The CHIKV and MAYV strains used in this study were obtained from the World Reference Center for Emerging Viruses and Arboviruses (WRCEVA) at the University of Texas Medical Branch (Galveston, TX), with the exception CHIKV SL-15649 (ECSA lineage), a kind gift from Dr. Mark Heise (Morrison et al., 2011). The viruses obtained from WRCEVA include MAYV 12A (genotype L), MAYV TRVL-4675 (genotype D), and CHIKV H-20235 (Asian lineage). African green monkey kidney (Vero) cells were used to propagate viruses. Vero cells were purchased from American Type Culture Collection (Manassas, VA; CCL-81) and maintained with Dulbecco’s Modified Eagle Medium (DMEM; Sigma Aldrich; St. Louis, MO) supplemented with L-glutamine, 5% fetal bovine serum (FBS), and 1% gentamycin (herein referred to as Vero cells maintenance media) and kept at 37°C with 5% CO₂. Vero cells were grown to ~85% confluency before being infected with the respective viruses at an MOI of 0.01 in viral diluent (RPMI-1640 media with 25 mM HEPES, 1% BSA, 50 μg/mL Gentamicin, 2.5 μg/mL Amphotericin B). The cells were infected for 1 h at 37°C, with rocking every 10–15 min. After 1 h of incubation, Vero cell maintenance media was added to the flask. Once 50–75% of cells demonstrated cytopathic effects (CPE), the cellular supernatant was collected and clarified by centrifugation at 4°C before storage at −80°C.

Virus stocks were titrated by plaque assay before use in subsequent experiments. Serial 10-fold dilutions of each sample were made in viral diluent and then added to confluent monolayers of Vero cells. After a 1 h incubation period, an overlay containing 1.5% methylcellulose was added. Plaques were visualized following formalin fixation and staining with crystal violet 3 days post-infection.

Purified sdAb samples were serially diluted in viral diluent to the appropriate concentration (10 μg/mL-0.01 μg/mL). Diluted sdAbs, as well as viral diluent (no sdAb) controls, were then mixed with an equal volume of the respective virus at 1,000 PFU/ml. Following a 1.5 h incubation at 37°C, sdAb/virus mixtures were then added to confluent monolayers of Vero cells. After a 1 h incubation period, an overlay containing 1.5% methylcellulose was added. Plaques were visualized following formalin fixation and staining with crystal violet 3 days post-infection. The number of plaques in the wells without sdAbs were counted and taken as control. The average number of plaques from replicates were then expressed as a percentage of that in the control wells.

To generate the piggyBac transposon-based plasmid that co-expresses both sdAbs (CA6 and CC3), herein referred to as “AE_CA6CC3,” a single coding sequence was designed in silico that contained the Dfurin1 gene of Drosophila melanogaster Meigen (Diptera: Drosophilidae) cleavage site (R-Q-K-R) (Cano-Monreal et al., 2010) and Porcine teschovirus-1 2A (P2A) self-cleaving peptide (Liu et al., 2017) inserted between the CA6 and CC3 sequences (Figure 2). Sequences for these sdAb can be found in reference [26] (Liu et al., 2019). The entire transcript was codon-optimized for Drosophila and commercially synthesized by GENEWIZ, Inc. (South Plainfield, NJ). We then cloned the gene synthesis product containing the Ae. aegypti carboxypeptidase A (CpA) promoter, Dfurin 1, P2A, and the sdAb sequences into piggyBac [polyUb GFP] via GeneArt Gibson Assembly cloning (ThermoFisher Scientific, Waltham, MA) per manufacturer’s protocol. We sequence-validated the final AE_CA6CC3 construct using Sanger sequencing. All primers for cloning, and subsequent Sanger sequencing, were designed using SnapGene (San Diego, CA) software and obtained from Integrated DNA Technologies. Inc. (IDT; Coralville, IA). AE_CA6CC3 sequence is available through GenBank (accession: OQ921683).

Ae. aegypti Liverpool (LVP) eggs were obtained from BEI resources (Manassas, VA). Mosquitoes were reared and housed at 26°C, relative humidity 60–80%, and 12/12 h light/dark photoperiod. Larvae were fed Nishikoi fish food (Essex, England) and adult mosquitoes were provided with a 10% sucrose solution administered through cotton balls. Mosquitoes were provided defibrinated sheep’s blood (Colorado Serum Company; Denver, CO) using artificial membrane feeders for all bloodmeals.

Authors acquired necessary permissions prior to the start of the mosquito experiments. All studies were approved under IBC protocol (18–084) and conducted within an Arthropod Containment Level-3 (ACL-3) facility.

All transformations were carried out by adapting protocols previously described (Chen et al., 2021). PiggyBac donor plasmid (300 ng/μl; piggyBac [AE_CA6CC3]) that contains an enhanced green fluorescent protein (eGFP) transformation marker driven by the Ae. aegypti polyubiquitin promoter was co-injected with an in vitro transcribed piggyBac mRNA (300 ng/μl) into less than 1 h old embryos of Ae. aegypti LVP. The piggyBac-hsp70-transposase plasmid (Handler et al., 1998) was used as a template for in vitro transcription using the mMessage mMachine T7 Ultra kit (Thermofisher), followed by MEGAclear (Thermofisher) column purification. Surviving G₀ females were mated to LVP males in pools of 20–25 mosquitoes. Each G₀ male was mated individually with five LVP females in individual cages to isolate each independent line. G₁ larvae were screened for green fluorescence using a Leica M165 FC fluorescence microscope. Positive G₁ individuals were out-crossed to Liverpool mosquitoes to ensure that all transgene cassettes were stably inherited to the G₂ generation.

Midguts from WT, AE1, and AE5 mosquitoes (n = 30/group; >G₃ generation) were dissected 16 h after a bloodmeal and protein was extracted with ice-cold radioimmunoprecipitation assay buffer (RIPA buffer; 50 mM Tris-HCl pH 7.4, 150 mM NaCl, 0.25% Na-deoxy- cholate, 1% NP-40, 1 mM EDTA) containing one cOmplete^™, Mini, EDTA-free Protease Inhibitor Cocktail tablet (MilliporeSigma, Burlington, MA). The entire sample was then homogenized and centrifuged at 4°C. Clarified supernatants were transferred to a fresh microcentrifuge tube, pulse-sonicated twice for 10 s each, and mixed with SDS loading buffer. All samples were then placed in a heating block at ~90–95°C for 10 min. Samples were run on a 4–20% SDS-PAGE and transferred to a nitrocellulose membrane. The membranes were then stained with Ponceau S stain to visualize protein transfer. Following destaining, membranes were probed with AffiniPure Goat Anti-Alpaca IgG, VHH domain primary antibody (128–005-232; Jackson ImmunoResearch, West Grove, PA) at a concentration of 15 μg/mL overnight at 4°C. The secondary antibody used was rabbit anti-goat HRP (HAF017; R&D Systems, Minneapolis, MN) at a dilution of 1:1000 per the manufacturer’s recommendation. Images were generated by applying Prometheus^™ ProSignal^™ Pico chemiluminescent ECL reagents (20-300B, Genesee Scientific, San Diego, CA) to the blots and visualized using an Azure c400 gel imaging system (Azure Biosystems, Inc., Dublin, CA).

Vector competence experiments were carried out by adapting protocols previously described (Bates et al., 2021). Female mosquitoes 5–7 days old (n = 50) were separated into cartons and sucrose starved for 16 h and H₂O starved for 12 h prior to infection to promote blood-feeding. Mosquitoes were fed virus-spiked bloodmeals containing defibrinated sheep’s blood (Colorado Serum Company; Denver, CO) using artificial membrane feeders, and only fully engorged mosquitoes were separated into new containers. Pre- and post-infection blood samples were collected for use as back-titer calculations. Seven days post-infection, mosquitoes were cold-anesthetized and midguts and legs/wings were collected in viral diluent. Saliva was collected through forced salivation for 30 min in immersion oil and then mixed with viral diluent. Samples were stored at −80°C until analyzed by plaque assay. For virus titrations, all virus-negative mosquito samples were given a value of half the limit of detection (LOD/2) for statistical analyses.

For vector competence experiments, data were analyzed via a two-tailed Fisher’s exact test. For virus titration experiments, data were analyzed by Kruskal Wallis test with multiple comparisons with undetectable mosquito samples given a value of half the limit of detection (LOD/2) for statistical analyses.

We previously showed that two anti-CHIKV sdAbs, clones CC3 and CA6, were potent at neutralizing CHIKV (Liu et al., 2019), and that CC3 also has strong neutralizing capacity against MAYV and other related alphaviruses (Liu et al., 2021). We then performed individual sdAb plaque reduction neutralization tests (Figure 3A; PRNTs) and combination PRNTs (Figure 3B). The individual PRNTs were performed to further validate CC3’s neutralization activity against MAYV and to demonstrate that CA6 has anti-MAYV activity (Figure 3C). We hypothesized that using both neutralizing sdAbs simultaneously would not interfere with neutralization capacity, thus we performed a series of combination PRNTs to evaluate the level of CHIKV and MAYV neutralization in the presence of both CC3 and CA6 (Figure 3D). These data demonstrate that the use of both CC3 and CA6 at concentrations greater than 0.16 μg/mL, maintain potent neutralization of both CHIKV and MAYV.

We next constructed a piggyBac transposon-based plasmid, herein referred to as “AE_CA6CC3,” to co-express both sdAbs in Ae. aegypti mosquitoes. AE_CA6CC3 was designed to express both CC3 and CA6 under the control of the Ae. aegypti carboxypeptidase A promoter (AeCpA) (Moreira et al., 2000) to provide bloodmeal-upregulated, gut-specific gene expression (Figure 2). The eGFP transformation marker, constitutively expressed via the Ae. aegypti polyubiquitin promoter (Ae polyUb) (Anderson et al., 2010), was also included in the design to aid in screening and selection of construct-positive mosquitoes (Figure 4). To generate transgenic mosquitoes, the AE_CA6CC3 donor vector and piggyBac mRNA were co-injected into Ae. aegypti embryos and the resulting G₀ individuals were outcrossed to wild-type (WT) mosquitoes. In the G₁, GFP-positive individuals identified from each cross in were crossed with WT mosquitoes of the opposite sex to establish multiple independent lines (n = 5). Upon further observation, one of the transgenic lines was removed from experimentation due to an apparent male-linked insertion of our construct. The resulting four transgenic lines (AE1, AE2, AE4, and AE5) were reared for multiple generations to expand the colonies for preliminary vector competence studies.

We first performed preliminary vector competence experiments to determine which virus strains should be used to evaluate the transgenic lines. We compared two strains of CHIKV (CHIKV H-20235 and CHIKV SL-15649) and two strains of MAYV (MAYV 12A and MAYV TRVL-4675) in WT Ae. aegypti LVP mosquitoes. We spiked bloodmeals with each virus and evaluated virus infection, dissemination, and transmission 7 days post-infection (dpi). Our results indicated the use of CHIKV H-20235 and MAYV 12A for subsequent studies due to the high transmission rates (83 and 75%, respectively; Table 1) in WT mosquitoes. Additionally, both CHIKV H-20235 (Accession: KX262991) and MAYV 12A (Accession: KP842796) were isolated within the last 15 years.

Following the selection of virus strains, the transgenic lines (n = 4) were subjected to a preliminary vector competence analysis to assess susceptibility to CHIKV and MAYV. Multiple lines were generated and analyzed because transgene integration is unpredictable when using transposon-based delivery systems and testing of multiple lines facilitates identification of lines with the optimal transgene expression phenotype(s) (Reid et al., 2021). Female mosquitoes from the WT colony and each transgenic line were exposed to either CHIKV or MAYV viremic bloodmeals and infection, dissemination, and transmission rates were calculated 7 days later. AE1 and AE5 were selected for further experimentation since transmission rates were the lowest for each virus (Table 2). Specifically, compared to WT mosquitoes, AE1 transmission was reduced for CHIKV and MAYV (p = 0.002 and p = 0.004, respectively) and AE5 had a significant reduction (p = 0.008) in transmission for CHIKV and a near significant reduction (p = 0.056) for MAYV. Interestingly, we observed no differences in dissemination of CHIKV in our transgenic lines, which could be to lower overall systemic titers or due to leaky expression outside of the gut. Further studies can be conducted to assess expression in different compartments.

To evaluate CA6 and CC3 transgene expression in vivo, transgenic Ae. aegypti mosquito lines AE1 and AE5 were subjected to Western blot analyses. Midguts from each transgenic line (G₃ and greater) and WT mosquitoes were dissected 16 h post-bloodmeal. Purified CC3 and CA6 sdAbs (a kind gift from Dr. Ellen Goldman) were used as a positive control (Figure 5). As expected, sdAbs were expressed in the midgut tissues of mosquitoes from AE1 and AE5 but not WT mosquitoes. Notably, banding patterns which may suggest unsuccessful ribosomal-skipping of the P2A linker sequence (Liu et al., 2017) thus resulting in a fusion protein of the CA6 and CC3 sdAbs (Figure 5).

With selection of the AE1 and AE5 transgenic lines, and evidence of successful sdAb expression in these lines, we performed a larger vector competence experiment to compare infection (virus present in midgut tissues), dissemination (virus present in legs/wings tissues), and transmission (virus present in saliva samples) rates to those of WT mosquitoes (Table 3). We observed a near-significant reduction in dissemination rate (p = 0.0547) and a significant reduction in transmission rate (p < 0.0001) for line AE1 with CHIKV. We also observed significant reductions in the dissemination and transmission rates for line AE5 (p = 0.0053 and p < 0.0001, respectively) with CHIKV. Interestingly, significant reductions in all rates were found with both AE1 and AE5 lines for the MAYV vector competence experiments (Table 3).

As mentioned, we did not see as strong of a reduction in dissemination of CHIKV in our transgenic lines and believed this could be due to lower overall systemic titers within the transgenic mosquitoes. Because of this, we sought to assess the replication of CHIKV and MAYV in these transgene-positive lines by measuring the viral titers present in each of the mosquito groups. We found significant reductions of CHIKV titers in AE1 and AE5 in infection (AE1 p < 0.0001; AE5 p < 0.0001), dissemination (AE1 p = 0.016; AE5 p = 0.008), and transmission (AE1 p < 0.0001; AE5 p < 0.0001) titers when compared to WT (Figure 6). Similarly, we observed significant reductions in MAYV replication in infection (AE1 p < 0.0001; AE5 p = 0.0008), dissemination (AE1 p = 0.0008; AE5 p = 0.0025), and transmission (AE1 p = 0.0003; AE5 p = 0.0024) (Figure 6). Alongside these data, we observed significant reductions in dissemination and transmission rates of infected transgenic mosquitoes when compared to WT mosquitoes (Supplementary Table S1). Altogether, these data demonstrate that both lines expressing sdAbs are more refractory to CHIKV and MAYV than WT mosquitoes.