UGAL/Rockefeller strain Ae. aegypti mosquitoes were kept at 28°C and 70% relative humidity under a light-dark cycle of 12:12 hours as previously described (22, 23). Hatched larvae were transferred to plastic containers with sufficient water and fed with yeast extract daily. Pupae were collected and transferred to a plastic container in an insect dorm. Emerged mosquitoes were fed using cotton balls soaked with a 10% sucrose solution. Female mosquitoes three to five days post- eclosion (PE) were used for our experiments. The sucrose-soaked cotton balls were removed at least 12 hours before blood feeding. Female mosquitoes were permitted to blood-feed on an anesthetized ICR strain mouse for 15 to 30 minutes. ICR strain mice were anesthetized with an intraperitoneal injection of Avertin at a dose of 0.2 mL per 10 g of weight. All animal procedures and experimental protocols were approved by AAALAC-accredited facility, the Committee on the Ethics of Animal Experiments of the National Taiwan University College of Medicine (IACUC approval No: 20200210).

Ae. albopictus C6/36 cells were cultured in DMEM/MM (1:1) containing 2% heat-inactivated fetal bovine serum (FBS) and 1× penicillin–streptomycin solution. For virus production, cells were infected with the DENV2 strain 16681 at 0.01 multiplicity of infection (MOI). The culture supernatant was harvested at 7 dpi and stored at −80°C. To determine the viral titer, the virus stock was subjected to examination with a plaque assay, as previously described (24). Approximately 1.0 × 10⁷ PFU/mL DENV2 was used to infect the mosquitoes.

The whole bodies of three to five mosquitoes or the midguts of 20 to 30 mosquitoes were collected in 1.5 mL tubes containing 0.5 mL Trizol Reagent (Invitrogen). Tissue was homogenized with a rooter-stator homogenizer at room temperature for 5 minutes and centrifuged at 13000 rpm for 10 minutes at 4°C. After centrifugation, the supernatant was transferred to a new micro-tube with 0.1 mL chloroform (J. T. Baker) and mixed thoroughly at room temperature for 3 minutes. Samples were then centrifuged at 13000 rpm for 15 minutes at 4°C and the supernatant was transferred carefully to a new micro-tube with 0.25 isopropanol (J. T. Baker). Samples were gently mixed and stored at −80°C for 30 minutes. After precipitation, the samples were once again centrifuged at 13000 rpm for 30 minutes at 4°C. The supernatant was discarded and 0.5 mL 75% ethanol (Taiwan Burnett International Co., Ltd) was used to wash the RNA pellet. All resulting samples were centrifuged at 8000 rpm for 5 minutes at 4°C and the supernatant was discarded. Finally, the RNA pellet was dried in a laminar flow hood and dissolved in DEPC-H₂O. After Baseline-ZEROTM DNase (Epicentre) treatment, the RNA sample was stored at −80°C.

The RNA concentration was quantified with a spectrophotometer (Nanodrop 2000, Thermo) and was diluted with DEPC-H₂O at a concentration of 1 μg/μL. The RNA samples were reverse-transcribed to cDNA with a High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems). The cDNA samples were stored at −20°C for further use. Gene expression was analyzed with a polymerase chain reaction (PCR) using ProTaq Plus DNA Polymerase (Protech). The ribosomal protein S7 gene was used as an internal control.

The qPCR system used in this study was the SYBR Green dye binding system. SYBR Green binds to the minor groove of DNA and the target gene is quantified by detecting the resulting fluorescence signal. The cDNA sample was quantified with the KAPA SYBR FAST Universal qPCR kit (KAPA) and the qPCR primers were designed using ABI Primer Expression Software. PCR consisted of an initial denaturation at 95°C for 3 minutes, followed by 40 cycles at 94°C for 3 seconds, and 40 seconds at 60°C. Fluorescence readings were measured at 72°C after each cycle. The target gene signal was detected and analyzed with the ABI 7900HT Fast Real-Time PCR System, and relative quantification results were normalized using the ribosomal protein S7 gene as an internal control.

RNAi primers were designed with the E-RNAi webservice (http://www.dkfz.de/signaling/e-rnai3//). The T7 promoter sequence (5′- TAATACGACTCACTATAGGG) was incorporated into all forward and reverse RNAi primers. The target gene fragment was amplified with Ex Taq DNA Polymerase (Takara). Fragments were amplified and cloned into a pCR 2.1-TOPO vector at 23°C for 30 minutes using a TOPO TA Cloning Kit (Invitrogen). The constructed plasmid was transformed into HIT-DH5α competent cells. Plasmids from positive colonies were purified using a FarvoPrep Plasmid DNA Extraction Mini Kit (Favogen) and sequenced to confirm that the cDNA was in frame.

The plasmid was digested by a restriction enzyme and fragments were separated using 1% agarose gel. Target fragments were isolated and purified from the gel using a FarvoPrep GEL/PCR Purification Kit (Favogen). The fragments were then amplified with Ex Taq DNA Polymerase (Takara) and purified with the FarvoPrep™ GEL/PCR Purification Kit (Favogen). The purified PCR product was used as the template for synthesizing the dsRNA in vitro using a T7-Scribe™ Transcription Kit (Epicentre). The reaction was performed at 37°C for 4 to 12 hours. A solution of 95 μL of DEPC-H₂O and ammonium acetate (stop solution) was added to stop the reaction and the supernatant was transferred into a new Eppendorf tube with 150 μL of a phenol/chloroform (AMRESCO) solution. Samples were centrifuged at 13000 rpm for 5 minutes at 4°C and the supernatant was transferred to a new Eppendorf tube with 150 μL of chloroform. After another centrifugation at 13000 rpm for 5 minutes at 4°C, the supernatant was transferred to a new Eppendorf tube with 110 μL isopropanol. Samples were gently mixed and stored at −80°C for 30 minutes. Finally, each sample was centrifuged at 13000 rpm for 30 minutes at 4°C. The dsRNA pellets were dried in a laminar flow hood and dissolved in DEPC-H₂O.

The dsRNA was diluted to a final concentration of 5 μg/μL. Between day three to five post-eclosion (PE), female mosquitoes were injected with 280 nL of dsRNA (5 μg/μL) using a Nanoject II AutoNanoliter Injecter (Drummond Scientific Company). dsRNA against LacZ was used as control dsRNA (dsLacZ). Silencing efficiency was confirmed by collecting the total RNA of mosquitoes three days post-injection for RT-PCR analysis.

The whole bodies of three to five mosquitoes or the midguts of 10 to 30 mosquitoes were collected in 1.5 mL Eppendorf tubes containing 100 µL of protein lysis buffer and homogenized with a rooter-stator homogenizer. Each homogenized sample was centrifuged at 13000 rpm for 30 minutes at 4°C and the supernatant was transferred to a QIAshredder column (QIAGEN). The eluted samples were collected and transferred to new Eppendorf tubes at −80°C. The protein concentration was quantified using the Bradford method with a Bio-Rad Protein Assay Dye Reagent (Bio-Rad Laboratories, Inc.). Each protein sample was mixed with the same volume of sample buffer, Laemmli 2× Concentrate (SIGMA), and adjusted to the same volume with 1× sample buffer. To denature proteins for electrophoresis, protein samples were incubated at 98°C for 18 minutes. The protein samples (10 µg in midguts or 60 µg in whole body mosquitoes per lane) were subjected to SDS-PAGE and blotted onto a PVDF membrane (Pall Corporation) for 1.5 hours. The membranes were blocked with 5% skim milk in PBST (1× phosphate-buffered saline, 0.4% tween 20) at room temperature for one hour. Afterwards, the membranes were incubated in the blocking solution with the primary antibody (Anti-NS1, anti-Anopheles gambiae TEP1, or Anti-GAPDH) overnight at 4°C. The anti-Anopheles gambiae TEP1 antibody used was a gift from Dr. Stephanie Blandin at the Institute of Molecular and Cellular Biology, French National Centre for Scientific Research (CNRS) in Strasbourg, France. Membranes were washed in a PBST solution and incubated with a secondary antibody (anti-rabbit IgG) in the blocking solution at room temperature for one hour. Finally, membranes were washed in PBST and developed using WesternBright Peroxide and ECL (Advansta Inc.) as the substrate for horseradish peroxidase following the manufacturer’s instructions.

Mosquito midguts were dissected in PBS and fixed in 4% paraformaldehyde (Electron Microscopy, Hatfield, PA) for at least four hours. The fixative was then removed and the midguts were rinsed in PBS, incubated for one hour in 0.1% Triton X-100 in PBS for cell permeabilization, and blocked with a PAT blocking buffer (1% Bovine serum albumin (BSA), 0.5% Triton X-100 in PBS) for one hour. A monoclonal mouse anti-NS1 antibody (YH0023) (Yao-Hong Biotechnology Inc., Taipei, Taiwan) was used as the primary antibody to examine DENV antigens in the midguts. They were then incubated with a 1:500 dilution of goat anti-mouse antibody conjugated with Alexa-488 fluorochrome (Molecular Probes Inc., Eugene, OR). Finally, midguts were mounted with a DAPI-containing medium for confocal microscopy (ZEISS, LSM 510 META Confocal Microscope).

The whole bodies or midguts were collected from TEP1 silenced, dsLacZ-treated, or wild type (control) mosquitoes in 100 μL serum-free medium with antibiotics (penicillin–streptomycin) and stored at −80°C. C6/36 cells were seeded in a 24-well tissue culture plate and incubated at 28°C overnight. The homogenized suspensions of infectious mosquitoes were centrifuged at 18,928 × g for 30 minutes and kept on ice. The cell monolayers were rinsed with PBS and 200 μL of the 10-fold serial dilutions of infectious mosquito suspensions were added for two hours. After viral adsorption, 500 μL 1% methyl cellulose cell media with antibiotics (penicillin–streptomycin) was added and the plates were kept in an incubator at 28°C for five days. The plates were fixed with 4% formaldehyde for one hour at room temperature and stained with 1% crystal violet for 30 minutes. Plaques were quantified via manual counting (24).

Female mosquitoes were allowed to lay eggs for 50 minutes three days after the blood meal. The DNA of donor and helper plasmids was mixed at the ratio of 500:300 ng/µL and diluted in a 1× injection buffer (2 mM KCl, 0.1 mM sodium phosphate, pH 6.8). Approximately 500 injected embryos were kept on the filter paper for four days before hatching. Each surviving male and female adult from the injected generation 0 (G0) was outcrossed with three control females or males at a male/female or female/male ratio of 1:3. eGFP fluorescence driven by the 3xp3 promoter manifests at the optic nerve and tracheal gills of G1 transgenic larvae, which were screened with the help of a stereoscopic fluorescent microscope (SZX10, Olympus) (25, 26).

The functional stem-loop structure of the artificial mir-based RNAi_TEP1 miRNA was created through the first primer sets, AeTEP1-mir-1-1/Ae-TEP1-mir-1-2 or AeTEP1-mir-2-1/Ae-TEP1-mir-2-2, by PCR. This functional stem-loop miRNA was then extended and flanking sequences with restriction enzyme sites were added with the second primer set, Mir6.1_5′EcoRI/BglII and Mir6.1_3′BamHI/XhoI, to get the precursor TEP1 miRNA unit. The BglII and BamHI restriction enzyme sites of the precursor TEP1 miRNA unit were used for assembling TEP1-miR-1 and TEP1-miR-2 to generate the TEP1-2miRNA cassette (27). Based on the above, following double digestion by the restriction enzymes, EcoRI/BamHI-TEP1-miR-1 and BglII/XhoI-TEP1-miR-2 were integrated concurrently into the EcoRI and XhoI sites of pMOS1_AePUb-Den3-4miR_3xp3-eGFP (GenBank accession: MG603748) to generate a pMOS1_AePUb-miR-TEP1-2miR_3xP3-eGFP transition plasmid with a truncated AePUb promoter (26). The AeCPA promoter was amplified from Ae. aegypti genomic DNA by using the following primers: pMOS1_fusion_AeCPA-pr-F and pMOS1_fusion_AeCPA-pr-R (28). Finally, the AeCPA promoter fragments were cloned into the FseI and EcoRI double-digested pMOS1_AePUb-miR-TEP1-2miR_3xp3-eGFP transition plasmid with In-Fusion HD Cloning technology (Clontech), generating the pMOS1-AeCPA-miR-TEP1-2miR-3xp3-eGFP vector.

All statistical analyses were performed using GraphPad Prism 5 software. Gene expression and fecundity data were analyzed using ANOVA for all independent experiments.

To identify the immune-responsive genes involved in DENV replication in mosquitoes, we selected several immune-responsive genes previously identified as potential inducible genes (29, 30). Total RNA was extracted from the midguts of mosquitoes at three and seven days after an infectious or normal blood meal. The transcriptional profiles of the immune-responsive genes from normal (BF) and infectious DENV2 blood-fed (DENV2) mosquito midguts were examined with qRT-PCR analysis. Interestingly, our results showed that transcription of TEP1 was significantly up-regulated three days post DENV2 infection ( Figure 1 ). This indicates that TEP1 is sensitive to DENV infection in the mosquito midgut. Therefore, we investigated the role of TEP1 in DENV2 replication further.

First, we examined the expression profiles of TEP1 between normal and infectious DENV2 blood-fed in mosquito midguts. Midguts from female mosquitoes were collected 6, 12, 24, 48, and 72 hours after a normal or infectious blood meal. Equal amounts of total RNA from each group were used for cDNA synthesis. The transcriptional profiles of TEP1 were examined with qRT-PCR analysis. Our results showed that TEP1 RNA expression was higher in the midgut of the mosquito after a blood meal ( Figure 2A ). The RNA expression level was significantly higher after an infectious blood meal. In order to examine the translational pattern of TEP1, total protein from the midgut of the mosquito was collected 6, 12, 24, 48, and 72 hours after a normal or infectious blood meal. Equal amounts of total protein from each group were used for western blot analysis. Our results show that TEP 1 was activated at 6 hours with a blood medium and decays at 12 hours after feeding, while the expression is higher at 6 hours when the blood includes virus and this was maintained up to 24 hours post infection ( Figure 2B ). To clarify the role of TEP1 in DENV2 replication, we used a reverse genetic approach by introducing TEP1 dsRNA into the mosquito to silence TEP1 expression. The viral genome and NS1 protein expression were activated in the absence of TEP1 ( Figures 2C, D ). These results indicate that TEP1 plays a crucial role in regulating DENV2 replication in the midgut of mosquitoes.

To investigate the function of TEP1 in mosquitoes, we generated a TEP1 loss-of-function transgenic mosquito using anti-TEP1 miR expression ( Figure 3A ). The expression of anti-TEP1 miR was regulated by a mosquito midgut-specific CPA promoter, which is activated after a blood meal. First, we examined the efficiency of TEP1 silencing in the transgenic mosquitoes. The total RNA of wild type and anti-TEP1 miR-expressed transgenic mosquitoes was collected 6, 12, 24, 48, and 72 hours after a normal blood meal. The expression levels of both TEP1 ( Figure 3B ) and GFP ( Figure 3C ) were determined by qRT-PCR. Compared with wild type mosquito, our results indicate that the TEP1 mRNA level was down-regulated 24 and 48 hours after the blood meal in the transgenic mosquitoes ( Figure 3B ). GFP expression was used as a marker for anti-TEP1 miR-expressed transgenic mosquitoes ( Figure 3C ). To investigate the role of TEP1 in DENV2 replication in the mosquito midgut, replication efficacy in transgenic mosquitoes was examined. Total protein from wild type or anti-TEP1 miR-expressed transgenic mosquitoes were collected 1, 2, 4, 6, 8, and 10 days after a normal or infectious blood meal ( Figure 3D ). Expression of the viral proteins were significantly increased in the transgenic mosquitoes after an infectious blood meal compared to the wild type mosquitoes ( Figure 3D ). Combined, our results suggest that TEP1 is crucial for boosting DENV2 replication in the midgut of mosquitoes.

We examined the effect of TEP1 on infectious virus particle production by comparing the efficiency of particle production between wild type and anti-TEP1 miR-expressed transgenic mosquitoes. Whole body samples of wild type or transgenic mosquitoes were collected 2, 4, 6, 8, 10, and 14 days after an infectious blood meal. They were examined with a plaque assay for infectious virus particle quantification. Our results show that, in response to TEP1 silencing, infectious virus particle production efficiency was higher in transgenic mosquitoes than in wild type mosquitoes after the infectious blood meal ( Figure 4A ). This supports the notion that TEP1 plays a key role in DENV replication. In addition, we examined the effect of TEP1 on infectious virus particle production in the mosquito midgut with a plaque assay ( Figure 4B ). Combined, our results indicate that TEP1 serves as a negative regulator for DENV2 replication in the midgut of mosquitoes.

A previous study reported that a TEP-related protein, Ae. aegypti macroglobulin complement-related factor (AaMCR), is an essential factor in resisting flaviviral infection in Ae. aegypti (18). Moreover, AaMCR interacts with DENV through a homologue of the scavenger receptor-C (AaSR-C), which interacts with DENV and AaMCR. The expression of AMPs is regulated by the AaSR-C/AaMCR complex for potent anti-DENV activity. Additionally, previous studies have also suggested that AMPs play crucial roles in mosquito immune defense against DENV (31–33). Therefore, we hypothesized that TEP1 controls DENV replication by regulating the expression of AMPs. To examine this hypothesis, total RNA was extracted from the midgut of either the wild type or anti-TEP1 miR-expressed transgenic mosquitoes 6, 12, 24, 48, and 72 hours after a normal blood meal. The transcriptional profiles of immune-responsive genes from normal blood-fed mosquitoes were examined with qRT-PCR analysis. Our results showed that Rel1 remained unchanged in TEP1 silenced mosquitoes ( Figure 5A ) whereas Rel2, and defensins were significantly suppressed in TEP1 silenced mosquitoes ( Figures 5B, C ). On the other hand, cecropins was also suppressed at PE stage in TEP1 silenced mosquitoes ( Figure 5D ). Our results point out the crucial role TEP1 plays in controlling immune response in Ae. aegypti.