Mimicking superinfection exclusion disrupts alphavirus infection and transmission in the yellow fever mosquito Aedes aegypti

Significance Arboviruses are a significant burden to human health worldwide. Due to the lack of licensed vaccines and antiviral therapeutics, current disease prevention mainly focuses on vector control by means of insecticides. Genetic control technologies have the potential to provide more sustainable means of mosquito vector control. We have investigated a virus-strain independent, low-resistance, synthetic biology approach with broad spectrum potential to control the transmission of alphaviruses such as chikungunya virus. The uniqueness of our approach is based on utilizing a mechanism these viruses use themselves to prevent infection of the same cell with multiple different alphaviruses—a phenomenon termed superinfection exclusion. A population of genetically modified mosquitoes exhibiting this mechanism could result in reduced disease transmission in a relevant area.

Multiple viruses, including pathogenic viruses, bacteriophages, and even plant viruses, cause a phenomenon termed superinfection exclusion whereby a currently infected cell is resistant to secondary infection by the same or a closely related virus.In alphaviruses, this process is thought to be mediated, at least in part, by the viral protease (nsP2) which is responsible for processing the nonstructural polyproteins (P123 and P1234) into individual proteins (nsP1-nsP4), forming the viral replication complex.Taking a synthetic biology approach, we mimicked this naturally occurring phenomenon by generating a superinfection exclusion-like state in Aedes aegypti mosquitoes, rendering them refractory to alphavirus infection.By artificially expressing Sindbis virus (SINV) and chikungunya virus (CHIKV) nsP2 in mosquito cells and transgenic mosquitoes, we demonstrated a reduction in both SINV and CHIKV viral replication rates in cells following viral infection as well as reduced infection prevalence, viral titers, and transmission potential in mosquitoes.

superinfection exclusion | alphaviruses | nsP2 protease | arbovirus transmission
Superinfection exclusion (SIE) describes a natural phenomenon by which a current infection with one virus prevents a subsequent infection by the same or a closely related virus.It has been described for a range of viruses including human pathogenic viruses, plant viruses, and bacteriophages (1)(2)(3)(4)(5).One of the groups displaying SIE is alphaviruses (1,(6)(7)(8)(9), a group containing several pathogenic members.Alphaviruses are enveloped, positive-sense RNA viruses belonging to the Togaviridae family (10,11).While some alphaviruses are known to only infect invertebrates (12)(13)(14), most members also infect vertebrate hosts and are transmitted between them via arthropod vectors (15).Currently, the most important alphavirus with respect to causing human disease is the chikungunya virus (CHIKV)-responsible for multiple disease outbreaks in Africa, Southeast Asia, and Americas in recent decades (16).
The exact mechanisms by which certain viruses achieve SIE are either unknown or poorly understood, but there appears to be a wide range of means by which this phenomenon is mediated (4,5,17,18).For alphaviruses, there is evidence that the phenomenon is, at least in part, mediated by the viral protease nsP2 of the primary infecting virus (17,19).In a natural alphavirus infection, nsP2 processes the nonstructural polyproteins P123 or P1234-translated from the viral genome-in a timely, organized fashion to form the viral replication complex (RC) via intermediate steps crucial to different stages of the alphavirus replication cycle.More specifically, nsP2 mediates cleavage, firstly of nsP3/nsP4, then nsP1/nsP2, and lastly of nsP2/nsP3.Thus, normally it is only after the final processing event that nsP2 is present in its free form and available to mediate SIE against incoming viruses (8,(20)(21)(22)(23).
Due to its natural properties, the phenomenon of SIE may present a robust route toward generating alphavirus-refractory systems in mosquitoes (24).Such a system would be particularly relevant in Aedes aegypti-the main vector for CHIKV and other pathogenic alphaviruses (25,26).Currently, there are no approved vaccines available to prevent infection of humans with any pathogenic alphaviruses and the only preventative method is vector control.Control via genetic modification of the mosquito vector to achieve a virus-refractory effect is a unique approach to overcome this problem (27)(28)(29).Of particular interest would be a broad-spectrum effect (i.e., simultaneously affecting multiple viruses) which could potentially be achieved by mimicking a naturally broad-spectrum phenomenon, such as SIE.Recent work suggests that such a route may be feasible: Using a trans-replication system (30) including Ross River virus and Mayaro virus (delivered as replicase plasmids) was investigated and both, albeit to varying extents, interfered with RC-mediated reporter expression (17).
Here, we describe our efforts to employ a synthetic biology approach aimed at developing a virus refractory system in A. aegypti which functions by mimicking the SIE effect displayed during natural alphavirus infection.We achieve this in both, A. aegypti cells (Aag2) and transgenic A. aegypti mosquitoes, through expression of a single viral protein, the viral protease nsP2.Through transiently expressing SINV nsP2 and CHIKV nsP2 in Aag2 cells, we observed a reduction in viral replication of SINV and CHIKV using a trans-replication system.Subsequently, we generated transgenic mosquitoes expressing either SINV nsP2 or CHIKV nsP2 and demonstrated a reduced infection prevalence, as well as significantly lower transmission potential of SINV.Finally, we provide evidence that the expressed nsP2 remains in the cytoplasm in Aag2 cells, the relevant site of action to interfere with the formation of viral RC of the incoming virus.

SIE in Aag2 Cells Using Two Different Fluorescently Tagged SINV.
To assess if SIE can be detected in A. aegypti-derived Aag2 cells, we infected these cells with mCherry-tagged SINV (SINV mCherry) at three different multiplicity of infection (MOI) 0.1, 1, and 10 (or mock-infected) and, after 24 h, we infected the same cells with ZsGreen-tagged SINV (SINV ZsGreen), at three different MOI for each of the previous treatments.After a further 24 h of incubation, ZsGreen fluorescence was measured.
Images of the cells taken at the same time point show that cells appear to be predominantly infected with either SINV mCherry (red) or SINV ZsGreen (green) (Fig. 1B: SINV mCherry MOI 0.1/SINV ZsGreen MOI 10 and 1C: SINV mCherry MOI 10/ SINV ZsGreen MOI 0.1).Double-infected cells in which both viruses were replicating express both fluorophores and therefore appear yellow.Yellow-appearing cells could be occasionally detected (none shown in Fig. 1B or Fig. 1C), thus indicating coinfection of the same cell is possible but presumably rare.
In Vitro Reduction of Viral RNA Replication via nsP2 Expression in a Split Reporter Assay.To assess whether expression of nsP2 alone can cause a SIE-like state, we designed nsP2 expression plasmids to transiently express nsP2 in Aag2 cells.We tested three different versions of the protease: 1) CHIKV nsP2, 2) SINV nsP2 with a mutation to increase protease activity (nsP2 ND ), and 3) SINV nsP2 with a mutation to abolish the protease activity (nsP2 CA ).The N614D mutation was chosen as previous work identified this mutation in SINV to enhance protease activity and disrupt viral replication (31).Additionally, more recent work using a trans-replicase approach (17) found that this mutation in transiently expressed SINV nsP2 delivered a slightly stronger effect compared to wild-type SINV nsP2 in C6/36 cells.The C481A mutation of SINV nsP2 was added to the panel to test whether the protease function of nsP2 is needed to achieve a SIE effect.SINV/CHIKV nsP4 expression plasmids were added to the panel to function as a negative control accounting for any nonspecific effect that the expression of a viral nonstructural protein, or a plasmid of this structure, might have on viral replication.
In the split reporter activation assay, viral RNA replication activity was measured indirectly, by analyzing expression levels of nanoluciferase (nLuc) encoded on an alphavirus reporter template and translated from replicase-generated subgenomic RNA (Fig. 2A and ref. 30).In this system, the reporter template resembles the positive-sense viral genome RNA containing the 5′ and 3′ UTR as well as the subgenomic promoter of a specific alphavirus.Nonstructural and structural proteins are exchanged for reporters.The reporter RNA cannot replicate itself but provides a substrate which can be replicated in trans by viral RNA replicase.Upon viral infection of a cell, viral components of the replication complex will be expressed; these will act on the reporter RNA leading to transcription and translation of the subgenomic RNA encoded nLuc reporter protein (30).Cells were cotransfected with either a SINV or a CHIKV reporter template plasmid, as appropriate for the subsequent virus infection, and CHIKV nsP2, SINV nsP2 ND , SINV nsP2 CA , SINV nsP4, or CHIKV nsP4 expression plasmids.We then measured nLuc levels as a measure of replication activity of SINV or CHIKV in the cotransfected and infected cells and compared to nLuc activity in SINV-or CHIKV-infected cells only transfected with the respective reporter template.As the reporter plasmid and the nsP2 plasmid are cotransfected into the same cells, this system is robust to the low transfection efficiency commonly seen in Aag2 cells as only transfected cells will be analyzed via detection of the nLuc versus all cells being analyzed in conventional cytopathic effect based assays (30).Transient expression of functional nsP2 (CHIKV nsP2 or SINV nsP2 ND ) led to a statistically significant reduction in nLuc expression after infection with either SINV or CHIKV compared to expression of the reporter template alone.Transient expression of SINV nsP2 CA did not show a statistically significant influence on nLuc expression while expression of either SINV nsP4 or CHIKV nsP4 led to an increase in expression of nLuc after infection with the respective virus (Fig. 2 B, Left: SINV, Right: CHIKV).In detail, cotransfection with SINV nsP2 CA did not cause a change in nLuc expression compared to nLuc expression in the absence of nsP2 CA expression plasmid neither after infection with SINV nor CHIKV (P > 0.05) in the respective transreplication system.Both cotransfection with SINV nsP4 and CHIKV nsP4 in the respective trans-replication system did not lead to a reduction in nLuc expression.In fact, cotransfection with SINV nsP4 caused a modest nLuc expression increase (t(13) = 2.464, P = 0.0284); similarly, coexpression of CHIKV nsP4 also caused an increase in nLuc expression (t(9) = 7.391, P < 0.0001).After infection with SINV, there was significantly less nLuc induction in cells cotransfected with one of the two fully functional nsP2 variants; more so with SINV nsP2 ND (t(10) = 4.939, P = 0.0006) than CHIKV nsP2 (t(10) = 4.087, P = 0.0022).After infection with CHIKV, similarly to what was observed in the SINV system, there was significantly less nLuc expression in cells cotransfected with one of the two fully functional nsP2 variants.Again, nLuc was reduced to a greater extent homotypically, in this case, more by CHIKV nsP2 (t(10) = 3.434, P = 0.0064) than SINV nsP2 ND (t(11) = 3.284, P = 0.0073).
nsP2 Is Predominantly Cytoplasmic in Aag2 Cells.Alphaviruses display a cytopathic effect in mammalian cells.This is thought to be mediated, at least in part, by translocation of a fraction of nsP2, once cleaved off of the P23 polyprotein precursor, into the nucleus and there causing host cell transcriptional shut off (32,33).Translocation of free nsP2 protease is mediated by nuclear localization signals (NLS) within the protease region (34).Mutations disrupting the NLS have been shown to result in reduced cytotoxicity (17,35,36).Mosquito cells do not typically display a cytopathic effect upon infection, similarly, infected mosquitoes do not display overt symptomatic disease upon infection.We investigated whether we could find evidence of translocation of transiently expressed nsP2 into the nucleus.Our hypothesis that nsP2 mediates a SIE-like state by prematurely processing virusencoded P123/P1234 of an infecting virus requires sufficient nsP2 to be present in the cytoplasm, so eliminating nuclear localization might increase the effect.
To determine the subcellular localization of nsP2 protein in Aag2 cells, we designed a further nsP2 variant, SINV nsP2 FLAG, containing a DYKDDDDK-tag sequence added to the C-terminus of the protein via a flexible linker (SGGSGG).We monitored intracellular localization of transiently expressed nsP2 in Aag2 cells.At three time points (24 h, 48 h, and 72 h post transfection), nsP2 was observed localized in the cytoplasm (shown in green, Fig. 2 C-E).The transformation marker DsRed containing a NLS can be seen (Fig. 2 C-E, in red) only from 48 h post transfection due to slow DsRed tetramer formation and maturation (37) and, as expected, shows a clear colocalization with the nuclear dye DAPI (shown in blue).

Virus Infections of nsP2 Expressing Transgenic Mosquitoes.
Transgenic female A. aegypti expressing either SINV nsP2 ND or CHIKV nsP2 driven by an A. aegypti polyubiquitin promoter (PUb, ref. 38), were offered an infectious blood meal containing either SINV (1 × 10 9 PFU/mL) or CHIKV (5 × 10 8 PFU/mL), 5 to 7 d after emergence as adults.At 7 days post infection (dpi), saliva and body samples (not including wings and legs) were collected from each of the mosquitoes.Saliva samples were analyzed for the presence of virus (transmission potential prevalence), body samples were analyzed for presence of virus (infection prevalence), and viral titers of positive body samples were calculated (TCID 50 ).Only saliva samples from corresponding virus-positive body samples were analyzed for virus prevalence.

Discussion
Although SIE is still a poorly understood phenomenon, it has been suggested that in alphaviruses, the proteolytic activity of nsP2 of a primary infecting virus plays a part in prohibiting infection of a second virus (6,39).We have shown that expression of this multifunctional protein reduces replication of two different alphaviruses, SINV and CHIKV, in Aag2 cells.We generated transgenic A. aegypti lines which express either SINV nsP2 or CHIKV nsP2 and demonstrated that by mimicking a persistent SIE-like state in this mosquito vector, subsequently infected mosquitoes exhibit a reduced infection prevalence, viral titers in infected mosquitoes, and transmission potential prevalence, particularly after infection with SINV.
Initially, using an A. aegypti-derived Aag2 cell culture system and two different SINV reporter viruses, we demonstrated that infection with one alphavirus inhibits replication of a second SINV reporter virus in the same cell.Alphavirus SIE had previously been demonstrated in mammalian (33) and other mosquito cell lines such as A. albopictus C6/36, U4.4, and C7-10 (6).Using fluorescent reporter viruses, we demonstrated that infection with SINV substantially inhibits replication of a second alphavirus (in this case, SINV ZsGreen), consistent with previous work in other mosquito cell systems (6, 8).As here, previous studies of SIE in cell culture did not show a complete prevention of viral replication of the secondary infecting virus (6,8).However, as analysis in previous studies was carried out by observation of cytopathic effect, it is unclear if replication of the two different viruses occurred in the same, or in different cells.To gain more information, we utilized fluorescent marked viruses and live-cell imaging and found the majority of cells expressed either the reporter of one virus (mCherry-red) or the other virus (ZsGreen-green), rather than both fluorophores (Fig. 1 B and C).This was indicative of replication of predominantly either, but not both, viruses within the same cell.
After demonstrating alphavirus SIE in Aag2 cells, we investigated whether expression of one viral protein alone, nsP2, could recreate this effect.Previous work in another mosquito cell line had established that expressed nsP2 plays a role in disrupting replicase-delivered RC formation (17).Using a modified trans-replication system (Fig. 2A), we demonstrated that SINV-and CHIKV-derived nsP2 greatly reduced nLuc expression (henceforth a proxy for viral RNA replication) after infection with either SINV or CHIKV, in both cases to a greater extent when in a homotypic infection scenario (e.g., SINV nsP2/SINV > SINV nsP2/CHIKV).Concurring with this proposed proteolytic SIE mode-of-action, cotransfection with a protease-dead SINV nsP2 CA mutant did not significantly affect viral replication.Additionally, cotransfection with either SINV or CHIKV-derived nsP4, included as a control to show that inhibition is a specific property of nsP2, actually increased viral RNA replication.This may be attributable to an increased availability of the usually limited nsP4 (due to a naturally occurring opal stop codon at the 3′ end of nsP3 limiting synthesis (reviewed in ref. 40)] combined with rapid degradation of nsP4 protein following the N-end rule (41) for replication of the used template RNA.Our results are in line with previous work (6,39) and concur with results obtained by Cherkashchenko et al. (17).However, while there the authors cotransfected plasmids expressing nsP2 and a ns-polyprotein as a source of viral RC, we instead first transfected with nsP2-expressing plasmid and subsequently (24 h later) infected those same cells with active virus.This was to provide a more realistic timeline of nsP2 expression as envisaged for virus infection experiments in transgenic mosquitoes.Together with lower plasmid transfection concentrations used in our study (32 to 60 times lower), the above difference in experimental design might account for variance in effect sizes observed in some of the tested experimental conditions compared to Cherkashchenko et al.
Interestingly, our observations and those of Cherkashchenko et al. (17) differ markedly from another recent study, by Boussier et al., in which no nsP2-mediated SIE effect against CHIKV was observed ( 11).This difference may be due to nsP2 protease activity depending on the correct N-terminal specification.Here, we ensured faithful recreation of this correct specification by including the DNA sequence encoding the final 10 amino acids of nsP1 upstream of the nsP2 sequence, from which nsP2 can then process itself, mimicking the natural production of nsP2 from a larger polyprotein.Contrastingly, Boussier et al. expressed nsP2 without this nsP1-derived sequence using an artificial start codon.Previous work has shown that an incorrect N-terminal residue or presence of even one additional residue can lead to severely reduced proteolytic activity of nsP2 (42,43).In addition, Boussier et al. used a mammalian cell culture system and had difficulties generating an nsP2-expressing stable cell line.We did not experience such difficulties in our insect cell culture system when transiently expressing nsP2 and observed a robust SIE-like effect of nsP2 against CHIKV in Aag2 cells.These earlier findings in combination with ours could indicate that SIE against CHIKV might be mediated via different mechanisms in mammalian versus insect systems or differences in the timing of polyprotein processing An ongoing question in alphavirus replication concerns the subcellular localization of free nsP2 and what impact this has on viral replication dynamics.In a mosquito cell line, it was observed that mutations altering the putative NLS of nsP2 did not impact the ability of this protein to reduce RC-dependent replication (17).Our nsP2 confocal imaging data provides an explanation for this observation: We could not detect translocalization of nsP2 into the nucleus in Aag2 mosquito cells, indicating that, unlike in mammalian cells, these NLS are nonfunctional in the Aag2 mosquito cell line.This could also potentially explain the absence of a cytopathic effect of nsP2 expression in mosquito compared to mammalian cells-an effect which has complicated functional studies of this important viral protein in mammalian systems (11,32,33).
In summary, as a proof-of-principle, we have successfully generated transgenic mosquitoes expressing either SINV nsP2 or CHIKV nsP2.No major adverse fitness effects, e.g., lethality or sterility, were observed in these transgenic mosquitoes, though more detailed studies would be needed to quantitatively assess more subtle effects.Upon viral challenge via infectious blood meal, we demonstrated a significant reduction in infection prevalence, viral titers of those mosquitoes that were infected and prevalence of transmission potential after infection with SINV compared to LVP females.The same trend, albeit not statistically significant for all parameters, was observed upon challenge with CHIKV.Interestingly, results from our cell culture experiments failed to accurately predict a SIE effect of nsP2 in transgenic mosquitoes.While, similarly to cell culture data, we saw a better homotypic than heterotypic SIE effect against both SINV and CHIKV, the SIE effect with respect to infection prevalence against CHIKV in transgenic mosquitoes was weaker than cell culture data would have suggested.In mosquitoes infected with CHIKV, viral titers were greatly reduced, however, not below viral detection threshold.This could be due to a number of different factors, including differences in the evolution of vector/virus interactions-A.aegypti is one of the main vectors for CHIKV but not SINV, which mainly is transmitted by ornithophilic mosquito species.The observed effect could conceivably be the result of interactions between transgene insertion site-specific tissue expression patterns (we attempted but did not manage to locate insertion sites for either of the two nsP2-expressing transgenic lines) and infection and replication patterns of the respective viruses [e.g., salivary gland lobe specific expression differences between SINV and CHIKV (44)(45)(46)].
While other mechanisms have been utilized to engineer arbovirus refractory mosquitoes such as RNA interference (28), expression of microRNAs (47), or expression of a single-chain variable fragment derived from a human monoclonal antibody (27), this study reports of engineered mosquitoes expressing a single viral protein to achieve a viral refractory effect.Alphavirus nsP2 is a highly conserved protein, and its function is crucial to alphavirus replication, making it an ideal target in such refractory systems for avoiding resistance evolution.Furthermore, this conservation may allow broad-spectrum application of identified components.Further work is required to identify key elements of the multifunctional alphavirus nsP2 to improve the identified refractory effect without risking inducing fitness costs to the mosquito.This study, together with recent work (17), provides evidence that an engineered SIE system holds the potential for a broad spectrum, virusstrain independent, low-resistance, synthetic biology approach to develop alphavirus refractory mosquitoes.

Materials and Methods
Construction of nsP2 Plasmids.Construction of nsP2 expression plasmids was performed as described in Cherkashchenko et al. (17).In short, sequences encoding full-length nsP2 of SINV (isolate Toto1101, ref. 48) or CHIKV (isolate LR2006OPY1, East/Central/South African genotype) were codon optimized for usage of A. aegypti and cryptic splice sites present in these sequences were removed.SINV nsP2 ND contains a single amino acid substitution at position 614 changing aspartic acid (D) to asparagine (N), resulting in an increased protease activity (31).SINV nsP2 CA contains a single amino acid substitution at position 481 changing cysteine (C) to alanine (A), resulting in inactivation of the protease activity of nsP2.Expression of the transgene is driven by A. aegypti polyubiquitin promoter (PUb) (38) and the plasmid contains a transcription terminator of hsp70 gene from Drosophila melanogaster, as well as a HR5/iE1 promoter driving expression of the red fluorescent protein DsRed functioning as transformation/transgenic marker.Sequences of all plasmids were confirmed using Sanger sequencing.The correct nsP2 N-terminal residue (glycine for CHIKV nsP2 and alanine for SINV nsP2) is important for its various functions, including its protease activity (42,43,49).We therefore encoded not only full nsP2 but also a 10 C-terminal amino acid residue of the respective nsP1 which after translation is removed by nsP2 protease cleavage leaving fully functional nsP2.
Virus Stocks.SINV AR339 stocks were propagated in BHK-21 cells by culturing for 48 h at 37 °C.SINV (Toto1101) harboring either a mCherry or ZsGreen marker and a duplicated subgenomic promoter in the intergenic region and CHIKV (isolate LR2006OPY1) were rescued from icDNA clones.icDNA plasmids containing SP6 promoters, were linearized, in vitro transcribed using the mMESSAGE mMACHINE™ SP6 Transcription Kit (Thermo Fisher) and the resulting capped RNA electroporated into BHK-21 cells using a Gene Pulser Xcell electroporator (BioRad) and cultured for 48 h at 37 °C.The culture media was harvested and clarified by centrifugation.Virus stocks used for mosquito infection studies were concentrated using Amicon™ 100 kDa Ultra-15 Centrifugal Filter Units (Merck) to achieve appropriate virus concentrations in the resulting blood meal.Virus stocks were aliquoted and stored at −80 °C.

SIE Exclusion in Aag2
Cells.To test whether SIE can be detected in A. aegypti cells, Aag2 cells were infected with SINV mCherry at three different MOI (0.1, 1, and 10), and after 24 h of incubation, a secondary infection with SINV ZsGreen was carried out.For each of the infections, the inoculum was left on the cell monolayer for 1 h and subsequentially exchanged for fresh L-15 media.Plates were imaged using the IncuCyte live cell imager (Sartorius) after 24 h of incubation following the secondary infection and images analyzed using the associated IncuCyte software (Sartorius).The experiment was carried out in eight technical replicates per treatment.
Transreplication nsP2 Assay.Aag2 cells were seeded into 96-well plates (55,000 cells/well).After 24 h, cells were cotransfected with the viral reporter template plasmid appropriate for the virus to be tested (10 ng/well), one of the plasmids in the panel to be tested (encoding for SINV nsP2 CA , SINV nsP2 ND , CHIKV nsP2, SINV nsP4, each 20 ng per well) and a firefly luciferase-expressing plasmid (HR5/IE1-firefly-SV40, 0.2 ng) as internal control.The used transreplication assay and reporter template plasmids have been described elsewhere (17,30,50).Transcription of both SINV and CHIKV reporter template plasmids is driven by a truncated A. aegypti PUb promoter and the plasmids contain either a SV40 (CHIKV) or P10 (SINV) transcription terminator (Fig. 2A).Transfection was performed using TransIT-PRO® Transfection Reagent & Kit (Mirus Bio).Post transfection, cells were incubated at 28 °C and infected with the respective virus inoculum at MOI1 24 h later.After a further 48 h of incubation, cells were harvested and luciferase assays were performed using the Nano-Glo® Dual Reporter Assay Kit (Promega) and the GloMax®-Multi Detection luminometer System (Promega).The experiment was carried out in eight technical replicates per treatment.
Immunocytochemistry. Aag2 cells were seeded in 24-well plates (120,000 cells/well), with each well containing a round coverslip.Cells were transfected and incubated as described above with a PUb-driven, FLAG-tag SINV nsP2-expressing plasmid.After 24 h, 48 h, and 72 h, a subset of cells was fixed using 10% formalin solution (Sigma) for 1 h.After fixation, cells were washed three times with PBS containing 0.1% Tween (PBST, Sigma) and unspecific binding sites were blocked with 3% bovine serum albumin solution in PBST (Sigma).Cells were incubated in mouse anti-DDDDK (FLAG)-tag antibody solution (1:500, Abcam) overnight at 4 °C.After three washes in PBST, cells were incubated in anti-mouse Alexa 488 antibody solution (1:200, Abcam) for 1 h at RT, washed again twice, and mounted using VectaShield Hard set mounting medium containing DAPI (2bScientific).Imaging was carried out using a STELLARIS 5 confocal microscope (Leica) at The Pirbright Institute Bioimaging facilities.
Generation of Transgenic Mosquitoes.Transgenic mosquito lines were generated as previously described (51)(52)(53).Briefly, A. aegypti Liverpool strain (LVP) preblastoderm embryos were microinjected with a mixture of plasmid expressing hyperactive piggyBac transposase under control of PUb promoter (300 ng/μL) and donor piggyBac transgene plasmid (SINV nsP2 ND and CHIKV nsP2, 500 ng/μL).Injection survivors were backcrossed to LVP individuals and progeny screened for the fluorescent transformation marker DsRed using a fluorescent stereomicroscope (Leica) to identify transformed individuals.Single male transformants (G 1 ) were then crossed to five LVP females each to establish independent transgenic lines.For each of the lines used for data presented in Fig. 3, approximately half of the offspring inherited the fluorescent marker, indicating single insertions.Insertion-site identification was performed as previously described (54); however, repeated efforts to determine this sequence were unsuccessful.To generate lines with a higher allele frequency for SINV nsP2 and CHIKV nsP2, transgenic individuals were crossed to transgenic individuals only, for a minimum of five generations.Transgene frequency in the lines used for virus challenge studies was between 85% and 90%.
Transgenic and LVP A. aegypti were reared and maintained at 26.5 °C (±1 °C) and 65% (±10%) relative humidity with a 12:12 hour light:dark cycle.Larvae were fed on finely ground TetraMin Ornamental Fish Flakes (Tetra GmbH), and adults on a 10% sucrose solution.Adult females were fed on defibrinated horse blood (TCS Biosciences).
SINV and CHIKV Infection of Mosquitoes and Virus Determination.All experiments with CHIKV were performed under Containment Level 3 (CL-3) conditions at The Pirbright Institute, UK.CHIKV LR2006OPY1 and SINV AR339 passaged in BHK-21 cells were used.LVP or transgenic (confirmed by red fluorescence body marker) female mosquitoes were allowed to feed on an infectious blood meal containing either SINV (1 × 10 9 PFU/mL) or CHIKV (5 × 10 8 PFU/ mL) via a temperature-controlled blood-feeding system (Hemotek Ltd.) inside a Microbiological Safety Cabinet class III.Mosquitoes were allowed to feed for 1 h and immobilized on ice, and engorged females were transferred into maintenance containers and maintained for 7 d at 27 °C and 12:12 light/dark cycle.For transmission potential prevalence, mosquitoes were immobilized on ice at 7 dpi and wings and legs were removed.Saliva was collected according to published methods (55), and the remaining body was collected.Infection prevalence and infection titers were analyzed using TCID 50 assays, and transmission potential prevalence was analyzed using cytopathic effect assays, both carried out on BHK-21 monolayers.Infection prevalence was defined as percent of positive samples over all investigated samples, infection titers were defined as TCID 50 titers of all positive samples, and transmission potential prevalence was defined as the percent of positive saliva samples out of the number of positive body samples.Experiments were carried out with 30 to 35 female mosquitoes per treatment.
Statistical Analysis.Analysis of SIE of SINV ZsGreen by SINV mCherry was carried out using Student's two-tailed t tests of combinations of interest.Analysis of SIE caused by transient expression nsP2 on SINV and CHIKV infection in Aag2 cells was carried out using Student's two-tailed test to compare treatment groups of interest.Analysis in mosquito infection studies was carried out using Fisher's exact tests for infection prevalence and transmission potential prevalence and one-way ANOVA for infection titers.All analysis was carried out using GraphPad Prism 9.3.1.
Data, Materials, and Software Availability.All plasmids have been previously published as referenced in text.Plasmids and transgenic mosquitoes are available upon request.
, Cherkashchenko et al. demonstrated that expression of Sindbis virus (SINV) nsP2 and CHIKV nsP2 was capable of reducing viral RC-dependent reporter expression in an Aedes albopictus cell line; the ability of SINV and CHIKV nsP2 to disturbed RC processing of several different alphaviruses OPEN ACCESS 2 of 8 https://doi.org/10.1073/pnas.2303080120pnas.org

Fig. 3 .
Fig. 3. Expression of nsP2 reduces SINV infection prevalence, viral titer, and transmission potential prevalence.(A and B) Percent of SINV (A) or CHIKV (B) infected mosquitoes from three different treatment groups are shown, LVP control, CHIKV nsP2, and SINV nsP2-expressing transgenic lines over total number of mosquitoes being provided an infectious blood meal (C and D).SINV (C) and CHIKV (D) TCID 50 viral titers are shown for the three different treatment groups for virus-positive samples.(E and F) Corresponding saliva samples were analyzed of positive whole-body samples and are shown as percent over total number of mosquitoes being provided an infectious blood meal.Significant differences are indicated with asterisks (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).