in one mosquito. (C) Gene silencing efficiency was confirmed on the day of infectious blood feeding for Rin and Rel2 transcript abundance (S1 File). Bars represent the level of transcript abundance of Rin and Rel2 relative to the dsGFP-treated control (defined as 1.0). For all panels, x-axis indicates dsRNA treatments. Results were from three independent replicates (A, B) with n = 72 mosquitoes per condition or (C) a pool of at least 6 mosquitoes per replicate. Analysis of infection prevalence was performed using a chi-squared analysis test. A two-tailed non-parametric unpaired Mann-Whitney test was performed to assess the statistical significance of the difference in viral titer in mosquitoes. Student t-test analysis, which compares the mean of two independent groups (control and treatment) were used to assess the statistical significance of the difference in transcript abundance. The P-values were considered significant if p < 0.05 ** P < 0.01, *** P < 0.001, **** P < 0.0001, ns non-significant. Error bars indicate SEM.

https://doi.org/10.1371/journal.ppat.1014423.g004

To query whether the proviral activity of Rin could potentially act by subverting Imd antiviral protection during ONNV infection, Rin and Rel2 were simultaneously silenced, followed by an ONNV infectious bloodmeal. The loss of Rel2 complemented Rin silencing and restored normal levels of infection prevalence as well as viral load, as compared to Rin silencing alone (Fig 4A & 4B). Thus, with Rel2 silenced, Rin no longer displays detectable proviral activity for ONNV, because infection prevalence and intensity of the double silencing are the same wild-type level as the dsGFP controls with normal Rin and Rel2 activity. This outcome indicates that achieving a normal level of ONNV infection is independent of Rin activity when Rel2 is silenced, and suggests that the presence of Rin is required for virus-dependent suppression of Imd pathway activity. Such a mechanism is further supported by examining the directionality of transcriptional control between Rin and Rel2 (Fig 4C). Silencing of Rin significantly decreases its own transcript abundance and also that of Rel2, while silencing of Rel2 decreases its own transcript but has no effect on Rin transcript abundance. Thus, Rin controls Rel2 by a hierarchical and unidirectional mechanism of transcript regulation, in which Rel2 does not branch back to influence Rin transcript levels.

Further detail is provided by examination of the infection phenotype caused by silencing of Rel2 alone (Fig 4A). Rel2 activity limits viral titer, but alone does not appear to reduce infection prevalence, although the high (80%) infection prevalence in the dsGFP controls probably does not provide statistical power to detect a significant effect in the dsRel2 treatment alone, in contrast to the phenotype produced by silencing of Rin alone. This suggests that the nearly double rate of mosquitoes displaying sterile immunity for ONNV (i.e., absence of detectable infection) after Rin silencing may require the combined contribution of additional antiviral mechanisms that are also controlled by Rin, and cannot be explained only by Rin control over Rel2 alone.

These results confirm a proviral role for Rin during the infection of An. coluzzii with ONNV after an infectious bloodmeal. However, the Rin proviral phenotype requires an intact Imd pathway. In contrast, suppression of the Imd pathway by silencing of positive regulator Rel2 rescues full ONNV infection even in the absence of Rin, indicating that Rin is unnecessary as a proviral factor in mosquitoes without an active Imd pathway. The simplest interpretation for these observations is that viral co-optation of normal Rin cellular function, particularly including its role in homeostasis of basal immunity, can subvert at least the important Imd, and potentially other antiviral immune pathways.

Rin biochemically interacts with ONNV and its non-structural protein nsP3 in Anopheles cells

In mammalian and Aedes mosquito cells, G3BPs and Rin have been shown to colocalize with the non-structural protein 3 (nsP3) of multiple alphaviruses including CHIKV and ONNV [14]. However, to our knowledge Anopheles Rin has not been examined. Here, we analyzed the colocalization and interaction of ONNV nsP3 with An. coluzzii Rin in 4a3A cells (Fig 5). To measure colocalization, we co-transfected a C-terminal streptavidin tagged construct of Anopheles Rin (Rin-strep) and a construct expressing nsP3 of ONNV, with expression of both driven by the Anopheles actin promoter (Fig 5A, top panel, nsP3 transfection). Colocalization of Rin and virally-produced nsP3 was also tested after direct infection of cells with ONNV, without nsP3 transfection (Fig 5A, lower panel, ONNV infection). At least 87% of cells counted displayed colocalization of Rin and ONNV nsP3 signal with both nsP3 sources (Fig 5B). There was no significant difference in colocalization between Rin and transfected nsP3 in uninfected cells, and viral origin nsP3 in ONNV-infected cells.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 5. Rin biochemically interacts with ONNV and nsP3 in Anopheles cells.

Immunostaining analysis of Rin colocalization with nsP3 and ONNV in Anopheles cells. (A) Top panel (nsp3 transfection): Anopheles cells were co-transfected with a C-terminal streptavidin-tagged construct of Anopheles Rin (Rin-strep) and a construct expressing nsP3 of ONNV, both driven by the Anopheles actin promoter. Rin was detected using streptavidin Strep-Tactin conjugated with dye-649 (red, #2-1568-050, IBA) and nsP3 was detected using a polyclonal antibody against ONNV nsP3 plus Alexa Fluor 555 conjugated secondary antibody (green) (#A27039, Invitrogen). Lower panel (ONNV infection): Anopheles cells were co-transfected with a C-terminal streptavidin-tagged construct of Anopheles Rin (Rin-strep), but not the construct expressing nsP3 of ONNV, and were infected with ONNV as the source of nsP3. Rin was detected using streptavidin Strep-Tactin conjugated with dye-649 (red), and the viral-source nsP3 was detected using the same polyclonal antibody against ONNV nsP3 plus Alexa Fluor 555 conjugated secondary antibody (green). Both panels, nuclei were stained with Hoechst 33342 dye (blue). Scale bars, 10 µm. Representative images are shown. (B) The colocalization of Rin and ONNV nsP3 signal was compared for transfection-source nsp3, or viral-source nsP3, by counting cells stained as above. 87% or 88% of cells counted, respectively, displayed colocalization of Rin and ONNV nsP3 signal under each condition of nsP3 source. There was no significant difference in colocalization for the two conditions of nsP3 source. ns, non-significant. (C) Streptavidin pull-down was used to affinity purify Rin-strep (Rin-strep) from 4a3A cells co-transfected with untagged nsP3, used as bait for nsP3 cellular partners. The eluates of the pull-down were analyzed by Western blot. Rin were detected using a streptavidin mouse monoclonal antibody conjugated with HRP, and nsP3 was detected using the same polyclonal antibody as above against ONNV nsP3 plus HRP-conjugated anti-rabbit secondary antibody. The top panel shows positive control detection of the expected Rin-strep band of 89 kDa detected by anti-streptavidin only in the cells transfected with the Rin-strep plasmid (lanes 2 and 4). The bottom panel shows the presence of a 70 kDa band corresponding to ONNV nsP3, which was only seen in the elution from cells co-transfected with Rin-strep and nsP3 (lane 4), but was not observed in the non-transfected cells (NT, lane 1), nor in cells transfected only with Rin-strep but not nsP3 (lane 2), nor in cells transfected only with nsP3 but not Rin-strep (lane 3). An unknown band of 110 kDa was also stained by the nsP3-polyclonal antibodies. NT, non-transfected control cells, Rin-strep, cells transfected only with Rin-strep, nsP3, cells transfected only with nsP3, Rin-strep + nsP3, cells co-transfected with both plasmids.

https://doi.org/10.1371/journal.ppat.1014423.g005

To confirm the interaction between Rin and nsP3, we used a streptavidin-pull down assay to purify Rin conjugated with streptavidin (Rin-strep) from Anopheles 4a3A cells co-transfected with untagged nsP3. The resulting Western blots were detected with antibodies directed against nsP3, or as a positive control against streptavidin (Fig 5C). A 70 kDa band corresponding to ONNV nsP3 was only seen in the elution from Rin-strep in co-transfected cells but was not observed in the eluate of only nsP3- or Rin-strep transfected cells, indicating that nsP3 was bound and pulled down by Rin-strep. An unknown non-specific band of 110 kDa was also stained by the nsP3-polyclonal antibodies. The expected control band of 89 kDa was detected by anti-streptavidin in the cells transfected with Rin-strep construct. Therefore, in Anopheles 4a3A cells, Rin forms a stable physical interaction with ONNV nsP3, either by direct binding or indirectly through accessory binding proteins.

ONNV nsP3 interaction with Rin alters the Rin-associated protein complex

ONNV infection alters Rin activity and causes Rin-dependent inhibition of Imd. ONNV nsP3 interacts stably with Rin in Anopheles cells, as it does with Rin/G3BPs in general, and this interaction is likely to be involved in the coopting of Anopheles Rin function. To shed light on the mechanism by which Rin regulates host immunity, we analyzed by mass spectrometry the Rin and Rin/nsP3 binding complex in Anopheles 4a3A cells. We first confirmed proviral activity of Rin in 4a3A cells, as was seen in mosquitoes, and found that silencing of Rin decreased the abundance of intracellular ONNV genomic RNA (S1A Fig) and infectious particles released into the supernatant (S1B Fig). Rin silencing in 4a3A cells was efficient (S2 Fig).

We then captured Rin and Rin/nsP3 binding complexes, using as bait Rin tagged with C-terminal streptavidin, transfected alone (Rin-strep) or co-transfected with ONNV nsP3 (Rin-strep+nsP3) (Fig 6, raw blot data in S3 Fig). Mass spectrometry was also carried out upon extensive controls for binding specificity to exclude irrelevant interactions, as follows: i) pulldown from non-transfected cell extract, to control for nonspecific protein binding to the Strep-Tactin magnetic beads, ii) pulldown from extract of cells transfected with streptavidin conjugated to the irrelevant protein, GFP (GFP-strep), to control for cellular proteins binding either to streptavidin, to the beads, and/or displaying general nonspecific protein binding activity, iii) pulldown from extract of cells transfected only with nsP3 without Rin-strep, to control for any binding of nsP3 without the streptavidin moiety to the beads, thus pulling down irrelevant proteins. Proteins found in the control conditions were considered non-specific and were not analyzed further.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 6. ONNV nsP3 binding alters Rin protein partners in Anopheles cells.

Eluates of cells transfected with Rin-strep or co-transfected with Rin-strep and nsP3 of ONNV (as in Fig 5C) were analyzed by mass spectrometry to identify protein partners. The proteins identified in the control conditions were classified as non-specific and were removed from the analysis. (A) Summary of the biological functions of the protein partners of Rin-strep, or (B) of Rin-strep in presence of nsP3 are shown. (C) Graph indicates the partner proteins in common between Rin-strep and the complex Rin-strep/nsP3. (D-E) The average abundance (Intensity-Based Absolute Quantification, IBAQ) scores of the protein partners of Rin-strep (dark blue) or Rin-strep/nsP3 complex (light blue) or their protein partners in common (plain blue) are indicated. Each bar represents the average IBAQ abundance of each protein partner of Rin-strep or of the complex Rin-strep/nsP3. Data are from four replicates per condition. Raw data and statistical analyses of mass spectrometry analysis are presented in S2 File.

https://doi.org/10.1371/journal.ppat.1014423.g006

Most Rin partners discovered in the proteomic screen are proteins with unknown function. Of the known proteins, Rin interacts in pathways related to RNA metabolism, protein synthesis components, ubiquitination pathways and signaling (Fig 6A; Raw data and statistical analyses of mass spectrometry analysis are presented in S2 File). The biological pathways of Rin partners are consistent with reported Rin involvement in Ras metabolism [30] and RNA metabolism [17]. However, when Rin is in complex with viral nsP3, the profile of protein partners is altered (Fig 6B). A total of 48 proteins interact with Rin in the presence or absence of nsP3, seven of them interacting only with Rin alone and 8 proteins interacting with Rin only when nsP3 is present (Fig 6C). These 8 new proteins become strong candidates to explain the biological effect of ONNV upon Rin function. The Rin-nsP3 complex includes 8 new protein partners that may be involved in protein synthesis, RNAi pathway and cellular/nuclear transport, although most of these proteins (5/8) have unknown function, and will require further work to understand, and fewer partners involved in GDP/GTP metabolism and in signaling pathways (Fig 6D and 6E). Therefore, the presence of nsP3 modifies the biological profile of Rin-interacting protein partners, likely in order to modify the function of Rin during ONNV infection.

Ethics statement

The protocol for the ethical treatment of the animals used in this study was approved by the research animal ethics committee of the Institut Pasteur, “C2EA-89 CETEA Institut Pasteur” as protocol number 202195.02. The Institut Pasteur ethics committee is authorized by the French Ministry of Higher Education and Research (MESR) under French law N° 2001–486, which is aligned with Directive 2010/63/EU of the European Commission on the protection of animals used for scientific purposes. The study was performed using practices and conditions approved by the Institut Pasteur Biosafety Committee as protocol number CHSCT 14.114.

Mosquito, cell and virus strains

The Anopheles coluzzii Ngousso colony was initiated in Cameroon in 2006 and has been maintained at the Institut Pasteur Center for the Production and Infection of Anopheles (CEPIA) facility since 2008. Mosquitoes are maintained at 26°C, 70% humidity. Female mosquitoes used for all experiments are less than 5 days post-emergence.

Anopheles coluzzii 4a3A cells originated from larvae and spontaneously immortalized [41]. 4a3A cells are maintained at 27°C without CO₂ in Insect-XPRESS Protein-free Insect Cell Medium complemented with heat-inactivated Fetal Calf Serum (FCS) (ThermoFisher Scientific). The C6/36 cell line (ATCC) [42] was used for viral titration and is maintained in Leibovitz’s L-15 medium (ThermoFisher Scientific) supplemented with MEM non-essential amino acid (ThermoFisher Scientific) and complemented with heat-inactivated Fetal Calf Serum (ThermoFisher Scientific) at 28°C without CO₂. Baby hamster kidney cell line (BHK-21) was used to produce virus particles from infectious clones and are maintained in Dulbecco’s Modified Eagle Medium (DMEM) (1X) + GlutaMax media (ThermoFisher Scientific) complemented with heat-inactivated Fetal Calf Serum (ThermoFisher Scientific) at 37°C with 5% CO₂.

The infectious clone of ONNV originated from ONNV strain Igbo Ora-IBH10964 (Igbo-Ora strain) and was isolated from a human febrile patient during the epidemic of 1966 in Nigeria (accession number AF079457) [43]. ONNV was either produced on Anopheles 4a3A cells and viral titer done on kidney epithelial cells from African green monkey (Vero cells) were measured at 7.3 x 10⁵ plaque forming units/mL (pfu/mL) for cell infection. Mosquito infection used infectious ONNV produced on BHK-21 cells and the titer measured on C6/36 cells is 5.64 x 10⁷ FFU/mL. Infectious clone cDNA was linearized by PmeI, and viral RNA was transcribed using T7 RNA polymerase (Ambion mMESSAGE mMACHINE SP6 Kit). Transcribed RNA was electroporated into BHK-21 or 4a3A cells. Virus was recovered after 72 h and titer on C6/36 cells.

Primers and plasmids

The dsRNA amplicons were constructed using primers (given in S1 Table) derived from the An. gambiae PEST reference genome available in the VectorBase database. DsRNAs targeting GFP act as a control for the dsRNA treatment. DsRNAs were synthesized using T7 primers and the transcription kit of MEGAscript T7 (Invitrogen) following manufacturer instructions.

Plasmids used for colocalization, and streptavidin pull-down assay were constructed using Rin codon-optimized AGAP000403 sequences and ONNV nsP3 protein from IBH10964 strains graciously provided by Andres Merits, University of Tartu, Estonia. A C-terminal streptavidin tag II (W-S-H-P-Q-F-E-K) was added at the end of Rin sequences. For their expression in 4a3A cells, these genes are expressed under the Actin promoter (Ac5) in the plasmid pAc5.1. 4a3a cells lines were transfected using lipofectamine LTX (ThermoFisher Scientific) with 0.25 µg to 0.5 of plasmids depending on the number of cells used.

RNA extraction, cDNA synthesis and qPCR analysis

RNA was extracted from 4a3A cells or mosquitoes using Trizol reagent (Sigma Aldrich) and RNA miniprep kit (Ozyme). Complementary DNAs were synthesized with M-MLV transcriptase inverse kit (Invitrogen) using 1 µg of RNA. Quantitative polymerase chain reaction (qPCR) primers used for detection of the gene silencing and for analyzing the transcript level of immune genes and viral RNA are shown in S1 Table. The qPCR primers were checked for specificity prior to the analysis. All qPCRs were performed using SYBR green supermix (KAPA SYBR FAST ABI, Sigma-Aldrich) and the CFX96 Touch Real-Time PCR Detection System (BioRad). The ribosomal protein rpS7 gene was used as the internal control and the analysis of transcript relative expression to rpS7 was performed according to the 2 − ΔΔCt method. PCR cycle conditions were 95°C for 10 min, then 39 cycles of 95°C for 15 sec, 60°C for 1 min.

Antibodies and fluorescent labels

Sera of rabbit containing polyclonal rabbit antibodies provided by Andres Merits, University of Tartu, targeting the macro and the AUD domain of nsP3 of ONNV were used at 1:100 overnight at +4°C for colocalization and western blot assay. Secondary goat anti-Rabbit IgG (H + L) antibodies (Invitrogen) conjugated with Alexa Fluor 555 were used during colocalization assay at 1:500 for 45 min at room temperature (RT) in 0.2% Triton X-100, 1% BSA, and PBS 1X. Secondary antibodies anti-rabbit conjugated with HRP (Invitrogen) were used for Western blot at 1:8000 for 1 h at RT in 5% BSA in PBS 1X. Monoclonal antibodies targeting the streptavidin tag II and conjugated with HRP (IBA Lifesciences) were used at 1:10000 for 1 h at RT during Western blot in 5% BSA in PBS 1X. Streptavidin Strep-Tactin molecule conjugated with dye-649 (IBA Lifesciences) were used at 1:100 for 3 h at RT during colocalization assay in 0.2% Triton X-100, 1% BSA, and PBS 1X. DAPI (4', 6-diamidino-2 '-phenylindole, dihydrochloride) (ThermoScientific) was used at 1:1000 for 10 min at RT in PBS 1X for staining the nucleus of 4a3A cells during nsP3-Rin cellular colocalization assays. Detection of viral titration used monoclonal mouse antibodies against CHIKV Virus-like particles (The Native Antigen Company), diluted at 1:1000 in 0.1% BSA PBS 1X solution for 45 min at 37°C. Secondary goat antibodies against mouse IgG and IgM (H + L) conjugated with Alexa 488 (Invitrogen) were diluted at 1:500 in PBS 1X and incubated at 37°C for 30 min.

Gene silencing

For gene silencing in mosquitoes, 500 ng of dsRNA in a final volume of 70 nL was injected into the thorax of cold-anesthetized 1–2 d old An. coluzzii females using a Nanoject II injector (Drummond Scientific) and glass capillary needle as previously described [28]. The gene silencing efficiency was verified by RT-qPCR using pooled RNA from six unfed mosquitoes collected 2 days after dsRNA injection, and is presented for all figures in S1 File. For gene silencing genes in cultured cells, 1 × 10e5 An. coluzzii 4A3A cells were seeded and left to adhere overnight. Cells were incubated for 30 min on a rocker with 5000 ng of dsRNA in 200 µL of Insect Xpress media without FBS. After incubation, 300 µL of Insect Xpress media with 10% (vol/vol) FBS was added, and cells were incubated until collection or infection.

Cell infection

The 4a3A cell line was infected with ONNV at a standard multiplicity of infection (MOI) of 0.01 (except as noted for MOI 0.5 for nsP3-Rin cellular colocalization experiments) for 1 h at 28°C without CO₂ in Insect-X-Press media without FCS. After 1 hour of infection, cells were washed three times with Insect-X-Press without FCS. Then 2% FCS Insect-X-Press media were added to cells for the time of incubation. Cells were then collected at 24h or 48h post-infection.

Infectious blood feeding

The experimental protocol is based on published methods [44]. The infectious blood meal was 2/3 vol/vol rabbit erythrocytes and 1/3 vol/vol of viral solution for a final titer of 10⁷ FFU/mL. Adenosine triphosphate (ATP) at 10 mM was also added to the infectious blood. Infectious blood was fed to mosquitoes using the Hemotek system with artificial skin maintained at 37°C. Mosquitoes were fed for 1 h, and unfed females were removed. Fed mosquitoes were maintained at 28°C and 80–90% humidity with access to 10% sugar solution. At three days post-infectious feeding, abdomens and thoraces of female mosquitoes were separated, homogenized, and centrifuged at +4°C for 5 min at 10,000 rpm. Supernatants were collected and stored at -80°C until analysis.

Viral titration by focus formation assay

Only infected mosquitoes were analyzed for the determination of viral titer, to avoid conflating measures of infection prevalence and intensity. Ae. albopictus C6/36 cells were seeded in 96-well plates at 1.25 x 10⁶ cells/ Leibovitz’s L-15 medium (ThermoFisher Scientific) supplemented with MEM non-essential amino acids (ThermoFisher Scientific) with 10% of FBS and incubated at 28°C without CO₂, 48h prior to titration. Virus-containing samples were serially diluted, and 50µL of the dilutions were added to cells in one well. After one hour of incubation at 28°C, 150 µL of 1:1 4% carboxymethylcellulose (CMC) and L-15 containing 5% FBS was layered onto cells, along with antibiotic-antifungal solution (Life Technologies) at a final concentration of 1.5X. Plates were incubated at 28°C without CO₂ for 3 days, then cells were fixed using 3.4% formaldehyde in PBS 1X for 20 min at room temperature. Cells were then washed three times in PBS 1X and stained with primary antibodies. After three washes with PBS 1X, cells were stained with secondary antibodies. Viral titers were counted by observing foci using a fluorescent microscope and were expressed as focus forming units/mL (FFU/mL).

Confocal microscopy

Cells were seeded in µ-slide 8-well plate (Ibidi GmbH) and transfected using lipofectamine with 0.25 µg of plasmids. Two days post-transfection, cells were fixated with 4% formaldehyde for 20 min at RT. Cells were then blocked and permeabilized using a solution of 0.2% Triton X-100, 1% BSA, PBS 1X for 1 h at RT, and were stained with antibodies diluted in 0.2% of Triton X-100, 1% BSA, and PBS 1X. Cells were washed using 0.2% Triton X-100 1% BSA PBS 1X. Colocalization analysis in 4a3A used a laser-scanning confocal microscope (LSM700, Carl Zeiss Jena) at the UTechS Photonic BioImaging C2RT platform (Institut Pasteur). Image acquisition parameters were as follows: 63X oil-objective, frame size 512x512, pixel dwell 1.57 µsec, scan time 5.78 sec, 16-bit depth, image size 32.9x32.9 µm and 0.06 µm pixel size. Laser parameters were as follows: the DAPI laser (450 nm) had an intensity of 2, a pinhole diameter of 0.6 µm, a master gain of 730 and a digital gain of 1. The Red A555 laser (514 nm) had an intensity of 2.6, a pinhole diameter of 0.6 µm, a master gain of 711 and a digital gain of 1. The Far-Red D647 laser (633nm) had an intensity of 2, a pinhole diameter of 0.8 µm, a master gain of 700 and a digital gain of 1. Images were analyzed using Fiji software [45]. Background signals were calculated with the mean of signals from 5 pictures of the negative controls of each fluorophore. Mean of background signals were then removed from the analyzed pictures. Colocalization was quantified by counting the number of cells with colocalization pattern as compared to cells co-transfected (harboring both signals). Between 30 and 50 cells were counted in the analysis.

Streptavidin-pull down assay

First, 0.5 µg of plasmid constructs were transfected in 4a3A cells using lipofectamine LTX with Plus Reagent (ThermoFisher Scientific). Two days post-transfection, cells lysis was performed using cell lysis buffer (containing Tris 1 M, NaCl 5M, EDTA, Igepal, Protease inhibitor and Phosphatase inhibitor) for 30 min at +4°C and samples were centrifuged at 16,000g for 15 min at +4°C. Pull-down assay was performed using MagStrep type 3 beads (Strep-Tactin XT coated magnetic beads, 5% (v/v) suspension, IBA Lifesciences) with a binding capacity up to 0.85 nmol/μL beads (corresponding to 25.5 μg of a 30 kDa protein). Beads were first washed with cell lysis buffer then cell lysate was added on the beads and incubated for 2 h at +4°C on a tube rotator. After 2 h of incubation, lysates were removed, and washes were performed using 1X W buffer (IBA Lifesciences). Following washes, either on-beads digestion at the Proteomic platform is done for Mass spectrometry analysis or the elution of proteins is performed using 1X BXT buffer containing biotin (IBA Lifesciences) for 10 min. Eluates are then analyzed by Western Blot in order to detect strep-tag proteins (Rin or GFP, bait) and partners (nsP3, prey) using specific antibodies.

Protein samples were reduced in 1X DTT and heated at 95°C for 5 min. Then samples were mixed with XT sample buffer (Bio-Rad) and loaded on 4–12% Criterion SDS-PAGE gels (Bio-Rad). SDS-PAGE is run at 100 V for 10 min (stacking gel) and 150 V for 1 h (running gel) in NOVEX NUPAGE MOPS SDS Running Buffer 20X (Life Technologies). Following electrophoresis, proteins contained on gel are transferred onto a 0.2 µm nitrocellulose membrane (Bio-Rad) using Bio-Rad Trans-Blot Turbo Transfer System. Transfer program is run for 7 min at 2.5A and 25V. After transfer, immunoblots are blocked in a solution containing 5% of Bovine Serum Albumin (BSA) in TBS 1X (Tris-buffered saline, 0.1% Tween 20). Then, immunoblots were probed with antibodies at specific dilutions in 5% BSA TBS 1X. Detection step was performed using the Enhanced chemiluminescence (ECL) system (Clarity Western ECL substrate, BioRad) following manufacturer instruction. Blots were photographed and revealed taking one picture every second for 10 sec using ChemiDoc Imaging System (Bio-Rad). Blots were then annotated, and molecular weights were measured related to the protein molecular weight marker (Precision PlusProtein All Blue Standards, BioRad) using the Image Lab analysis software (BioRad). The molecular weight marker lane is labeled “M” on relevant blots. A positive control was performed using a commercial GFP-strep (Green Fluorescent Protein) (IBA LifeSciences) in order to confirm the success of the Western Blot.

Mass spectrometry analysis

The products of the streptavidin-pull down assay were analyzed for differential protein abundance by liquid chromatography mass spectrometry (LC-MS/MS). First, for on-bead digestion of samples, proteins were resuspended in 50 mM ammonium bicarbonate pH 8.0 and reduced for 30 min at room temperature (RT) with 5 mM TCEP and then alkylated at 50 mM iodoacetamide (Sigma) for 30 min at RT in the dark. Proteins were then digested with 0.5 µg Sequencing Grade Modified Trypsin (Promega). Following overnight digestion at 37°C with agitation at 800 rpm in a Thermomixer C (Eppendorf), the sample tubes were placed on a magnet for 1 min, and the supernatant containing the released peptides was transferred to a new tube. Resulting peptides were acidified at 1% with formic acid, desalted with an AssayMAP Bravo robot using C18 column and AssayMAP Peptide Cleanup Protocol (Agilent). All samples were dried in a Speed-Vac and peptides were resuspended in 2% ACN, 0.1% FA prior to LC-MS/MS analysis. LC-MS/MS analysis of digested peptides was performed on an Orbitrap Eclipse mass spectrometer (Thermo Fisher Scientific) coupled to an EASY-nLC 1000 (Thermo Fisher Scientific). Peptides were loaded (at constant pressure of 900 bars) and separated at 250 nl.min-1 on a home-made C18, 30 cm capillary column picotip silica emitter tip (75 μm diameter filled with 1.9 μm Reprosil-Pur Basic C18-HD resin, (Dr. Maisch HPLC GmbH) equilibrated in solvent A (2% ACN, 0.1% FA). Peptides were eluted using a gradient of solvent B (80% ACN, 0.1% FA) from 2% to 7% in 3 min, 7% to 31% in 42 min, 31% to 62% in 10 min, 62% to 95% in 5 min (total length of the chromatographic run was 70 min). The column was equilibrated with 10 µL of A at 900 bars. Mass spectra were acquired in data-dependent acquisition mode with the XCalibur software (Thermo Fisher Scientific). MS spectra were acquired at orbitrap resolution of 60k (at m/z 400), AGC target 800000, a custom max injection time. The scan range was limited from 300 to 1500 m/z. Peptide fragmentation was performed using higher-energy collision dissociation (HCD) with the energy set at 27 NCE. The MS/MS spectra acquired at a resolution of 30k (at m/z 400). Isolation window was set at 1.6 m/z. All acquisitions were done in profile and positive mode. Dynamic exclusion was employed within 30 sec. Raw data and statistical analyses of mass spectrometry analysis are presented in S2 File.

Peptide identification and statistical analysis were as follows. Raw data were analyzed using MaxQuant software version 2.0.3.0 [46] using the Andromeda search engine [47]. The MS/MS spectra were searched against a custom database from the Vectorbase repository for the PEST reference genome of Anopheles gambiae (download 25/04/2022). Usual known mass spectrometry contaminants and reversed sequences of all entries were included in the search. The Andromeda search was performed choosing trypsin as specific enzyme with a maximum number of two missed cleavages. Possible modifications included carbamidomethylation (Cys, fixed), oxidation (Met, variable), Nter acetylation (variable). The mass tolerance in MS was set to 20 ppm for the first search then 4.5 ppm for the main search and 20 ppm for the MS/MS. Maximum peptide charge was set to seven and seven amino acids were required as minimum peptide length. The “match between runs” feature was applied for samples having the same experimental condition with a maximal retention time window of 0.7 minute. One unique peptide to the protein group was required for the protein identification. A false discovery rate (FDR) cutoff of 1% was applied at the peptide and protein levels. For the differential analyses of one condition versus another, Intensity-Based Absolute Quantification (IBAQ) quantification was used. Proteins identified in the reverse and contaminant databases and proteins “only identified by site” were first discarded from the list of identified proteins. Only proteins with at least three quantified intensities in a condition were kept. The proteins of interest are therefore those which emerge from this statistical analysis supplemented by those which are considered to be present from one condition and absent in another.

Statistical analysis

All experiments are the results of at least three independent replicate experiments. For RT-qPCR analysis of RNA transcript levels relative to rpS7, the 2 − ΔΔCt method was used, and the difference in deltaCt distribution across biological replicates was statistically tested using unpaired t-test. Differences in the infection rates were analyzed using a chi-squared test. A two-tailed non-parametric unpaired Mann-Whitney test was performed to assess the statistical significance of the difference in viral titer in mosquitoes. The P-values were assessed with a null distribution. The P-values were considered significant if p < 0.05 (*p < 0.05; **p < 0.01; ***p < 0.005, ****p < 0.0001). Bar plots were created using GraphPad Prism (Dotmatics Software).